SM-II Developer Kit

VIDEO Developer Kit
Reference Manual
i n t e r p e rs o n a l -
c o m p utin g G m b H
Screen Machine II Developer Kit
NeXT Version 2.0

November 1995
The firm interpersonal computing GmbH reserves all rights in regard to the Video Developer
Kit. The acquisition of the Video Developer Kit does not authorize the distribution of the kernel
driver. Any licencing agreements for a kernel´s Runtime version are granted solely by interper-
sonal computing GmbH. Any failures to comply with the reserved rights will result in an im-
mediate termination of all held licencing rights or agreements.
The interpersonal-computing GmbH reserves the right to change or vary in any way the kernel
and/or the Kit´s functions without notice. The firm interpersonal computing GmbH assumes no
responsibility or liability for any arising complications or incompatibilities as a result of these
changes.
Screen Machine II! is a registered trademark of FAST Multimedia AG, München.

Movie Machine ! is a registered trademark of FAST Multimedia AG, München.
NeXT", NEXTSTEP" and NEXTIME" are registered trademarks of NeXT Computer Inc., Redwood City.
Copyright# interpersonal-computing GmbH, 1993 München.
interpersonal-computing GmbH
NeXT Center München Tel. : ++49 (0)89 22 33 75
Oettingenstraße 2 Fax. : ++49 (0)89 22 33 76
80538 München e-Mail : [email protected] (international <20 kB)
-2 -
Video Developer Kit
Contents
1 Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
The VDOboards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
The Display Details . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Things you should always do . . . . . . . . . . . . . . . . . . . 13
Split Memory Mode . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Integration Into NEXTSTEP . . . . . . . . . . . . . . . . . . . . 16
Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Animations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Classes Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
VDOboards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
SMControl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
MMControl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
SMControlAudio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
SMControlChooserController . . . . . . . . . . . . . . . . . . . 62
SMWindow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
SMView . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
SMPaletteView . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
SMViewDragging . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
SMYUVImageRep . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
SMFLMImageRep . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
MMFLMImageRep . . . . . . . . . . . . . . . . . . . . . . . . . . 101
SMTVControl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
SMTV_AFC_Table . . . . . . . . . . . . . . . . . . . . . . . . . . 117
SMTV_AFC_TableList . . . . . . . . . . . . . . . . . . . . . . . 120
Contents 3
Video Developer Kit
3 SMPalette . . . . . . . . . . . . . . . . . . . . . . . . . 123
InterfaceBuilder . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Palettes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Building an Application . . . . . . . . . . . . . . . . . . . . . . . 130
4 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . 133
Variable Types and Constants . . . . . . . . . . . . . . . . . 134
The FLM file format . . . . . . . . . . . . . . . . . . . . . . . . . 141
The YUV Color Model . . . . . . . . . . . . . . . . . . . . . . . 144
Contents 4
Chapter 1
Concepts
Video Developer Kit
Introduction
The Video Developer Kit is a collection of powerful objects bringing live video to the desktop
and into your applications. Live video is what multimedia is all about. Where up to now it was
only possible to have heterogeneous networks, e-mail and distributed objects, there are now the
VDOboards, a suite of video overlay boards of the second generation, with dedicated special
purpose video processors that result in a unrivaled picture quality; developed by FAST Multi-
media AG a leader in video integration.
Together with NEXTSTEP, the VDOboards driver and the developer kit, the world of live vid-
eo is at the developers fingertips. Any application is a candidate for live video now; where you
used to integrate e-mail and drag&drop you can now place live video.
We have built a well integrated environment with a straight forward class hierarchy that makes
development as easy as possible. Much care has been taken to make the integration into NEXT-
STEP as seamless as possible.
The examples supplied only show a subset of the possibilities that the VDOboards offers. They
are the best way to learn about the VDOboards capabilities. From then on the road is open to
develop applications for picture analysis, security observations, video capture, documentation
or just for watching CNN in a window for the latest events while trading bonds.
1-6 Concepts - Introduction

Video Developer Kit
The VDOboards
One of the new features of the Relase 2.0 kit is that it now supports more than just the Screen
Machine II. Here is a list of the boards supported by the VDOboards driver (current version with
the release of this kit). All these boards support video overlay on graphic cards with a maximum
resolution of 1280x1024.
Board Special Features
Screen Machine II 1MB Video memory, only board capable of capturing both fields in full
resoution. The video is digitized in 4:2:2.
Movie Machine Inexpensive, includes Tuner & Videotext. (Not available in the US).
The video is digitized in 4:1:1.
Movie Machine PRO Supports video out, and has a builtin Tuner. The video is digitized in
4:1:1.
Movie Machine II Supports Video out uses 4:2:2 versus the 4:1:1 of the MM PRO. A
JPEG and MPEG option can be installed on the board. The boards sup-
ports mixing of input channels with special effects and alpha dissolves
and wipes. Has built in Tuner & Videotext.
FPS 60 Has a JPEG option built in (compression & decompression) as well as
video out. The MPEG option can be installed on this board. No special
effects on video out. The video is digitized in 4:2:2.
NOTE: To use the options additional drivers are necessary. The options are not supported by
this release of the kit. Contact us if compression board support is needed.
How the VDOboards work
graphics card /
I/O Interace previous SM
luma
SVHS black monitor /
red input processor output
chroma next SM
yellow
memory
In the input stage the video signal is converted into the digital YUV 4:2:2 (4:1:1 in the case of
the MovieMachine and MovieMachine PRO) format. Video is transmitted in YUV, a format in
which Y (the brightness) and UV (the color) are separated, unlike in the RGB or CMYK for-
Concepts - The VDOboards 1-7

Video Developer Kit
mats. The final conversion for grabbed images from YUV to RGB is not handled by the board,
but integrated into the SMYUVImageRep.
For a description of the YUV format see the Appendix of the YUV Image Format.
The conversion from the video signal to 4:2:2 YUV takes place at the rate of the incoming vid-
eo, for PAL/SECAM at 50 fields and for NTSC at 60 fields per second. The type of conversion
depends on the settings of the input converter. The processor then applies any scaling and fil-
tering to the data and transfers it into the onboard memory.
The video processor also transfers the data from memory to the output stage, which mixes the
video data with the graphics signal. The output stages handles features like brightness, satura-
tion and contrast. This is why these settings will have no affect on the image in memory, and
hence the grabbed images. (This is not entirely true, the SMFLMImageRep considers these pa-
rameters when converting from YUV to RGB). To retrieve data from the board the processor
has a fourth interface to the I/O bus of the PC and this is the only bottleneck. It allows a maxi-
mum of 2-3 MB/s to be transferred, which is a far cry from the 20.8 MB/s needed for full frame
rate of PAL, or the 17.6 MB/s for NTSC.
The I/O interface supports two modes for transferring images to and from a VDOboard. One is
the so called I/O port access and is only 8 bit. The other is memory mapping (only supported by
some chipsets when the system has more than 16 MB). Memory mapping is 16bit and suffers
from a lot less overhead, increasing performance by a factor of 2 to 3.
How to use it
After correct setup and a connected video source everything is ready to be put into action. For
a simple application we need only an instance of the control class.
In this application the video will be displayed somewhere on screen, outside a window :
#import <objc/Object.h>
#import <smkit/VDOboards.h>
#import <smkit/MMControl.h>
main(int argc, char**argv)
{
NXRect aRect;
char tmp[128], tmpChar;
id control;
// We need an instance of NXApp for the control class

[Application new];
// Get instance of control class

control = [VDOboards
newWithSelectionAndBoards:VMCBOARD_ALL_BOARDS
withFeatures:VMCBOARD_ANY_FEATURES];
// Set input Possible values are
// SM_INPUT_BLACK SM_INPUT_RED
// SM_INPUT_YELLOW SM_INPUT_SVHS
[control setVideoInput:SM_INPUT_BLACK];
1-8 Concepts - The VDOboards

Video Developer Kit
// Set output window of 360 by 240 at 20.0,20.0

NXSetRect(&aRect,20.0,20.0,360.0,240.0);
[control setWindowFrame:aRect];
// Turn video on
[control setVideoOn:YES];
// Wait for keypress

gets(&tmpChar);
// Turn video off

[control setVideoOn:NO];
// Free control & App

[control free];
[NXApp free];
}
With a simple loop you could animate the window:

for(i=20.0;i<100.0;i+=10.0)
{
aRect.origin.x = i;
[control setWindowFrame:aRect];
}
A more interesting feature are the readImage: and writeImage: methods. They allow you to cap-
ture the live video images or to display images in the live video window:
id image;
NXStream *saveStream;
// Grab the current live picture as image

// Returns a newly allocated instance of NXImage
image = [control readImage];
// Open stream
saveStream = NXOpenMemory(NULL,0, NX_READWRITE);
// Write to stream
[image writeTIFF:saveStream];
NXFlush(saveStream);
NXSaveToFile(saveStream, "video.tiff");
// Free image
[image free];
// Close stream
NXCloseMemory(saveStream,NX_FREEBUFFER);
Concepts - The VDOboards 1-9

Video Developer Kit
Here is some code to display an image in the video window:
NXStream *imageStream;
SMYUVImageRep * myImageRep;
NXImage *myImage;
myImageRep = [SMYUVImageRep alloc];

// Open stream (substitue MyImage.tiff with the name
// of your favourite picture.
imageStream = NXMapFile("MyImage.tiff", NX_READONLY);
if([myImageRep initTIFFDataFromStream: imageStream])
{
myImage = [NXImage alloc];
[myImage useRepresentation: myImageRep];
[control writeImage: myImage];
[myImage free];
}
// Close stream
NXCloseMemory(imageStream,NX_FREEBUFFER);
For a closer look at what can be done with the SM outside of NeXTSTEP take a look at the SM-
Parser example. It uses almost all the features of the SMControl class, and together with its
script capability, is quite a interesting tool.
1-10 Concepts - The VDOboards

Video Developer Kit
The Display Details

Video Source VDOboard Windowserver
768
100
VideoView
max 600
zoom frame
576/480
window frame
video frame
PAL, SECAM / NTSC
5/11
0
19/21 768/640 0 100
0
0 max 800
0
0 1024
Here is an overview of the three frames that can be set and how they determine the appearance
on screen.
The video frame determines what portion of the digitized video gets written into the VDOboards
memory. The maximum values depend on what system the input source is. Theoretically it is
possible to set a selection of video to be displayed, although for this purpose it is better to use
the zoom frame, unless VDOboards memory usage is a problem (see splitMemoryMode:). The
SMControl class presets the video frame according to the specified system. If it is desired to
digitize more or less of the bandwidth of the video, it is possible to change it with the setVide-
oFrame: method (This is best done in a subclass of SMControl in the setSystem: method). Al-
though the theoretical maximum for PAL is 768 by 576, the default settings are 736 by 552 as
most video sources do not have more bandwidth.
Here are the default values the SMControl class uses for the systems:
NTSC
x = 5
y = 11
width = 640
height = 480
horizontal range:0 - 650
vertical range:0 - 500
PAL/SECAM
x = 19 // Offset to skip noise
y = 24 // Offset to skip videotext
width= 736
height= 560
horizontal range:0 - 774
vertical range:0 - 590
Concepts - The Display Details 1-11

Video Developer Kit
The window frame sets the size at which the video is displayed on screen. The maximum size
and position depend on the screen size. This is one of the major differences to the NeXT Di-
mension board, where the maximum size was determined by the type of board (PAL or NTSC).
The live video is scaled to the size of the window frame. So live sizes from 16x8 to 1280x1024
are possible. If only a part of the video should be visible, there is the possibility to specify a
zoom frame.
The zoom frame changes the portion of the video displayed on screen without having to change
what is kept in memory. It is specified in percent of the window or video frame (doesn't make
a difference). The portion of the video specified by the zoom frame is then mapped to the screen
to the coordinates specified by the window frame.
To recap, the video frame is how much is digitized into VDOboards memory, the zoom frame
is how much of that is displayed, and the window frame is where and how big it is displayed.
1-12 Concepts - The Display Details

Video Developer Kit
Things you should always do

To use live video of a VDOboard in your application there are at least two objects that you have
to integrate. The first is a SMControl (or MMControl if you need video out) object for every
VDOboard that you are going to use, and at least one SMView (or a subclass of it) for each SM-
Control object to display the live video connected to the VDOboard. It is recommended to use
a SMView or a subclass due to proper handling of the video size and position on screen.
The SMControl or MMControl object is allocated and initialized using factory methods of the
VDOboards class, giving response whether or not any VDOboards are available (because other
applications might already be using them) and asking if a VDOboard should be used anyway.
Using the +newWithSelectionAndBoards:withFeatures: methods results in a panel being
popped up, allowing the selection of a VDOboard if multiple boards are present.
To tell the SMControl which views it is going to handle, register every SMView to a SMControl
by sending an addView: message to the SMControl. Any allocated and initialized SMView
needs to know which SMControl it is using by sending a setControl: message to it. After that
you can turn the video on by sending start: to a SMView.
If you turn video on by using the setVideoOn: method of the SMControl class, you should have
specified one SMView to be the active view through a setActiveView: message sent to the ap-
propriate SMControl.
Note: We do not recommend to start video display using a SMControl method if a SMView is
used for display handling. Undesired effects may occur in further use.
Generally methods of the SMView should be preferred to those in the SMControl object, in
which the view appearance is altered.
These first steps did nothing more than turn on video of a VDOboard controlled by a SMControl
object in the bounds of a SMView.
There is one last thing you have to take care of. Every SMControl object has to be carefully
freed before the application terminates, to make the VDOboard controlled by a SMControl ob-
ject available for further use, and to remove the overlay from the screen.
NOTE: All instances of SMControl classes have to be freed using the -free method. It is not
sufficient to use zone destruction.
Concepts - Things you should always do 1-13

Video Developer Kit
Split Memory Mode

Unfortunately it gets more complicated from here. Due to the fact that on high-res screens (for
theVDOboards this is anything above 800x600) the video can be displayed only at a reduced
bandwidth. The grabbed images are also of a reduced bandwidth, although the Screen Machine
II always digitizes the full bandwidth. To make it possible to display the video and grab images
at full quality there is a special feature built into the SMControl class, the split memory mode.
The trick is to split the 1 MB of memory on the board into two regions, one for the live image
and one for the still or grabbed image. While viewing the video on screen the video is digitized
into one part of the memory (from hereon referred to as the upper memory) of the board and
displayed from there. When the picture is set to still or an image is grabbed, two fields of the
live video are written into the other part (from here on referred to as lower part) of the memory,
at the full bandwidth in the size desired. The limitations of this method are the amount of mem-
ory available. A full size PAL or SECAM frame is 864kB and NTSC 600kB, so for NTSC it is
possible to store a full quality image in lower memory and store a half size image in upper mem-
ory for the live display (600kB + 300kB < 1Meg). For PAL/SECAM this is not possible, (864kB
+ 432kB > 1Meg). In this case trade-offs have to be made, either the full quality grabbed image
has to be reduced in size, or the quality of the live picture reduced, to accommodate both in the
SM memory. Even in split memory mode there are no limitations (up to the screen size) on how
big the live (displayed) picture can be set. However it will tend to get coarse if size of the video
in the upper memory region is significantly smaller than the displayed size.
The way it works is that you specify the size and the maximum width of the picture in the lower
memory (size/width = height). With the setSplitMemGrabSize: call you can set the actual size
of the picture to be grabbed. Setting sizes larger than the split memory region can result in un-
predictable images. Setting grabSize wider than the maximum width, even if the total image
would fit, also leads to unrecognizable images. The rest of the memory is used to display the
video on screen.
The call to do this is setMemoryMode:mode offset:offset width:width.

The arguments are:
mode is SM_SPLIT_MEM to split the memory, or SM_MEM_NORM to reset.
offset is the start of the live video memory region (maximum width * maximum number of
lines).
The value is in pixels, not bytes.
width is the maximum width of the grabbed image in the lower memory region.
This value has to be a multiple of 16. Usually this should be the value set with the -setVideoF-
rame: method. If you haven't set a video frame, query the active one with -videoFrame.
To recap, inorder to use the split memory mode you have to set the mode and divide the memory
using the -setMemoryMode:offset:width : call. With this call you set the maximum size that is
possible to be digitized in the lower memory. The actual size of the image is set with the set-
SplitMemGrabSize: call.
1-14 Concepts - Split Memory Mode

Video Developer Kit
Example: Split memory mode for a PAL frame
// Set split mem bounds

//
// A PAL frame has a maximum size of 768x576. But we have a video frame
of 736x560 set.
// And we want to be able to capture the complete frame, so we calculate
the following
// value: 736*560 = 412160
[control setMemoryMode: SM_SPLIT_MEM offset: 412160 width: 736]
// Set size of grabed image

//
// We now et the size of the image that will be digitize. The only
restriction is
// the maximum size we determined for the setMemoryMode:offset:width:
call.
// We want a half PAL frame = 368 x 276
// Width * height < offset
// The width has to be a multiple of 2
aSize.width = 368
aSize.height = 276
[control setSplitMemGrabSize: aSize];
You can also try this with the SMParser:
smp> mem 2 406272 736

smp> grabsize 368 276
524288
upper mem
406272
lower mem
736 x 552
grab size
368x276
0
(Sizes are in pixels, 1 pixel = 2 bytes)
Concepts - Split Memory Mode 1-15

Video Developer Kit
Integration Into NEXTSTEP

So far we have described the basic functionality of the board, now lets focus on how to integrate
the functionality in the existing NEXTSTEP environment. Ideally it should be possible to have
the live video in a window, to drag around, miniaturize and hide it. For that purpose there are
two more classes (in addition to the SMControl class) in the SMKit that handle just that and
more. The dragging and miniaturizing is handled by the SMWindow class. It behaves just like
a regular Window with a few extra features. These extra features are that it can handle more than
one SMView (even if connected to different instances of the SMControl class), it can display
the live video in the MiniWindow and can make resizing and miniaturizing smoother than the
regular window class.
A SMView is a subclass of View that handles the resizing, printing and correct display of the
video in a view. Additionally, it also enables drag&drop, dropping TIFF and FLM images on
the view then will be displayed in the SMView, or dragging the stopped live video as a FLM,-
TIFF or the FILENAME (Pastboardtypes) out of the view.
In conjunction with an SMView and a SMWindow, the SMControl class also makes clipping
possible. Clipping is the process of masking out the parts of the video that are obscured by other
objects. Noticeably, these are windows and panels that pop up, or are placed above the active
video view.
The basic steps to create a NEXTSTEP application with live video from a VDOboard are to
place a SMView in a SMWindow and connect the view to an instance of the SMControl or
MMControl class. (For details take a closer look at the SMDemo example). Even easier is the
use of the supplied SMPalette with its SMPaletteView. Just drag a SMPaletteView into a
SMWindow, connect it to an instance of the SMControl class, connect a button to the start:
method of the view and you have live video in the InterfaceBuilder test mode.
1-16 Concepts - Integration Into NEXTSTEP

Video Developer Kit
Examples
The source can be found in smkit/Examples.
1. SMDemo
SMDemo is a simple application with the following features :
• Uses SMWindow, SMView and SMControl

• Implements saving of images
• Allows various settings
The SMDemoControl.m file contains action methods for the various settings. For example, The
radio buttons are connected to clickedInput: method.
- clickedInput:sender
{
switch ([[sender selectedCell] tag])
{
case 0: [smCtrl setVideoInput:SM_INPUT_BLACK];
break;
case 1: [smCtrl setVideoInput:SM_INPUT_RED];
break;
case 2: [smCtrl setVideoInput:SM_INPUT_YELLOW];
break;
case 3: [smCtrl setVideoInput:SM_INPUT_SVHS];
break;
}
Concepts - Examples 1-17

Video Developer Kit
return self;
}
The demo uses a SMView to display live video. For this we have to send the setVideoInput:
messages directly to the SMControl object used for this SMView.
The most important thing to mention is to free any SMControl allocated in an application.
The SMDemo takes care of that in the appWillTerminate: method.
- appWillTerminate:sender
{
[smCtrl free];// it's very important to free the
used SMControl class
// to free the sm for further use
return self;
}
2. SMPaletteDemo
The SMPaletteDemo is the last example, and by far the simplest. All you need is about 10 min-
utes and the SMPalette.palette loaded into your InterfaceBuilder (and of course a VDOboard).
This is what we are set out to create.
Start a new application in InterfaceBuilder. (Command-n)
1. Delete the window supplied.

2. Drag an SMControl object from the SMPalette into the object suitcase.
1-18 Concepts - Examples

Video Developer Kit
2. Drag a SMWindow onto the screen.

3. Adjust its size to 498 by 338.
4. Drag a SMPaletteView into the window and size it to 640 by 480.
5. Drag a connection from the SMView to the SMControl object and connect the views's
smControl outlet to the object:
6. Choose the SMView, bring up its Inspector (Command-1) and change its attributes to:
7. Group the view, change the title to 'No Title', offsets of zero and to a bezeld border.
8. Choose Size To Fit in Layout menu.
9. Group in ScrollView. Adjust the ScrollView to be just the size of the SMView.
10. Change size of ScrollView to 200 by 180.
11. Put a slider and a CheckButton (Call it mute) into the Window.
12. Drag a connection to the SMView and connect the target to the setAudioVolume:
target method.
13. Drag a connection from the button to the view and connect the target to the
setAudioMute: target method.
14. Group them and title it Audio.
15. Place it below the ScrollView.
16. Add a radio matrix of 2 by 2 to the window. Name them Black, Red, Yellow and
S-VHS and assign them tags 0-3 respectively.
17. Drag a connection from the matrix to the SMView and connect the target to the
setVideoInput: target method.

Video Developer Kit
18. Group the matrix and title it Input. Place it below the Audio box.
18. Create two buttons and call them 'Start' and 'Stop'. Place them next to the Input box.
19. Drag a connection from the 'Start' button to the SMView and connect the target to the
start: target method.
20. Drag a connection from the 'Stop' button to the SMView and connect the target to the
stop: target method.
21. Group everything (ScrollView, Audio box, Input box and the two buttons). Title the
group Video in a ScrollView.
22. Copy the complete group and place it next to it on the right side.
23. Rename it to Scalable Video.
24. Select the SMView and size it to 200 by 180
25. Bring up its Inspector (with Command-1) and change its attributes to:
26. Select the ScrollView and choose ungroup (Command-G) from the menu.
27. Adjust the view so it is placed correctly.
28. SAVE
29. You are ready. Hit Command-r to try it.
Note: For resizing to work correctly change the groups to be resizeable.
3. SplitIt
SplitIt is a application with the following features :
• Uses split memory mode

• Allows to change the spilt memory mode settings online
The boards of ///FAST use the YUV color space. The SM has a color depth of 4:2:2. This means
that a sequence of 4 pixel has 4 Y (luminance), 2 U and 2 V (both chrominance) components,
which leads to 8 components for a sequence of 4 pixel ( = 2 components for 1 pixel ). A 4:2:2
YUV color space is compared with a 24bit RGB color space. The YUV modes are described in
Appendicies/A_.. and Appendicies/B_.. .
If you use higher screen resolutions (scaled switch is marked in Configure.app->setup) the SM
scales the live video down to the half of its width. This has nothing to do width the memory the
card has on board. This is only a problem of memory speed. Due to that fact the SM grabs only
one half of the image information and the kit software doubles the width again (-> you get only
the half image quality). One solution for that problem would be switching video scaling off just
right before grabbing an image (setHorizontalScaled:NO). This will work quite well if you have
set setVideoOn:NO before but does not solve the problem grabbing a already stopped video
frame. In that case you have a "down scaled" image in the SM memory which can't be improved
through scale switching.
Due to the reduced quality problem we integrated the split memory mode.
In split memory mode the memory is divided into two regions:

1. A "visible" region (upper mem) which contains the image data that is displayed on screen. If
1-20 Concepts - Examples

Video Developer Kit
you size the live video larger than the size of this region the display quality will get worth be-
cause the video will be scaled up to that size (same problem when sizeing the video larger than
640x480 in SM_MEM_NORM).
2. An "invisible" region (lower mem) in which the video data is composed in full quality (with-
out scaling more than necessary to the recommended size). This happens whenever the card gets
a grab or stop (still...) message. This "unscaled" data can be accessed using a readImage: meth-
od.
The split memory mode is designed to be used further on in the same application, when once
set . You normally set it once at application startup. That's why you have to set the maximum
width in a setMemoryMode::: call. The maximum width in combination with the offset speci-
fies the maximum image size you can set ( in setSplitMemGrabSize: ) and request. Because of
internal card memory management the maximum width has to be a multiplier of 16.
Calculate the offset :

offset=(width*2) * height
This equals the way you would calculate the memory amount a RGB picture would take : rgb_-
mem=(width*3) * height. Remember (as described above) that in YUV 4:2:2 you have 2 Bytes
per pixel.
In your case the maximum width shouldn't be more than 640 (appropriate for NTSC) and with
a maximum height of 480 you get an offset of 614000. 434176 Bytes of memory are left for the
upper mem. If you are using Hi-Res (horizontalScaled==YES) the SM needs only half memory
size (as described above) for displaying live video. That means we can calculate the upper mem
as:
maximum_lossless_size=((width/2)*2)*height=width*height
maximum_lossless_size<=434176=upper_mem
To retain the NTSC ratio = 640/480=4/3 we get :

width*height<=upper_mem
(height * (4/3))*height<=upper_mem
height<=sqrt( 3/4 * upper_mem)
In this case :
height=570 => width=760
This is more than we need for displaying a NTSC signal.
When using horizontalScaled==NO we have to double the width for the calculation and we get :
height=403, width=537
Do not misunderstand this result. You can size the display video larger but you get only image
information for that size. For every larger size the video is scaled up (loss of information and
quality implied).
Conclusion : The split memory mode offers the opportunity to grab images in different sizes
and if horizontalScaled is used, in better quality than the live video on your monitor has. We
know that this calculations are not easy to explain and to understand but for a further enhance-
ment of the split memory mode (split the memory in more than 2 regions) it is necessary. We're
thinking about the implementation of additional methods in the kit to make the handling easier.
But even with "high level" methods you have to understand what's going on in the SM memory
to get the results you want to have.

Video Developer Kit
Animations
Since Steven Jobs has introduced NEXTIME at the NeXTWORLD EXPO in 1993, there has
been a great push for video from the harddisk. The VDOboards driver is capable of both record-
ing live video and playing back live video from harddisk, either in real time or at full resolution.
Why not both? Well a 10 second video spot (NTSC which is 640x480 at 30 frames a second)
would be 176 MB. This exceeds what an ISA bus can handle by an order of magnitude. Even if
it would be possible, it isn't very desirable. Who has a 6 gigabytes for an hour of video? So the
only reasonable solution is compression. For real time recording at the full resolution, the com-
pression has to be on board. This is something we support with the CODY option for the Screen-
Machine II and MovieMachine Series ( a JPEG compressor/decompressor board). The
compressed video is a factor of 10 to 20 times smaller. The problem with this kind of compres-
sion algorithm, is that it is asymmetrical. Meaning that it takes longer to compress the video than
to decompress. This is why NeXT has developed a symmetrical algorithm based on wavelets.
Not only is it faster, but the quality is also higher at the same rate of compression. Another nice
feature is that the lossyness and hence the data rate can be adjusted during decompression, in-
dependent of the settings at compression time. This means that the video can be compressed at
full quality and played back at a lesser quality on less powerful machines from the same set of
data. Together with the CODY option and our VDOconverter productr you can record JPEG
movies and transcode them into a wavlet compressed animation, capable of being played back
using NEXTIME.
The next problem is to define a framework to play back the video with sound. Video and sound
have to be synchronized and if the video falls back, frames have to be skipped and the frame
rate or quality have to be adjusted. It has to define a file format, or a set of file formats supported.
NEXTIME will be just such a framework. It supplies a compression technique, allowing other
compression algorithms to be integrated. It uses the Quictime file format, and can be extended
to use others like AVI. But most importantly it handles the play back and recording with sound.
Now it would be possible to come up with a similar, possibly even compatible framework, but
the time it would take to develop such a thing exceeds the time in which NEXTIME will hope-
fully become available from NeXT. When that happens, we will supply the adaptors and other
software to integrate the VDOboards driver into NEXTIME. Until then it is not really possible
to grab sequences of video and play them back from harddisk.
Latest word we have, is that the neccessary tools for NEXTIME will become available with
NEXTSTEP 4.0
1-22 Concepts - Animations

Chapter 2
Reference
Video Developer Kit
Classes Overview
VDOboards
Kernelspace
Userspace
SMKit
SMControl MMControl
Object Responder Window SMWindow
View SMView
NXImageRep SMYUVImageRep SMFLMImageRep
SMTVControl
SMTV_AFC_Table
SMTV_AFC_TableList
The Kernel / Driver

The kernel is a driverKit driver that is configured using the standard Configure.app. It is the low
level connection to all boards. One kernel can service all installed boards. Access to the kernel
is restricted to the SMControl class.
The SMControl class

The SMControl class is the connection to the board on the user level side. It controls all the fea-
tures of the board as well as the access to the individual boards. There has to be one instance per
connection to a board. This means if you have 3 Screen Machines II (of a maximum of 4) in
your computer you have to have 3 instances of the SMControl class to use all 3 of them. It is
not advised, though possible, to have more than one instance of a SMControl class control the
same board (as in two separate applications) as this can lead to some unpredictable results. For
the moment the Screen Machine is a non-shareable resource.
MMControl
The MMControl class extends the SMControl class to support the MovieMachine video out ca-
pabilities.
2-24 Reference - Classes Overview

Video Developer Kit
SMControlAudio
These methods control the optional audio hardware that can be connected to or be on board a
VDOboard. This is either the optional Audio-on-Bracket, a TV-Tuner or the integrated audio of
the Movie Machines.
SMControlChooserController
The SMControlChooserController is a simple way to let the user choose which VDOboard to
use if multiple are present in the system.
SMWindow
SMWindow is a special subclass of Window that handles events necessary to ensure correct be-
havior when a window is moved, hidden or miniaturized. The window has two special features,
one is that it allows the window to be dragged by the content view and the other that the icon of
the window has the live video displayed in it.
SMView
SMView is a subclass of View that handles the display of live video on screen. It sets the posi-
tion and size of the live video overlay and makes printing possible. It also implements
drag&drop in the form that TIFF, EPS and FLM images can be dropped on the view, which will
then be displayed in the video view by a video board.
SMPaletteView
SMPaletteView is a subclass of SMView that has a number of action methods corresponding to
SMControl methods, allowing the direct connection to outlets in InterfaceBuilder. In this way
it is possible to develop simple applications in IB without having to write a line of code.
SMPalette
The SMPalette is an InterfaceBuilder palette with a SMWindow, SMPaletteView and a SM-
Control class. With it simple applications can be built in minutes and run just using Interface-
Builder.
SMYUVImageRep, SMFLMImageRep, MMFLMImageRep

SMYUVImageRep + SMFLMImageRep are subclasses of NXImageRep that store the image
data from the Screen Machine in YUV format. The SMFLMImageRep also saves the data in the
FLM image format.
SMTVControl
SMTVControl is for the onboard tuners of the MovieLine or the optional tuner for the Screen-
MachineII. It hast methods for tuning channels as well as reading predefined channel tables.
SMTV_AFC_Table
SMTV_AFC_Table is a class for holding afc tables for specific countries and television stand-
ards. This class can read the supplied afc tables ( in Resources/AFCTables).
SMTV_AFC_TableList
SMTV_AFC_TableList is a class that read the supplied Countries.afcTable table and allows
user selection by county and television standard.
Reference - Classes Overview 2-25

Video Developer Kit
VDOboards
Inherits From: none
Declared In: <smkit/VDOboards.h>
Class Description
This is an abstract class, as in that it has no instance methods. Its sole purpose is to return an
initialized instance of the appropiate control class. (In the moment MM or SMControl). It also
provides methods to query the VDOboards present. To make it possible to develop applications
that are indpendant of a specific board present, we now support querying and initializing boards
by a list of features. The VDOboards driver already supports over 5 different boards, with a sub-
stantial set of common functionality. In the ever changing world of PC hardware, this ensures
that your application will probaly run on the newest board when it becomes available. So we
strongly encourage you to use the feature flags instead of hardcoding your application to a spe-
cific board type.
Method Types
Factory methods + boardAvailable:
+ boardsPresentOfType:withFeatures:
+ closeAllBoards
+ getFeaturesOfBoard:
+ getKernelVersion:minor:
+ newForBoardNum:
+ newForBoardNum:fromZone:
+ newWithSelectionAndBoards:withFeatures:
+ newWithSelectionAndBoards:with-
Features:fromZone:
Class Methods
boardAvailable:
+ (BOOL)boardAvailable:(short)num
Returns YES if the VDOboard with the number num is available.
2-26 Reference - VDOboards

Video Developer Kit
boardsPresentOfType:withFeatures:
+ (unsigned int)boardsPresentOfType:(unsigned int)boardType
withFeatures:(unsigned int)features
Returns the number of VDOboards present in the system that match the boardType and features.
Possible values for boardTypes are:
VMCBOARD_ALL_BOARDS If the type of board does not matter (use this together
with the features flag).
VMCBOARD_MMPRO For a Movie Machine PRO
VMCBOARD_SMII8 For a Screen Machine II
VMCBOARD_MM For a Movie Machine
VMCBOARD_MMII For a Movie Machine II
VMCBOARD_FPS60 For a FPS60 board.
The different board types can be ORed together for a greater selection.
Alternatively one can specify VMCBOARD_ALL_BOARDS and a list of features the board
should support.
Possible flags for features are:
VMCBOARD_INPUTS The minimum number of inputs the boards
should have (0-7)<<8
i.e. 0x0300 for three inputs or more
VMCBOARD_OUTPUTS The number of inputs the boards should have (0-7) << 16
i.e. 0x020000 for two outputs or more.
Additionally the type of each input or output can be specified
VMCBOARD_INPUT_SVHS If any of the inputs supports SVHS
VMCBOARD_INPUT_NTSC If any of the inputs support NTSC sources.
VMCBOARD_INPUT_PAL If any of the inputs support PAL sources.
VMCBOARD_INPUT_SECAM If any oft he inputs support SECAM sources.
VMCBOARD_OUTPUT_SVHS If any of the outputs supports SVHS
VMCBOARD_OUTPUT_NTSC If any of the outputs support NTSC sources.
VMCBOARD_OUTPUT_PAL If any ofthe outputs support PAL sources.
VMCBOARD_OUTPUT_SECAMIf any of the outputs support SECAM sources.
And there is a list of special features each board supports.
VMCBOARD_AUDIO That the board has audio output (usually from the Tuner).
In case of the Screen Machine II this is the optional audio.
VMCBOARD_TUNER That the board has a Tuner.
Reference - VDOboards 2-27

Video Developer Kit
VMCBOARD_VT If the board has Videotext (also referred to as teletext).

VMCBOARD_OVERLAY If the board supports overlay (or inlay).
VMCBOARD_422 If the board uses 4:2:2 YUV sampling.
VMCBOARD_SQP If the board has square vs CCIR 601 pixels.
VMCBOARD_FRAMES If the board has enough memory to store full PAL
frame instead of just fields.
VMCBOARD_ALPHA If the board supports superimpsoing two video channels
on video out with alpha. (Currently only supported
by MM II)
The individual feature flags can be ORed together to select a more specific board.
If you want a specific board, specify VMCBOARD_ANY_FEATURES as the feature flag.
Example:
// Select any board that has at least 2 inputs and one output, a tuner
and that uses 4:2:2 YUV sampling.
int num = [VDOboards boardsPresent:VMCBOARD_ALL_BOARD

features:0x0200|0x010000|VMCBOARD_TUNER|VMCBOARD_422];
// Select any board that has at least 2 inputs and where one of the
inputs support SVHS
int num = [VDOboards boardsPresent:VMCBOARD_ALL_BOARD

features:0x0200|VMCBOARD_INPUT_SVHS];,
closeAllBoards
+ (void)closeAllBoards
Resets the usage count of all boards to 0. Essentially making all boards available again. The use
of this methods should not be necessary, as the usage count is decermented automatically when
a control class is freed, or when the PID that has last locked the board is not present anymore.
getFeaturesOfBoard:
+ (unsigned int)getFeaturesOfBoard:(short)num
Returns the features associated with the board number num.
2-28 Reference - VDOboards

Video Developer Kit
getKernelVersion:minor:
+ getKernelVersion:(short*)major minor:(short*)minor
Returns the major and minor version of the VDOboards driver.
newForBoardNum:
+ newForBoardNum:(unsigned int)i
Returns a newly alloacted and initialized instance of the appropiate control class, if the board is
present in the system. Regradless if the board is available or not. This is either an instance of
the SMControl or MMControl class.
newForBoardNum:fromZone:
+ newForBoardNum:(unsigned int)i fromZone:(NXZone*)zone
Same a +newForBoardNum, except that the zone from which the control class is allocated can
be specified.
newWithSelectionAndBoards:withFeatures:
+ newWithSelectionBoard:(unsigned int)boardType
Returns a newly alloacted and initialized instance of the appropiate control class, according to
the boardType and features specified. If more than one board os the requested type (and fea-
tures) is present in the system, the user is presented with a modal panel, to select the board to
use. If present in the same bundle as the calling class the SMMultiChoose.nib file is loaded from
that bundle. If this .nib file is not present it will NOT display a panel and return the first avail-
able board instead.
See also: - boardsPresentOfType:withFeatures:
newWithSelectionAndBoards:withFeatures:fromZone:
+ newWithSelectionBoard:(unsigned int)boardType
fromZone:(NXZone*)zone
Same as +newWithSelectionBoard:withFeatures:, except that the NXZone zone can be spec-
ified from which the instnce of the control class will be instantiated.
See also: newWithSelectionBoard:withFeatures:
Reference - VDOboards 2-29

Video Developer Kit
SMControl
Inherits From: Object
Declared In: smkit/SMControl.h
Class Description
The SMControl class is the connection to the VDOboards driver. We have chosen not to docu-
ment the driver interface to ensure compatibility even when the driver changes due to new op-
erating system releases or performance tuning. Each instance of the class controls one
VDOboard. The class provides access to all available functions of the board.
It can (among other things):
• Control the size and location of the video on screen.
• Control the appearance of the video on screen (color, brightness, contrast, ...).
• Control the optional audio hardware.
• Select the video input.
• Select TV system (NTSC, PAL, SECAM).
• Acquire single images or sequences from live video.
• Various digital effects (mosaic, posterization, chroma-keying, inverting, ...).
• Adjust the VDOboard overlay to the graphics system.
It is possible to write applications using just an instance of this class. To provide easier handling
of the NEXTSTEP environment there are implementations of subclasses of the standard View
and Window classes, as well as NXLiveVideoView compatible class. These are respectively the
SMView and SMWindow classes. For further details, please refer to the documentation of the
named classes.
2-30 Reference - SMControl

Video Developer Kit
Ranges and Default Values

Neutral
Min Max Off/Auto Default Stepsize Comments
Red gain 0.00 2.00 1.00 1.00 0.031 Adds or subtracts red from image
Green gain 0.00 2.00 1.00 1.00 0.031 Adds or subtracts green from image
Blue gain 0.00 2.00 1.00 1.00 0.031 Adds or subtracts blue from image
Output Stage (do
Brightness 0.00 1.00 0.65 0.65 0.016
not affect Quality
grabbed images) Saturation 0.00 1.00 0.67 0.65 0.016
Contrast 0.00 1.00 0.48 0.49 0.016
Hue 0.00 1.00 0.00 0.00 0.004 0.0 = 0˚, 1.0 = 360˚
Sharpness 0.00 1.00 0.50 0.50 0.016
Chroma intensity 0.00 1.00 0.50 0.52 0.010 Shifts lookup table for chroma values
Quality
Luma intensity 0.00 1.00 0.50 0.46 0.010 Shifts lookup table for lume values
Noise filter 0.00 1.00 0.00 0.00 0.250 Corrects for noisy input. (0 = Auto)
InputStage (affect Input filter 0.00 1.00 0.00 0.00 0.200 Corrects scaling of lines. (0 = Off, 0.2 = Auto)
Filters
grabbed images) Bandpass 0.00 1.00 0.00 0.00 0.250 Enables bandpass filtering. (0 = Off)
Prefilter 0.00 1.00 0.00 1.00 1.000 Enables prefiltering for a softer image
Mosaic 1 32 1 0.00 1 Size of square which pixels are summed. (0 = Off)
Effects
Posterization 0.00 1.00 0.00 0.00 0.143 Reduces ammount of distinct colors/shades in image
Volume 0.00 1.00 0.00 0.00 0.010 Volume
Fader 0.00 1.00 0.50 0.50 0.010 Distribution front to back (0 = front, 1 = back)
Audio Balance 0.00 1.00 0.50 0.50 0.010 Distribution left to right (0 = left, 1 = right)
Treble 0.00 1.00 0.50 0.50 0.010
Bass 0.00 1.00 0.50 0.50 0.010
Instance Variables
port_t smPort;
NXRect videoFrame;
NXRect windowFrame;
NXRect zoomFrame;
NXSize grabSize;
short screenMachineNum;
List *viewList;
id activeView;
int port_address;
struct _Setup {
short pll;
unsigned char system:2;
unsigned char interlaced:1;
unsigned char scaled:1;
unsigned char hasaudio:1;
short xoffset;
short yoffset;
} Setup;
struct _Mode {
unsigned char mixer:1;
unsigned char live:1;
unsigned char video:1;
unsigned char input:2;
unsigned char color:1;
unsigned char prefilter:1;
unsigned char timebase:1;
unsigned char invertChroma:1;
Reference - SMControl 2-31

Video Developer Kit
unsigned char invertLuma:1;

unsigned char flip:1;
unsigned char frametype:2;
unsigned char memory:2;
} Mode;
struct _Clipping {
unsigned char isClipping;
} Clipping;
struct _ImageQuality {
unsigned char red;
unsigned char green;
unsigned char blue;
unsigned short brightness;
unsigned char saturation;
unsigned char hue;
unsigned char contrast;
unsigned char sharpness;
unsigned char posterization;
unsigned char bandpassfilter;
unsigned char noisefilter;
char inputfilter;
char lumaintensity;
char chromaintensity;
} ImageQuality;
struct _Image {
char mode;
char type;
NXSize size;
} Image;
struct _Audio {
unsigned char input;
unsigned char mute;
unsigned char volume;
char balance;
char fader;
char treble;
char bass;
} Audio;
smPort Port to kernel server
videoFrame Size of video acquisition
windowFrame Size of displayed video on screen
zoomFrame Size of crop frame relative to output
grabSize Size of grabbed image
screenMachineNum Number of VDOboard, the control class is acting on
viewList List object with views connected to control class
activeView Currently active view
port_address Base address of boards for port IO
Setup.pll Phased Lock Loop, controls width and position of video
Setup.system System (PAL,NTSC,SECAM)

Video Developer Kit
Setup.interlaced Interlaced video

Setup.hasaudio Audio present
Setup.xoffset Horizontal offset of video
Setup.yoffset Vertical offset of video
Mode.mixer Mixer state
Mode.live Video live state
Mode.video Video on state
Mode.input Current input
Mode.color Color or BW display
Mode.prefilter Input prefiltered
Mode.timebase Timebase correction
Mode.invertChroma Chroma inverted
Mode.invertLuma Luma Inverted
Mode.flip Picture upside down
Mode.frametype Number and type of fields
Mode.memory Split memory mode
Clipping.isClipping Masking of windows
ImageQuality.red Red color gain
ImageQuality.green Green color gain
ImageQuality.blue Blue color gain
ImageQuality.lumaintensity Intensity of luma
ImageQuality.chromaintensityIntensity of chroma
ImageQuality.brightness Brightness
ImageQuality.saturation Saturation
ImageQuality.hue Hue
ImageQuality.contrast Contrast
ImageQuality.sharpness Sharpness
ImageQuality.posterization Posterization
ImageQuality.bandpassfilter Bandpass filter
ImageQuality.noisefilter Noise filter
ImageQuality.inputfilter Input filter
Image.mode Format of image data
Image.type Type of image rep
Image.size Size of image
Audio.input Audio input
Audio.mute Audio muted
Audio.volume Audio volume
Audio.balance Audio balance
Audio.fader Audio fader
Audio.treble Audio treble
Audio.bass Audio bass

Video Developer Kit
Method Types
Init - init
- init:
- reset
- free
- smNum
- getKernelVersion: minor:
- getSerial:
Setup - setPll:
- pll
- setSystem:
- system
- setInterlaced:
- isInterlaced
- setHorizontalScaled:
- isHorizontalScaled
- setHorizontalOffset:
- horizontalOffset
- setVerticalOffset:
- verticalOffset
- setVideoFrame:
- getVideoFrame:
- setWindowFrame:
- getWindowFrame:
Mode - setInputFrameType:
- inputFrameType
- setVideoInput:
- videoInput
- setStill:
- isStill
- setMixerOn:
- isMixed
- setVideoOn:
- isVideoOn
- setColorOn:
- isColor
- setPreFilter:
- isPreFiltered
- setVCRTimebase:
- isVCRTimebase
- setChromaInvert:
- isChromaInverted
- setLumaInvert:
- isLumaInverted
- setFlipped:
- isFlipped
- inputType:

Video Developer Kit
- outputType:
- numberOfInputs
- numberOfOutputs
Imagequality - setRedGain:
- redGain
- setGreenGain:
- greenGain
- setBlueGain:
- blueGain
- setLumaIntensity:
- lumaIntensity
- setChromaIntensity:
- chromaIntensity
- setBrightness:
- brightness
- setSaturation:
- saturation
- setHue:
- hue
- setContrast:
- contrast
- setSharpness:
- sharpness
- setBandpassFilter:
- bandpassFilter
- setNoiseFilter:
- noiseFilter
- setInputFilter:
- inputFilter
Acquisition - getImageSize:
- readImage
- writeImage:
- readImageSelection:
- setMemoryMode:::
- memoryMode
- setSplitMemGrabSize:
- setImageType:
- imageType
Effects - setPosterization:
- posterization
- setMosaicWidth:
- mosaicWidth
- setZoomFrame:
- getZoomFrame:
Audio (In SMControlAudio Category) - hasAudio
- setAudioMute:

Video Developer Kit
- isAudioMute
- setAudioInput:
- audioInput
- setAudioVolume:
- volume
- setAudioBalance:
- balance
- setAudioFader:
- fader
- setAudioTreble:
- treble
- setAudioBass:
- bass
Instance Methods
bandpassFilter
- (float)bandpassFilter
Returns current value of bandpass filter, a float between 0.0 and 1.0 in steps of 0.25. The default
is OFF (0.0).
See also: - setBandpassFilter:
blueGain
- (float)blueGain
Returns value of the blue gain in effect, a float between 0.0 and 2.0, in steps of 0.032, with a
default of 1.0. A value of 1.0 is neutral and means that no alteration is in effect. Beware that
values greater than 1.0 will cause the SMFLMImageRep to no longer accurately convert the im-
ages from FLM to TIFF.
See also: - setBlueGain:
brightness
- (float)brightness
Returns value of the brightness in effect, a float between 0.0 and 1.0, in steps of 0.016, with a
default of 0.65. A value of 0.65 is neutral and means that no alteration is in effect.
See also: - setBrightness:

Video Developer Kit
chromaIntensity
- (float)chromaIntensity
Returns value of the chroma intensity in effect, a float between 0.0 and 1.0, in steps of 0.010,
with a default of 0.52.
See also: - setChromaIntensity:
contrast
- (float)contrast
Returns value of the contrast in effect, a float between 0.0 and 1.0, in steps of 0.016, with a de-
fault of 0.49. A value of 0.49 is neutral and means that no alteration is in effect.
See also: - setContrast:
free
- free
Frees the SMControl class and releases VDOboard for use with another SMControl class. It also
unregisters all connected views. Since the SMControl is connected to hardware, it is important
that the class is freed with this method, and not by the destroy zone mechanism used by the App-
Kit. To ensure proper handling call this method in the terminate: method of the Application
class.
See also: - init,-init:,-initWithSelection
getImageSize:
- getImageSize:(NXSize *)size
Assigns size the proportions that an image would have if it would currently be read from the
VDOboard.
See also: - setSplitMemGrabSize,-setWindowFrame
getKernelVersion: minor:
- getKernelVersion:(short *)major minor:(short *)minor
Sets the two variables passed to the major and minor revision of the kernel.
e.g. if major = 2 and minor = 4 the kernel revision is 2.4.
See also: + getKernelVersion:...

Video Developer Kit
getSplitModeGrabSize:
- getSplitModeGrabSize:(NXSize *)grabSize
Assigns grabSize the currently set size for grabbed images in lower memory.
See also: - setSplitMemGrabSize:
getVideoFrame:
- getVideoFrame:(NXRect *)theRect
Assigns the currently active video frame to theRect.
See also: - setVideoFrame:
getZoomFrame:
- setZoomFrame:(NXRect *)aRect
Assigns aRect the currently active zoom frame. Returns self.
See also: - setZoomFrame:
getWindowFrame:
- getWindowFrame:(NXRect *)aRect
Assigns the currently active window frame to aRect.
greenGain
- (float)greenGain
Returns value of the green gain in effect, a float between 0.0 and 2.0 with a default of 1.0. A
value of 1.0 is neutral and means that no alteration is in effect. Beware that values greater than
1.0 will cause the SMFLMImageRep to no longer accurately convert the images from FLM to
TIFF.
See also: -blueGain:,-redGain,- setBlueGain:,- setGreenGain,- setRedGain
horizontalOffset
- (unsigned short)horizontalOffset
Returns the horizontal offset between the coordinate system of the VDOboard and the screen
coordinate system.
See also: - setHorizontalOffset:

Video Developer Kit
hue
- (float)hue
Returns the current value of hue in the range of 0.0 to 1.0 with a default of 0.0.
See also: - setHue:
imageType
-(char)imageType
Returns the type of ImageRep used by the readImage: and readImageSelction: methods. Pos-
sible values are SM_IMAGE_YUV and SM_IMAGE_FLM.
See also: - setImageType:,- readImage:,- readImageSelection:
init
- init
Looks for the next available VDOboard and initializes it to the default values. If no VDOboard
is available it returns nil.
init:
- init:(short)boardNum
Initializes VDOboard with number boardNum. If the VDOboard with number boardNum is not
available or not present it returns nil.
inputFilter
- (float)inputFilter
Returns current value of input filter in the range of 0.0 to 1.0 in steps of 0.200. A value of 0.0
means that automatic input filtering is enabled, a value of 0.200 that input filtering is disabled
and anything above different levels of input filtering.
See also: - setInputFilter:
inputFrameType
- (unsigned char)inputFrameType
Returns currently active input frame type, one of SM_FRAME_BOTH, SM_FRAME_EVEN
or SM_FRAME_ODD.
See also: - setInputFrameType:

Video Developer Kit
inputType
- (unsigned int)inputType:(unsigned int)inputNum
Returns the type of input, input inputNum is, one of TYPE_FBAS, TYPE_SVHS or TYPE_F-
BAS_SVHS. If a input is of type TYPE_FBAS_SVHS (i.e. supports both FBAS and SVHS)
you select the FBAS input by setting input inputNum, and the SVHS input by setting inputNum
+ number of inputs the board has.
See also: - numberOfInputs
isChromaInverted
- (BOOL)isChromaInverted
Returns YES if input chroma values are being inverted.
See also: - setChromaInverted:
isColor
- (BOOL)isColor
Returns YES if color is enabled. Says nothing about whether color source is connected or if a
color image is actually displayed. Returns NO if color is disabled.
See also: - setColorOn:
isFlipped
- (BOOL)isFlipped
Returns YES if vertical axis of image is flipped, otherwise NO; is mutually exclusive with
setMosaicWidth:.
See also: - setFlipped:, - setMosaicWidth:
isHorizontalScaled
- (BOOL)isHorizontalScaled
Returns YES if horizontal scaling is active, otherwise NO.
See also: - setHorizontalScaled:

Video Developer Kit
isInterlaced
- (BOOL)isInterlaced
Returns YES if interlaced input mode is selected. Otherwise returns NO.
See also: -setInterlaced:
isLumaInverted
- (BOOL)isLumaInverted
Returns YES if input luma values are being inverted.
See also: -setLumaInverted:
isMixed
- (BOOL)isMixed
Returns YES if mixer in the output stage is active, otherwise NO.
See also: -setMixerOn:
isPreFiltered
- (BOOL)isPreFiltered
Returns YES if input is prefiltered, otherwise NO.
See also: -setPreFiltered:
isStill
- (BOOL)isStill
Return YES if live video is halted, otherwise NO. Video is halted by setStill: or when an image
is loaded into the VDOboard.
isVCRTimebase
- (BOOL)isVCRTimebase
Returns YES if timebase correction for VCRs is enabled, otherwise NO.

Video Developer Kit
isVideoOn
- (BOOL)isVideoOn
Returns YES if live video is activated, otherwise NO. Says nothing about whether the correct
input is selected or if a valid source is connected.
See also: -setVideoOn:
lumaIntensity
- (float)lumaIntensity
Returns value of the luma intensity in effect, a float between 0.0 and 1.0, in steps of 0.010, with
a default of 0.46.
See also: -setLumaIntensity:
memoryMode
- (unsigned char)memoryMode
Returns currently active memory mode. One of SM_MEM_NORM or SM_MEM_SPLIT.
See also: -setMemoryMode:
mosaicWidth
- (short)mosaicWidth
Returns currently active mosaic width in the range of 1 to 16.
See also: -setMosaicWidth:
noiseFilter
- (float)noiseFilter
Returns current value of noise filter.
See also: -setNoiseFilter:
numberOfInputs
- (unsigned int)numberOfInputs
Returns the number of inputs the board has.

Video Developer Kit
numberOfOutputs
- (unsigned int)numberOfOutputs
Returns the number of outputs the board has.
noiseFilter
- (float)noiseFilter
Returns current value of noise filter.
See also: -setNoiseFilter:
outputType:
- (unsigned int)outputType:(unsigned int)outputNum
Returns the type of output, output outputNum is, one of TYPE_FBAS, TYPE_SVHS or
TYPE_FBAS_SVHS. If a output is of type TYPE_FBAS_SVHS (i.e. supports both FBAS and
SVHS) you select the FBAS output by setting output outputNum, and the SVHS output by set-
ting outputNum + number of outputs the board has.
NOTE: Most boards only have 1 output, or if they have 2 outputs then they
are active simultaneously.
See also: - numberOfOutputs
pll
- (unsigned short)pll
Returns value for phased locked loop of output mixer in the range of 0 to 1024.
See also: -setPll:
posterization
- (float)posterization
Returns current value of posterization in the range of 0.0 to 1.0 in steps of 0.143. 0.0 means no
posterization is in effect.
See also: -setPosterization:

Video Developer Kit
readImage
- (NXImage *)readImage
Returns a newly allocated NXImage with a ImageRep representation of the current live picture
displayed by the VDOboard. Whether the image contains a YUV or FLMImageRep is set by
the setImageType: method. This is not a shared image, it is up to the application to free the
returned NXImage.
See also: -readImageSelection:, -writeImage:
readImageSelection:
- (NXImage *)readImageSelection:(NXRect)imageRect
Returns a newly allocated NXImage containing a selection, specified by imageRect, of the cur-
rent live picture displayed by the VDOboard. Whether the image contains a YUV or SM-
FLMImageRep is set by the setImageType: method. This is not a shared image, it is up to the
application to free this representation. The imageRect is specified in the window frame coordi-
nate system, meaning that the lower left is the origin and that the maximum width and height
are those of the current window frame.
See also: -readImage:, -writeImage:
redGain
- (float)redGain
Returns value of the blue gain in effect, a float between 0.0 and 2.0 with a default of 1.0. A value
of 1.0 is neutral and means that no alteration is in effect. Beware that values greater than 1.0 will
cause the FLMImageRep to no longer accurately convert the images from FLM to TIFF.
See also: -setRedGain,- setGreenGain, -setBlueGain:
reset
- reset
Resets values of the VDOboard to defaults.
See also: -init:,-init,-initWithSelection
saturation
- (float)saturation
Returns value of the saturation in effect, a float between 0.0 and 1.0, in steps of 0.016, with a
default of 0.65. A value of 0.50 is neutral and means that no alteration is in effect.
See also: -setSaturation:

Video Developer Kit
setBandpassFilter:
- setBandpassFilter:(float)val
Sets value of bandpass filter to val. The range is 0.0 to 1.0, in steps of 0.25. 0.0 turns bandpass
filtering off.
See also: -bandpassFilter:
setBlueGain:
- setBlueGain:(float)val
Changes the output image blue component according to val. The range is from 0.0 to 2.0, in
steps of 0.031, with 1.0 being neutral. A smaller value means less blue and a larger more blue
in the picture. Has no effect on the image in memory of the VDOboard.
See also: -blueGain, -redGain, -greenGain, -setGreenGain:, -setRedGain:
setBrightness:
- setBrightness:(float)val
Sets brightness of output image according to val. The range is from 0.0 to 1.0 in steps of 0.016,
with 0.65 being neutral. A smaller value means a darker picture and a larger a lighter picture.
Has no effect on the image in memory.
See also: -brightness
setChromaIntensity:
- setChromaIntensity:(float)val
Sets the chroma intensity of the input decoder to val. A higher value means a more saturated
representation, and a lower a colorless/grayscale representation. The range is from 0.0 to 1.0 -
neutral is 0.50 and the default 0.52. Belongs to the set of controls affecting the input stage and
hence how the image is digitized into memory.
See also: -setBandpassFilter:
setChromaInvert:
- setChromaInvert:(BOOL)state
If state is YES, the chroma values of the input will be inverted. Affects the image stored in
memory. If state is NO, the input remains unaltered.
See also: -isChromaInverted

Video Developer Kit
setColorOn:
- setColorOn:(BOOL)state
If state is YES, the chroma of the live video is enabled. For a color picture to appear, a color
source has to be connected, the correct system has to be set, and the quality controls have to be
set accordingly. (Notably the chroma intensity and saturation.) Another reason for a lack of
color is if the black input is selected with a SVHS source connected. If state is NO, color is dis-
abled. The video can be color regardless of the current WindowServer depth, as long as a color
monitor is connected.
See also: isColorOn
setContrast:
- setContrast:(float)val
Sets contrast of output image according to val. The range is from 0.0 to 1.0 in steps of 0.016,
with 0.49 being neutral. A smaller value reduces the contrast and a larger a higher contrast of
the picture. Has no affect on the image in memory.
See also: -contrast
setFlipped:
- setFlipped:(BOOL)state
If state is set to YES, the vertical axis of the live picture will be flipped at the input stage. This
results in the live picture and the image in memory being upside down. If state is NO, the image
will remain unchanged. Do not use when mosaic is active, since it will produce unpredictable
results.
See also: -isFlipped
setGreenGain:
- setGreenGain:(float)val
Changes the output image green component according to val. The range is from 0.0 to 1.0, with
0.5 being neutral. A smaller value means less green and a larger value more green in the picture.
It has no affect on the image in memory.
See also: -greenGain, -blueGain, -redGain, -setBlueGain:, -setRedGain:

Video Developer Kit
setHorizontalOffset:
- setHorizontalOffset:(unsigned short)val
Sets horizontal offset of the VDOboard coordinate system to coordinate system of the screen to
val. It is used to align the two coordinate systems.
See also: -horizontalOffset
setHorizontalScaled:
- setHorizontalScaled:(BOOL)state
Sets input decoder to half the width of the video image. Reduces the bandwidth of picture by
50%. Used on system above a screen resolution of 800x600 to allow the PLL to set to the correct
aspect ratio. Setting state to YES enables horizontal scaling, NO disables it.
See also: -isHorizontalScaled
setHue:
- setHue:(float)val
Sets hue of output image according to val. The range is from 0.0 to 1.0, with 0.0 being neutral.
0.0 corresponds to 0 change and 1.0 a 360 change in the color angle.
See also: -hue
setImageType:
- setImageType:(char)type
Sets type of ImageRep used by the readImage: and readImageSelection: method. Possible
values are SM_IMAGE_YUV and SM_IMAGE_FLM. Returns self.
See also: -imageType, -readImage, -readImageSelection:
setInputFilter:
- setInputFilter:(float)val
Sets value of input filter to val. Range is from 0.0 to 1.0, where 0.0 is the default. A value of 0.0
turns automatic input filtering on.
See also: - inputFilter

Video Developer Kit
setInputFrameType:
- setInputFrameType:(unsigned char)val
Sets input frame type to val. Possible values are SM_FRAME_BOTH, SM_FRAME_EVEN,
SM_FRAME_ODD. With the input frame, one can select if the image is generated only for the
odd or even fields, or from both fields. Both fields result in a better image resolution, but, on
images with a lot of movement, it leads to stripes in the image caused by movement in the image
between fields. For pictures with fast movement, one should select odd or even fields.
See also: -inputFrameType
setInterlaced:
- setInterlaced:(BOOL)state
Sets input decoder to interlaced mode. Effectively halves the bandwidth by halving the height
of the video image in memory. Only useful for interlaced video systems, which are currently
not supported by NEXTSTEP. If state is set to YES, interlaced mode is selected.
See also: - isInterlaced
setLumaIntensity:
- setLumaIntensity:(float)val
Sets the luma intensity of the input decoder to val. A higher value means a brighter representa-
tion and a lower a darker a representation. The range is from 0.0 to 1.0, neutral being 0.5 and
the default 0.46. Belongs to the set of controls affecting the input stage and hence how the image
is digitized into memory.
See also: - lumaIntensity
setLumaInvert:
- setLumaInvert:(BOOL)state
If state is YES, the luma values of the input will be inverted. Affects the image in memory of
the VDOboard. If state is NO, the input values will remain unchanged.
See also: - isLumaInverted

Video Developer Kit
setMemoryMode:
- setMemoryMode:(char)mode offset:(long)offset width:(short)width
Sets memory mode of the VDOboard to mode, where mode is one of SM_MEM_SPLIT or
SM_MEM_NORM. When mode is SM_MEM_NORM parameters offset and width are ig-
nored.
e.g. [control setMemoryMode:SM_MEM_NORM offset:0 width:0];
For the SM_MEM_SPLIT mode, offset and width specify the amount of memory set aside for
grabbing high quality images independant of the size of the live video view.
offset and width are in pixels. width should be set to the maximum width that the images will
have. (Set by setSplitMemGrabSize:). width has to be a multiple of 16 and has to be in the
range of 16 < width < videoFrame width.
offset = width * maximum height of grabbed images.
The Screen Machine II has room for 524288 pixels in memory (other boards: 262144). So the
offset has to be less than 524288. Also keep in mind that the Screen Machine II requires some
memory to display the live video. The remaining memory 524288 - offset will be used for this
purpose. If the offset is set to high the live video quality will degrade noticeably.
See the Concepts section of the manual for more details on the split memory mode.
Note: At the moment SM_MEM_SPLIT only works on Screen Machine, FPS60 and Movie Ma-
chine II boards.
See also: - setSplitMemGrabSize:
setMixerOn:
- setMixerOn:(BOOL)state
If state is YES the mixer of the output stage will be active, resulting in the data from the graphics
card being mixed with the data from the live video. Live video will be displayed where the
graphics are black, and no live video where the graphics are white. Useful if graphics should be
visible without clipping. If state is set to NO, the live video data takes precedence, meaning that
the graphics data is discarded at the corresponding coordinates. Useful if the video should cover
the graphics, without having to change window order.
Note: When the mixer is turned off, the cursor will not be visible in those places where the video
is displayed.
See also: - isMixed
setMosaicWidth:
- setMosaicWidth:(short)width
Set mosaicking scale to width. Mosaicking causes width pixels to be summed and displayed as
a single pixel. Effectively reduces the resolution of the displayed image. Only works on output
stage; has no affect on image in memory. Do not use if video is flipped using the setFlipped:
method.
See also: - mosaicWidth, - isFlipped, - setFlipped:

Video Developer Kit
setNoiseFilter:
- setNoiseFilter:(float)val
Sets value of input noise filter to val. The Range is from 0.0 to 1.0 in steps of 0.25. The default
is 0.0.
See also: - noiseFilter:
setPll:
- setPll:(unsigned short)val
Sets the phased locked loop of the SM to val. Valid range is from 0-1024. Realistic values range
from 500-800. This value controls the width and horizontal scale of the live video picture. It acts
solely on the output stage of the SM; it has no effect on the digitized images. Setting to extreme
values can cause the picture to collapse (just stripes of distorted video on screen). On a system
with a resolution above 800x600 the horizontal scale of the input has to be halved by setting
setHorizontalScaled: to YES, to achieve the correct aspect ratio of the live video on screen.
This is usually done in the Configure.app which stores it values in the Instance0.table in
./usr/Devices/SMDriver.config. See appendix for example data.
See also: - pll
setPosterization:
- setPosterization:(float)val
Sets posterization of input image according to val. The range is from 0.0 to 1.0. 0.0 disables
posterization. A larger value reduces the amount of distinct shades of color.
See also: - posterization
setPreFilter:
- setPreFilter:(BOOL)state
If state is YES, the input source is prefiltered, otherwise NO.
See also: - isPreFiltered
setRedGain:
- setRedGain:(float)val
Changes the red component in the output image according to val. The range is from 0.0 to 1.0,
with 0.5 being neutral. A smaller value means less red and a larger value more red in the picture.
Has no affect on the image in memory.
See also: - redGain, - blueGain, -greenGain, - setBlueGain:, -setGreenGain:

Video Developer Kit
setSaturation:
- setSaturation:(float)val
Sets saturation of output image according to val. The range is from 0.0 to 1.0 in steps of 0.016,
with 0.5 being neutral. A smaller value means a less saturated and a larger a more saturated pic-
ture. Has no affect on the image in memory.
See also: - saturation
setSharpness:
- setSharpness:(float)val
Sets sharpness of output image according to val. The range is from 0.0 to 1.0, with 0.5 being
neutral. A smaller value reduces the sharpness and a larger increases the sharpness of the pic-
ture. Has no affect on the image in memory.
See also: - sharpness
setSplitMemGrabSize:
- setSplitMemGrabSize:(NXSize)grabSize
Sets size of the following read images from the VDOboard to grabSize. Only works in SM_-
SPLIT_MEM mode. grabSize should not exceed the available lower memory, reserved for the
images to be read, on the board.
See also: - readImage, - readImageSelection:
setStill:
- setStill:(BOOL)state
If state is YES, the live video is halted. If the split memory mode is active, a frame is written
into the lower part of memory before halting the live picture. If state is NO, the picture returns
to the live mode. Results in a one frame offset between picture in lower and live picture (upper
memory) if split memory mode is active. Setting to YES disables the writing of video data into
memory. This means that the controls on the input side have no effect while set YES. These are
all the filters and luma and chroma intensity.
See also: - isStill

Video Developer Kit
setSystem:
- setSystem:(unsigned char)val
Sets system of input decoder to val. Possible values are SM_SYSTEM_PAL, SM_SYSTEM_-
SECAM and SM_SYSTEM_NTSC. This also affects the current videoframe, which is reset to
the maximum resolution according to the system. These values are 756x567 for PAL and SE-
CAM and 640x480 for NTSC.
See also: - system
setVCRTimebase:
- setVCRTimebase:(BOOL)state
If state is YES, timebase correction is applied. Useful if mechanical video sources like video
recorders are connected. If state is NO, timebase correction is disabled.
See also: - isVCRTimebase
setVerticalOffset:
- setVerticalOffset:(unsigned short)val
Sets vertical offset of VDOboard coordinate system to coordinate system of the graphics sub-
system to val. Used to align the two coordinate systems. Usually set by the Configure.app, the
value is stored in the Instance0.table and read by the control class at initialization and reset.
See also: - verticalOffset
setVideoFrame:
- setVideoFrame:(NXRect)theRect
Set the video frame to theRect. The video frame is the size and position at which the video image
from the input decoder gets transferred into memory. The video frame is usually set by the set-
System: method to the currently active system and its size. The offset is usually set to skip
invalid or noisy data at the beginning of a field.
See also: - getVideoFrame:

Video Developer Kit
setVideoInput:
- setVideoInput:(unsigned char)val
Sets active input source for the VDOboard to val. Possible values are non negative integers
starting with 0. The range of values depend on the VDOboard you are using. For Screen Ma-
chine and SM_INPUT_SVHS the special connector cable has to be used, with luma connected
to the black input and chroma to the red input. If an audio device is connected, it also automat-
ically activates the according audio channel.
See also: - videoInput, - numberOfInputs, - inputType:
setVideoOn:
- setVideoOn:(BOOL)state
If state is YES, a live video will be displayed on screen. If state is NO, the live video will be
disabled and removed from the screen. Whether an actual picture is displayed depends on if
there is a valid video signal at the selected input, and if the quality adjustments are set correctly.
The overlay has to be adjusted with the system function for a correct output as well. Setting vid-
eo to off does noticeable disable the writing of the video data into memory; use the setStill:
method to prevent new data from being written. It is possible to grab video without it being dis-
played on screen.
See also: - isVideoOn
setWindowFrame:
- setWindowFrame:(NXRect)theRect
Sets window frame to theRect. The window frame is the size and position of the live video im-
age on screen. The minimum size is 16x1; maximum size is the current screen size. The window
frame affects the size of the image in memory if it is smaller than the video frame.
See also: - getWindowFrame:
setZoomFrame:
- setZoomFrame:(NXRect)theRect
Sets zoom frame to theRect. The zoom frame is the portion of the video in memory of the VDO-
board that is displayed on screen in the currently active video frame. The zoom frame is speci-
fied in percent from 0.0 to 100.0, and therefore independent of the current window and video
frame.
See also: - getZoomFrame:

Video Developer Kit
sharpness
- (float)sharpness
Returns current value of sharpness.
See also: - setSharpness:
smNum
- (short)smNum
Returns number of the VDOboard used by the SMControl class instance. Returns -1 if no VDO-
board is assigned.
system
- (unsigned char)system
Returns currently set system, one of SM_SYSTEM_PAL, SM_SYSTEM_SECAM or SM_-
SYSTEM_NTSC
See also: - setSystem:
verticalOffset
- (unsigned short)verticalOffset
Returns the vertical offset between the VDOboard coordinate system and the screen coordinate
system.
See also: -setVerticalOffset:
videoInput
- (unsigned char)videoInput
Returns the currently active video input, a positive integer starting with 0. The range of values
depend on the VDOboard you are using.
See also: - setVideoInput:, - numberOfInputs, - inputType:
writeImage:
- writeImage:(NXImage *)anImage
Displays anImage in the view of the VDOboard. The image is scaled to the current view dimen-
sions. Used to preview images in 24 bit, independent of graphics card. anImage has to contain
a SMYUVImageRep or SMFLMImageRep representation.
See also: - readImage, - readImageSelction:, SMYUVImageRep, SMFLMImageRep

Video Developer Kit
MMControl
Inherits From: SMControl
Declared In: <smkit/MMControl.h>
Class Description
This class extends the SMControl class for those VDOboards that support video out. In partic-
ular this is the MovieMachine PRO, MovieMachine II, and to a limited amount the FPS60. The
MovieMachine PRO supports video out with one background channel that is of fixed size and
has to have the same video system as the input source (the input itself can be different). The
additional foreground channel is scalable, supports loading of images and with the optional
CODY board, playback of live video form harddisk . The Movie Machine II is identical to the
Movie Machine PRO, except that it has a better quality (4:2:2) and supports alpha compositing
of the foreground and background channel. The FPS60 is different in that it has no foreground
channel.
Instance Variables
Method Types
- setVideoOut:
- setVideoOutInput:
- setVideoOutInvert:
- setVideoOutPattern:
- setVideoOutScale:
- setVideoOutSystem:
- setVideoOutXofs:
- setVideoOutYofs:
- videoOutInput
- videoOutSystem
- videoOutXofs
- videoOutYofs
- writeImage:
Reference - MMControl 2-55

Video Developer Kit
Instance Methods
setVideoOut:
- setVideoOut:(BOOL)state
If state is YES then the overlay will be disabled and displayed on the video out channel instead.
If state is set to NO the foreground channel is disabled on video out and displayed in the overlay
again. The mixer settings ( - setMixerOn:) has no effect on video out.
See also: - setVideoOn: (SMControl)
setVideoOutInput:
- setVideoOutInput:(short)nInp
Selects the input for the background channel on video out. Possible values are VID-
EO_OUT_INPUT_A, VIDEO_OUT_INPUT_B, VIDEO_OUT_INPUT_TV or VID-
EO_OUT_INPUT_NONE
See also: -videoOutInput, - setVideoInput: (SMControl)
setVideoOutInvert:
- setVideoOutInvert:(BOOL)bInv
If bInv is set to YES, the vertical axis of the live picture will be flipped at the input stage. This
results in the live picture and the image in memory being upside down. If bInv is NO, the image
will remain unchanged.
See also: - setFlipped (SMControl)
setVideoOutScaled:
- setVideoOutScaled:(BOOL)state
Controls how the foreground channel is displayed when the input and outputs systems differ(e.g
input video system is NTSC and the output system is PAL) . This is useful when converting be-
tween video standards. If state is set to YES the forground channel is scaled to fit the out system
(i.e PAL is shrunk if output is NTSC and NTSC is enlarged if output is PAL)
See also: - setVideoOutSystem, - setSystem: (SMControl)
setVideoOutSystem:
- setVideoOutSystem:(short)nSys
Sets video system of output background channel. Possible values are SM_SYSTEM_PAL,
SM_SYSTEM_SECAM and SM_SYSTEM_NTSC.
See also: - videoOutSystem, - setSystem (SMControl)
2-56 Reference - MMControl

Video Developer Kit
setVideoOutXofs:
- setVideoOutXofs:(short)nXof
Moves the foreground channel on video out relative to background channel. This is used to ad-
just the foreground channel to the background channel, and is normally done in the driver con-
figuration.
See also: - videoOutXofs, - setHorizontalOffset (SMControl)
setVideoOutYofs:
- setVideoOutYofs:(short)nYof
Moves the foreground channel on video out relative to background channel. This is used to ad-
just the forground channel to the bagrouind channel, and is normally done in the driver config-
uration.
See also: - videoOutYofs, - setVerticalOffset (SMControl)
videoOutInput
- (short)videoOutInput
Returns the currently active video out input, one of VIDEO_OUT_INPUT_A, VID-
EO_OUT_INPUT_B, VIDEO_OUT_INPUT_TV or VIDEO_OUT_INPUT_NONE.
See also: - setVideoOutInput, -videoInput (SMControl)
videoOutSystem
- (short)videoOutSystem
Returns currently set video out system, one of SM_SYSTEM_PAL, SM_SYSTEM_SECAM or
SM_SYSTEM_NTSC.
See also: - setVideoOutSystem, - system (SMControl)
videoOutXofs
- (short)videoOutXofs
Returns the horizontal offset between the foreground channel on video out relative to back-
ground channel.
See also: - setVideoOutXofs, - verticalOffset (SMControl)
Reference - MMControl 2-57

Video Developer Kit
videoOutYofs
- (short)videoOutYofs
Returns the vertical offset between the foreground channel on video out relative to background
channel.
See also: - setVideoOutYofs, - horizontalOffset (SMControl)
writeImage:
- writeImage:(NXImage*)aImage
Writes aImage into the foreground channel if the MMControl class is set to video out. The max-
imum image size supported is 704x576 for PAL and 640x480 for NTSC. Larger images will be
cropped, smaller images will be centered. aImage has to contain a MMFLMImageRep repre-
sentation. If not set to video out see writeImage: (SMControl).
See also: writeImage: (SMControl), (MMFLMImageRep)
2-58 Reference - MMControl

Video Developer Kit
SMControlAudio
Category of: SMControl
Declared in: smkit/SMControl.h
Category Description
These methods control the optional audio hardware that can be connected to or be on board a
VDOboard. This is either the optional Audio-on-Bracket, a TV-Tuner or the integrated audio of
the Movie Machines. Use the hasAudio: call to determine if an audio device is connected and
functional.
Instance Methods
audioInput
- (unsigned char)audioInput
Returns active audio input, SM_AUDIO_INPUT_BLACK, SM_AUDIO_INPUT_RED or
SM_AUDIO_INPUT_YELLOW.
See also: - setAudioInput:
balance
- (float)balance
Returns current audio balance, a value between 0.0 and 1.0, with 0.5 as the default (equal bal-
ance).
See also: - setAudioBalance:
bass
- (float)bass
Returns current value of audio bass, a value between 0.0 and 1.0, with 0.5 as the default.
See also: - setAudioBass:
Reference - SMControlAudio 2-59

Video Developer Kit
fader
- (float)fader
Returns current value of audio fader, a float between 0.0 and 1.0, with 0.5 as the default (equal
output front and back).
See also: - setAudioFader:
hasAudio
- (BOOL)hasAudio
Returns YES if an audio device is connected to the VDOboard, otherwise NO.
isAudioMute
- (BOOL)isAudioMute
Returns YES if audio is muted, otherwise NO.
See also: -setAudioMute:
setAudioBalance:
- setAudioBalance:(float)val
Sets balance of audio to val. Range is from 0.0 to 1.0, with 0.5 being equal volume at left and
right, 0.0 results in the right muted, and 1.0 in the left muted.
See also: -balance:
setAudioBass:
- setAudioBass:(float)val
Sets bass of audio to val. Range is from 0.0 to 1.0, with 0.5 being neutral. 0.0 results in less bass
and 1.0 in increased bass in the output audio.
See also: -bass:
setAudioFader:
- setAudioFader:(float)val
Sets audio fader (balance between front and back outputs) to val. Range is 0.0 to 1.0 with 0.5
being neutral (equal distribution between front and back). 0.0 is all to the front, and 1.0 all to
the back output.
See also: -fader:
2-60 Reference - SMControlAudio

Video Developer Kit
setAudioInput:
- setAudioInput:(unsigned char)input
Set audio input to be active to input, SM_AUDIO_INPUT_BLACK, SM_AUDIO_INPU-
T_RED or SM_AUDIO_INPUT_YELLOW.
See also: -audioInput:
setAudioMute:
- setAudioMute:(BOOL)state
If state is YES, the audio output is muted, otherwise the audio is played at the current set vol-
ume.
See also: -audioMute:
setAudioTreble:
- setAudioTreble:(float)val
Sets treble of audio to val. Range is from 0.0 to 1.0, with 0.5 being neutral. 0.0 results in less
treble and 1.0 in increased treble in the output audio.
See also: -treble:
setAudioVolume:
- setAudioVolume:(float)val
Sets audio volume to val. Range is from 0.0 to 1.0.
See also: -volume:
treble
- (float)treble
Returns current value of audio treble, a float between 0.0 and 1.0, with 0.5 as the default.
See also: - setAudioTreble:
volume
- (float)volume
Returns current value of volume, a float between 0.0 and 1.0, with 0.5 as default.
See also: - setAudioVolume:
Reference - SMControlAudio 2-61

Video Developer Kit
SMControlChooserController
Declared In: smkit/SMControlChooserController.h
Class Description
The SMControlChooserController is a simple way to let the user choose which VDOboard to
use if multiple are present in the system. It loads the file SMMultiChoose.nib. An example nib
file is supplied with the class, but it is possible to create a custom one. The custom nib file has
to have the name SMMultiChoose.nib, and the following outlets have to be connected: idRadi-
os, idStatusMatrix, idPanel.
Further the cancel: and choose: actions methods should be called.
Instance Variables
id idRadios;
id idStatusMatrix;
id idPanel;
short choice;
idRadios A Matrix of 4 elements reflecting the choice of the
user. The cells should have tags 0 through 3. The class
determines the users choice by querying the matrix
with the call [[idRadios selectedCell] tag]. Multiple
selections should not be allowed.
idStatusMatrix A Matrix of 4 text cells with tags 0-3. The text cells
are set with the current status of the VDOboard in the
system.
idPanel The panel or window the matrices reside in. This pan-
2-62 Reference - SMControlChooserController

Video Developer Kit
el is run in a modal loop by the chooseSM:status:-

pids: method.
choice Internal variable that holds the users choice
Method Types
- cancel:
- choose:
- chooseSM:status:pids
Instance Methods
cancel:
- cancel:sender
Call when user has aborted choice. This action method is called by the Cancel button in the sup-
plied nib file.
choose:
- choose:sender
Call when user has made choice. This action method is called by the Choose button in the sup-
plied nib file.
chooseSM:status:pids:
- (short)chooseSM:(short)numSM status:(short *)statuses pids:(int *)pids
Method invoked by VDOboards newWithSelectionAndBoards:withFeatures: method when
more than one VDOboard is present in the system. It is passed statuses an array of numSM
shorts that contain the usage count of the respective VDOboard and an array of numSM ints that
contains PID of the last application that has registered itself to use the respective VDOboard.
Also passed is numSM, which is the number of VDOboards present in the system. This method
is responsible for loading the nib file and running the panel in a modal loop until the user has
made a choice (by calling the choose: method) or cancelled the process (by calling the cancel:
method).
Returns number of the chosen VDOboard or -1 if the process was cancelled.
See Also: newWithSelectionAndBoards:withFeatures: (VDOboards)
Reference - SMControlChooserController 2-63

Video Developer Kit
SMWindow
Inherits From: Window : Responder : Object
Declared In: smkit/SMWindow.h
Class Description
The SMWindow is intended to be an "easy to use" object to display a video in its miniwindow,
without having to write a single line of additional code to keep the "look and feel" of a NEXT-
STEP application.
The SMWindow always behaves like a traditional Window no matter if a SMView is registered
or not, despite its miniwindow representation. The default setting for a SMWindow is to display
the live video or a grabbed image, if the video is set still, in the miniwindow, when the SMWin-
dow is miniaturized. If more than one SMViews or subclass from it are registered, the video dis-
played is the active video of the first SMControl object the SMWindow received when
registering a video view. For this reason, any settings concerning the image of the miniwindow
are ignored if the default setting of the SMWindow is kept. If you want to remove the video dis-
play in the miniwindow, use the showVideoInMiniWindow: method. Sending a NO in this
method switches the miniwindow to a normal representation using the image set for the mini-
window.
Important information:
A SMWindow can not have a backingType NX_BUFFERED. The default setting for the
SMWindows backingType is NX_RETAINED. This implies more considerations for the devel-
oper to redraw the content when needed, but was necessary because the Interceptor in the cur-
rent release does not support buffered windows.
Instance Variables
List *viewList;
List *smcList;
int nViews;
int nSMCs;
struct _smwFlags {
unsigned char hasVideo:1;
unsigned char movingInSMWindow:1;
unsigned char movingInVideoView:1;
unsigned char showVideoInMiniWindow:1;
unsigned char reserved:4;
} smwFlags;
viewList A list object containing the registered SMViews
smcList A list object containing the SMControl objects be-
longing to the SMViews
nViews The number of registered SMViews
nSMCs The number of different SMControl objects control-
2-64 Reference - SMWindow

Video Developer Kit
ling the SMViews

smwFlags.hasVideo True if at least one SMView has been registered
smwFlags.movingInSMWindow True if dragging the SMWindow from anywhere is
enabled (default: NO)
smwFlags.movingInVideoView True if dragging the SMWindow by clicking in the
SMView is enabled (default: NO)
smwFlags.showVideoInMiniWindow True if the video should be played in the MiniWin-
dow when the SMWindow is miniaturized (default:
YES)
Method Types
Managing the video - registerVideoView:
- unregisterVideoView:
- videoViews
- hasVideo
- showVideoInMiniWindow:
- showVideoInMiniWindow
Managing the dragging - setMovingInSMWindow:
- enabledMovingInSMWindow
- setMovingInVideoView:
- enabledMovingInVideoView
Instance Methods
enabledMovingInSMWindow:
- (BOOL)enabledMovingInSMWindow
Returns YES if the SMWindow can be moved from anywhere within its content area, not only
from the title bar. Otherwise NO (default).
See also: - setMovingInSMWindow:
enabledMovingInVideoView:
- (BOOL)enabledMovingInVideoView
Returns YES if the SMWindow can be moved from within a SMView. Otherwise NO (default).
See also: - setMovingInVideoView
Reference - SMWindow 2-65

Video Developer Kit
hasVideo
- (BOOL)hasVideo
Returns YES if at least one SMView is present or NO if not.
An SMView is present if it has previously been registered by using registerVideoView:, and is
absent if there is none or it was unregistered by using unregisterVideoView:.
See also: - registerVideoView, - unregisterVideoView: - videoVideos
registerVideoView:
- registerVideoView:aVideoView
Registers aVideoView which has to be a SMView or subclass of SMView. This method does
not insert the SMView into the view hierarchy of the SMWindow; it simply notifies the
SMWindow about the presence of a (or another) SMView. After registering the SMView, the
SMWindow will manage everything concerning the video display: switching the video off
while miniaturizing, setting the correct frame for the video display and so on.
See also: - unregisterVideoView, - videoViews - hasVideo
setMovingInSMWindow:
- setMovingInSMWindow:(BOOL)enabled
Determines whether the SMWindow can be dragged from anywhere within its content area, de-
pending on enabled.
See also: - enabledMovingInSMWindow
setMovingInVideoView:
- setMovingInVideoView:(BOOL)enabled
Determines whether or not the SMWindow can be dragged from within its SMView, depending
on enabled. If dragging from within the SMWindows content area is disabled this setting has
no effect.
See also: - enabledMovingInVideoView
showVideoInMiniWindow
- (BOOL)showVideoInMiniWindow
Returns YES (default) if the video should be displayed in the miniwindow when the SMWin-
dow is miniaturized. Otherwise NO.
See also: - showVideoInMiniWindow:
2-66 Reference - SMWindow

Video Developer Kit
showVideoInMiniWindow:
- showVideoInMiniWindow:(BOOL)doIt
Determines whether or not the video should be displayed in the miniwindow when the SMWin-
dow is miniaturized depending on doIt.
See also: - showVideoInMiniWindow
unregisterVideoView:
- unregisterVideoView:aVideoView
Unregisters aVideoView which has been previously set by registerVideoView: to inform the
SMWindow that it no longer has to manage aVideoView. This method does not remove the SM-
View from the view hierarchy of the SMWindow.
See also: - registerVideoView, - videoViews - hasVideo
videoViews
- (List *)videoViews
Returns the list object containing the ids of the SMViews previously registered using regis-
terVideoView:.
See also: - registerVideoView, - unregisterVideoView: - hasVideo
Reference - SMWindow 2-67

Video Developer Kit
SMView
Inherits From: View : Responder : Object
Declared In: smkit/SMView.h
Class Description
SMView is a view class allowing scaled live video to be displayed in a window and should be
used in conjunction with a SMWindow to ensure proper handling. It is possible to use the SM-
View in a standard window, but view handling might not be as smooth, and when moved par-
tially offscreen, unexpected results can occur. A SMView has to be connected to a SMControl
class in order to be able to display and control the video. It is primarily for handling the size and
geometric appearance of the video. It also makes printing possible, as well as clipping (masking
of views obscuring live video). A SMView can be part of a scroll view. For easy use there is the
SMPaletteView which has action methods to control almost all of the SMControls methods, and
it can be used in InterfaceBuilder.
The SMView class controls those aspects of the video that have to do with its geometric appear-
ance on screen, and only those. In particular those are in terms of SMControl class methods, the
window and zoom frame, as well as the setStill: method. In order to avoid conflicts with the
SMControl class, the following methods of the SMControl class should not be used when using
a SMView class.
setWindowFrame (automatically controlled by SMView)
setZoomFrame (automatically controlled by SMView)
setStill (use view’s start: and stop: methods instead)
setVideoOn (use view’s start: and stop: methods instead)
See the SMViewDragging category for drag&drop of images in the SMview class.
Note: Printing
When the printPSCode: method is invoked it usually brings up a PrintPanel, and only when
the user continues the drawSelf:: method will be sent to the view. This means the point in time
at which the video is grabbed for printing is relatively undetermined. So it is suggested that the
view is stopped (with the stop:) method before invoking printPSCode:. One way to accom-
plish this is to subclass the printPSCode: method and stop the video there.
Note: Clipping
The following methods reset the clipping when invoked. If several of these methods are called
in succession (like when initializing the view) it is suggested that the clipping is suspended with
the suspendClipping: method. Use suspendClipping: rather the setClippingOn: since setClip-
pingOn: also resets the clips. Resetting clips involves a flush of the VDOboards memory,
which is a time consuming process.
Don't use the clipping methods provided by the SMControl class. They only affect the clipping
mask of the VDOboard not the state of the view. Effects like the view blanking, or that the live
2-68 Reference - SMView

Video Developer Kit
video does not follow the window around can happen if the clipping of the VDOboard has been
turned off without the view being aware of it.
Note: Windows
All windows that contain a SMView that are of type NX_BUFFERED will behave like a win-
dow of type NX_RETAINED. The usual precautions, and drawing optimizations for retained
windows apply.
Instance Variables
id smControl;
struct _flags {
unsigned char scrollers:1;
unsigned char fixedratio:1;
unsigned char fixedsize:1;
unsigned char clipping:1;
unsigned char animated:1;
unsigned char grabonstop:1;
unsigned char zoom:1;
unsigned char stopped:1;
unsigned char dragfrom:1;
unsigned char dragto:1;
unsigned char isReallyObscured:1;
unsigned char wasOffScreen:1;
unsigned char willResize:1;
unsigned char pad:3;
} flags;
NXRect windowFrame;
NXRect zoomFrame;
NXSize screenSize;
NXImage *idGrabbedImage;
char tmpfilename[MAXPATHLEN+1];
int windowNum;
smControl The SMControl class, whose video the view is displaying
flags.scrollers YES if view should use scrollers on size mismatch
flags.fixedratio YES if video height to width ratio should always be 0.75
flags.fixedsize YES if video should not resize with view
flags.clipping YES if view should mask out windows obscuring the view
flags.animated YES if dragging and resizing of view should resize the vid-
eo at the same time
flags.grabonstop YES if view should grab video and display the grabbed im-
age instead of the still video
flags.zoom YES if view is not completely visible and video should be
adjusted accordingly
flags.stopped YES if video is still
windowFrame Size of visible part of view displaying video
zoomFrame Size of zoom rectangle of full video size
screenSize Size of current screen
Reference - SMView 2-69

Video Developer Kit
idGrabbedImage The grabbed image
Method Types
Initializing and freeing SMView objects - initFrame:control:
- free
Modifying control - setControl:
- control
- hasControl
- controlWillFree:
Changing behaviour - setFixedRatio:
- isFixedRatio
- setScaledVideo:
- isScaledVideo;
- setVideoSize:
- getVideoSize:
- setAnimatedMovement:
- isAnimatedMovement
- setGrabOnStop
- doesGrabOnStop
- start:sender
- stop:sender
Acquiring an image from view - grab
Clipping - setClippingOn:
- isClippingOn
- suspendClipping:
- resetClips- redrawClips
Dragging (In SMViewDragging Category) - canDragFrom
- canDragTo
- setDragFrom:
- setDragTo:

Video Developer Kit
Instance Methods
control
- control
Returns id of SMControl class the view is currently connected to.
See also: - setControl:, initFrame:control:
controlWillFree:
- controlWillFree:sender
Sent by SMControl to every view in its list when it receives a free message. When the view loos-
es a control it also looses its live video representation. Can be used to give feedback to the user.
doesGrabOnStop
- (BOOL)doesGrabOnStop
Returns YES if an image is grabbed and displayed when video is stopped, otherwise NO.
getVideoSize:
- getVideoSize:(NXSize *)aSize
Returns size of video in aSize. This is the size of the video acquisition rectangle, not the size as
displayed on screen. Useful if the view is set to fixed size and the displayed video size should
be changed.
See also: - setVideoSize:, - setScaledVideo:, - isScaledVideo
grab
- (NXImage *)grab
The current frame or field is captured in the size of the view. Returns an id of a newly allocated
NXImage, containing a SMYUV- or SMFLMImageRep with the captured data. The receiving
object should free the image.
See also: - setGrabOnStop:, - doesGrabOnStop

Video Developer Kit
initFrame:control
- initFrame:(NXRect *)aRect control:aControl
Initializes SMView to size specified by aRect and sets control to aControl. Returns self.
See also: - initFrame: (View)
isClippingOn
- (BOOL)isClippingOn
Returns YES if windows obscuring the view are masked out. Otherwise returns NO. Do not
confuse with the views setClipping: method.
See also: - setClippingOn:
hasControl
- (BOOL)hasControl
Returns YES if view has current control (i.e live video), otherwise returns NO. If this method
returns NO, it does not mean that the SMView instance does not have a valid instance of a SM-
Control class. It is possible for more than one SMView to be connected to the same SMControl
(like the live video in the SMMiniWindow). This method can be used to determine which of
these views is currently displaying the live video (since only one can at a given time). This
method will also return NO if no SMControl is present.
isScaledVideo
- (BOOL)isScaledVideo
Returns YES if video is scaled to view dimensions. Otherwise returns NO. When NO, video
size is determined by setVideoSize:, and the visible portion by visible rectangle of the view.
See also: - setVideoSize:, - setScaledVideo:, - getVideoSize
redrawClips
- redrawClips
Redraws all clips. This means that all areas obscuring the live video view will be masked out,
preventing the live video from 'ghosting' ontop of the obscuring windows. This method will not
remove areas that have accidentally been masked out although there is no window obscuring
the view. Use resetClips instead if you want to get rid of wrongly clipped areas. For perform-
ance reasons it it usally prefreable to call suspendClipping: with YES prior to this call, so the
automatic clipping mechanism doesn't conflict with this manual method. Don't forget to reean-
able clipping with a suspendClipping: with NO afterwards.

Video Developer Kit
[view suspendClipping:YES];
[view redrawClips];
[view suspendClipping:NO];
See also: - setClippingOn:, - redrawClips:, - suspendClipping:
resetClips
- resetClips
Resets clipps. This is the 'expensive' version, it will clear the entire video view prior to redraw-
ing the clips. For a 'cheaper' version use redrawClips. For performance reasons it it usally pre-
freable to call suspendClipping: with YES prior to this call, so the automatic clipping
mechanism doesn't conflict with this manual method. Don't forget to reeanable clipping with a
suspendClipping: NO afterwards.
[view suspendClipping:YES];
[view resetClips];
[view suspendClipping:NO];
See also: - setClippingOn:, - redrawClips:, - suspendClipping:
setClippingOn:
- setClippingOn:(BOOL)state
Turns clipping on when state is YES. This method should be used to use clipping, not the meth-
od in the control class, since it doesn't change the view behaviour.
See also: - isClippingOn, - resetClips, - redrawClips
setControl:
- setControl:anObject
Sets the SMControl class, which the view belongs to, to anObject. Reinitializes the VDOboard
associated with the control class to the dimensions of the view.
See also: - hasControl, - initFrame:control:
setGrabOnStop
- setGrabOnStop:(BOOL)state
Controls whether a grabbed image is displayed, instead of the still video. Set state to YES for
image, NO for still video. Has to be set to YES to enable dragging images from view.
See also: - doesGrabOnStop:, - grab, - setDragFrom:

Video Developer Kit
setScaledVideo:
- setScaledVideo:(BOOL)state
Controls if video is always scaled to view dimensions. If state is set to YES, video is scaled to
view dimensions. If set to NO, video is cropped when the view is smaller than videosize and the
views maximum size is constrained to videosize. The videosize is set with the setVideoSize:,
and can be determined with the getVideoSize: method.
See also: - isScaledVideo, - setVideoSize:, - getVideoSize
setVideoSize:
- (BOOL)setVideoSize:(NXSize)aSize
Sets size of video when the setScaledVideo: is set to NO. If aSize is smaller than the current
view bounds, the view will not be completely filled with video. If aSize is larger than the current
view bounds only part of the video can be seen. Use a scrollview to make everything visible.
See also: - isScaledVideo, - setScaledVideo:, - getVideoSize
start:
- start:sender
Turns video on and live. Removes grabbed image from view if present. Use instead of SMCon-
trols setStill: and setVideoOn: methods.
See also: - stop
stop:
- stop:sender
Sets video to still. If set to setGrabOnStop:, displays grabbed image in view instead of still
video.Use instead of SMControls setStill: and setVideoOn: methods.
See also: - start:, - doesGrabOnStop, - setGrabOnStop:
suspendClipping:
- suspendClipping:(BOOL)state
If state is YES all clipping activity is suspended until this method is called with a state of NO.
Use this method to temporarily disable the clipping mechanism without the performance hit. Es-
pecially useful when calling more than one method that resets clipping. (See class description
for details).
See also: - setClippingOn:, - isClippingOn

Video Developer Kit
SMPaletteView
Inherits From: SMView : View : Responder : Object
Declared In: smkit/SMPaletteView.h
Class Description
The SMPaletteView has been designed to build simple applications using InterfaceBuilder.
Most of its methods work by connecting Controls (e.g. Buttons, Matrices … ) to it. The really
new thing is that you can set defaults for the video input and the frame type that will be taken,
when the SMPaletteView gets a start: message. If the VDOboard is connected to an additional
audio feature, default values for the volume and the mute state are also set when receiving the
start: message. If one of these values is changed through methods of the SMPaletteView, they
will be updated as default values. The reason behind this behaviour is that you can set different
default values for different SMPaletteViews in a Window, which will be switched when another
SMPaletteView is clicked with the mouse to become active.
For further descriptions see the corresponding methods of the SMControl class.
Instance Variables
unsigned char defInput;
unsigned char defFields;
float defVolume;
BOOL defAudioMute;
defInput One of four possible video input values the VDO-
board offers
defFields One of the possible three inputFrameTypes
defVolume A default volume setting for the optional audio fea-
ture
defAudioMute The default state of the optional audio feature whether
audio is on or not
Method Types
Managing defaults - setDefaultAudioMute:
- defaultAudioMute:
- setDefaultFields:
- defaultFields
- setDefaultInput:
- defaultInput
- setDefaultVolume:
- defaultVolume
Reference - SMPaletteView 2-75

Video Developer Kit
Action methods - saveImageAs:

- setAudioBalance:
- setAudioBass:
- setAudioFader:
- setAudioMute:
- setAudioTreble:
- setAudioVolume:
- setBlueGain:
- setBrightness:
- setChromaIntensity:
- setChromaInvert:
- setColorOn:
- setContrast:
- setFlipped:
- setGreenGain:
- setHue:
- setLumaIntensity:
- setLumaInvert:
- setLumaSharpness:
- setMixerOn:
- setRedGain:
- setSaturation:
- setSharpness:
- setStill:
- setVideoFields:
- setVideoInput:
- setVideoOnOff:
Instance Methods
defaultAudioMute
- (BOOL)defaultAudioMute
Returns YES (default) if the optional audio feature is mute when the SMPaletteView gets a
start: message. Otherwise this method returns NO.
See also: - setDefaultAudioMute:
defaultFields
- (unsigned char)defaultFields
Returns SM_FRAME_ODD, SM_FRAME_EVEN or SM_FRAME_BOTH (default) as default
for the frameInputType for this SMPaletteView.
See also: - setDefaultFields:
2-76 Reference - SMPaletteView

Video Developer Kit
defaultInput
- (unsigned char)defaultInput
Returns a psitive integer (starting with 0) as the default video input for this SMPaletteView. The
range of values depend on the VDOboard you are using.
See also: - setDefaultInput:, - numberOfInputs (SMControl), - inputType: (SMControl)
defaultVolume
- (float)defaultVolume
Returns the default volume for the additional audio option for this SMPaletteView. The default
value if not changed is 0.3.
See also: - setDefaultVolume
saveImageAs:
- saveImageAs:sender
This method causes the SMPaletteView to grab the actual live video and activates a save panel
to store the newly grabbed image.
setAudioBalance:
- setAudioBalance:sender
sender has to be a Control object (e.g. Scroller) offering a floatValue: method. The range of the
float value has to be between 0.0 and 1.0.
setAudioBass:
- setAudioBass:sender
setAudioFader:
- setAudioFader:sender

Video Developer Kit
setAudioMute:
- setAudioMute: sender
sender has to be a Control object (e.g. Button in toggle mode) offering a state: method. The
returned value for state has to be YES (selected) or NO (not selected).
setAudioTreble:
- setAudioTreble:sender
setAudioVolume:
- setAudioVolume:sender
setBlueGain:
- setBlueGain:sender
setBrightness:
- setBrightness:sender
setChromaIntensity:
- setChromaIntensity:sender
setChromaInvert:
- setChromaInvert:sender

Video Developer Kit
setColorOn:
- setColorOn:sender
setContrast:
- setContrast:sender
setDefaultAudioMute:
- setDefaultAudioMute:(BOOL)audioMute
This method has effect only if an optional audio feature is connected to the VDOboard.
Through this method, the mute state of the audio feature can be set. Sending NO (default) keeps
the audio feature turned off when the SMPaletteView receives a start: message, YES will turn
audio on.
See also: - defaultAudioMute
setDefaultFields:
- setDefaultFields:(int)fields
This method sets the inputFrameType for this SMPaletteView that will be set when a start:
message is sent.
Possible values are : SM_FRAME_ODD, SM_FRAME_EVEN and SM_FRAME_BOTH.
See also: - defaultFields
setDefaultInput:
- setDefaultInput:(int)input
This method sets the video input for this SMPaletteView that will be set when a start: message
is sent.
Possible values are positive integers starting with 0(default). The range of values depend on the
VDOboard you are using.
See also: - defaultInput, - numberOfInputs (SMControl), - inputType: (SMControl)

Video Developer Kit
setDefaultVolume:
- setDefaultVolume:(float)volume
This method has effect only if an optional audio feature is connected to the VDOboard.
Setting a default volume between 0.0 and 1.0 will set the volume of the audio feature to volume
when a start: message is sent.
See also: - (float)defaultVolume
setFlipped:
- setFlipped:sender
setGreenGain:
- setGreenGain:sender
setHue:
- setHue:sender
setLumaIntensity:
- setLumaIntensity:sender
setLumaInvert:
- setLumaInvert: sender

Video Developer Kit
setLumaSharpness:
- setLumaSharpness: sender
setMixerOn:
- setMixerOn: sender
setRedGain:
- setRedGain: sender
setSaturation:
- setSaturation: sender
setSharpness:
- setSharpness: sender
setStill:
- setStill: sender

Video Developer Kit
setVideoFields:
- setVideoFields: sender
sender has to be a matrix of Controls (e.g. Matrix of radiobuttons) where the tags have to be set
to represent possible inputFrameTypes of the VDOboard. (tag value 1 = SM_FRAME_ODD,
2 = SM_FRAME_EVEN, 3 = SM_FRAME_BOTH). sender has to be a matrix because the
mechanism used to set the value is :
[[sender selectedCell] tag].
setVideoInput:
- setVideoInput: sender
sender has to be a matrix of Controls (e.g. Matrix of radiobuttons) where the tags have to be set
to represent possible video inputs of the VDOboard. The range of values depend on the VDO-
board you are using. (e.g. Screen Machine : tag value 0 = SM_INPUT_BLACK, 1 = SM_IN-
PUT_RED, 2 = SM_INPUT_YELLOW, 3 =SM_INPUT_SVHS). sender has to be a matrix
because the mechanism used to set the value is :
[[sender selectedCell] tag].
setVideoOnOff:
- setVideoOnOff: sender

Video Developer Kit
SMViewDragging
Category of: SMView
Declared in: smkit/SMView.h
Category Description
SMViewDragging is an addition to the SMView implementing the dragging protocol. It allows
dragging from images out of the view into other applications. Any stopped, grabbed image can
be dragged (enabled with setGrabOnStop:). Dragging of live video or stopped video only is
not supported.
Images can also be dragged from the workspace into the view, provided they are type FLM,
TIFF or EPS. This provides a convenient way to preview images in 24bit regardless of the win-
doserver color depth.
Instance Methods
canDragFrom
- (BOOL)canDragFrom
Returns YES if dragging a stopped grabbed image is enabled. Only works if setGrabOnStop:
is enabled as well. When the view is stopped, the user can drag an icon out of the view repre-
senting the stopped video into any application accepting NXFilenamePboardType or NXTIFF-
PboardType
See also: - setDragFrom:
canDragTo
- (BOOL)canDragTo
Returns YES if dragging an image into the view is enabled. If the dragged pasteboard contains
a filename ending in .FLM, .TIFF or .EPS the image will be scaled and displayed in the view.
See also: - setDragTo:
setDragFrom
- setDragFrom:(BOOL)state
If set to YES, dragging a stopped image form view is enabled otherwise this feature is disabled.
See also: - canDragFrom
setDragTo:
- setDragTo:(BOOL)state
If set to YES, dragging an image from Workspace into view is enabled otherwise this feature is
disabled
Reference - SMViewDragging 2-83

Video Developer Kit
SMYUVImageRep
Inherits From: NXImageRep:Object
Declared In: smkit/SMYUVImageRep.h
Class Description
The SMYUVImageRep simplifies the use of the YUV data obtained from a VDOboard. The
YUV format (Y is luminance, U and V are chrominance) is commonly used in video processing
and broadcasting. It makes it possible to support both black and white and color video at the
same time as the color and monochrome information are stored separately. Computer images,
and those in the other image representations, use the RGB color space (like TIFF uses). The
SMYUVImageRep bridges the gap between the YUV and RGB color space by converting ei-
ther way. Be aware that in SMYUVImageRep the settings of all gains, brightness, contrast, sat-
uration and hue values are not taken into consideration when converting to and from YUV.
The class employs a lazy conversion strategy - data is only converted into the other format when
requested (i.e when the instance is initialized with YUV data with the initData:pixelsWide:-
pixelsHigh:YUVMode: method, the data is not converted into RGB until requested by a meth-
od like writeTIFF:), or to SM_YUVMODE6 if necessary during initialization. Internaly the
data is hold in mode SM_YUVMODE6, therefore this mode is the best mode to read and write
data. This allows faster data handling during a process like sequence grabbing.
Instance Variables
int imgMode;
imgMode YUV mode of the image data.
Instance Constants
Constant Name Pixel Ordering
SM_YUVMODE0 YYVUYY
SM_YUVMODE1 Y
SM_YUVMODE2 YYVU
SM_YUVMODE3 YVU
SM_YUVMODE4 YYVUYY
SM_YUVMODE6 YUYV
SM_RGBMODE RGB (24 bit)
2-84 Reference - SMYUVImageRep

Video Developer Kit
Method Types
Initializing a new SMYUVImageRep object- init
- initData: pixelsWide: pixelsHigh:YUVMode:
- initDataFromStream: pixelsWide:pixelsHigh:
YUVMode:
- initFromPasteboard:
- initTIFFDataFromStream:
Freeing an SMYUVImageRep - free
Checking data types + canLoadFromStream:
Setting the size of the image - setSize:
- imageSizeFromMode:
Representation attributes - bitsPerSample
- hasAlpha
- isOpaque
- mode
- numColors
- setAlpha:
- setBitsPerSample:
- setNumColors:
- setOpaque:
- setPixelsHigh:
- setPixelsWide:
- updateDisplayData
Getting image data - getData: withMode:
- imageDataInMode6
Writing a TIFF representation of the image - writeTIFF:
- writeTIFF: usingCompression:
- writeTIFF: usingCompression: andFactor:
Setting/checking compression types + getTIFFCompressionTypes:count:
- getCompression:andFactor:
- setCompression:andFactor:
Archiving - read:
- write:
Reference - SMYUVImageRep 2-85

Video Developer Kit
Class Methods
canLoadFromStream:
+ (BOOL)canLoadFromStream:(NXStream *)stream
Tests whether the receiving class can initialize an instance of itself from stream. Currently this
method always returns YES.
See also: - initDataFromStream:, - initTIFFDataFromStream:
getTIFFCompressionTypes:count:
+ (void)getTIFFCompressionTypes:(const int **)list count:(int *)numTypes
Returns, by reference, an integer array representing all available compression types that can be
used when writing a TIFF image. The number of elements in list is represented by numTypes.
list and numTypes, belonging to the SMYUVImageRep class; it shouldn't be freed or altered.
The following compression types are supported:
NX_TIFF_COMPRESSION_NONE 1
NX_TIFF_COMPRESSION_LZW 5
NX_TIFF_COMPRESSION_JPEG 6
Instance Methods
bitsPerSample
- (int)bitsPerSample
Currently always returns 8.
See also: - setBitsPerSample:
free
- free
Deallocates the SMYUVImageRep. This method will not free any image data that the object
merely references, that is, raw data that was passed to it in a initData:... , initDataFrom-
Stream:... or initTIFFDataFromStream:... message. Returns nil.

Video Developer Kit
getCompression:andFactor:
- (void)getCompression:(int *)compression andFactor:(float *)factor
Returns, by reference, the receiver's compression type and compression factor. Use this method
to get information on the compression type for the source image data. compression represents
the compression type used on the data and corresponds to one of the values returned by the class
method getTIFFCompressionTypes:count:. factor is usually a value between 0.0 and 255.0,
with 0.0 representing no compression.
See also: + getTIFFCompressionTypes:..., - setCompression:andFactor:
getData: withMode:
- getData:(unsigned char*)data withMode:(int)aMode
Copies the image data in the mode aMode to the array data. aMode can be one of the modes
declared in mode. The size of data depends on the given mode aMode and should be the size
given by imageSizeFromMode:. Returns self.
See also: - mode, - imageSizeFromMode
hasAlpha
- (BOOL)hasAlpha
Currently always returns NO.
See also: - setAlpha:
imageDataInMode6
- (unsigned char*)imageDataInMode6
This method gives an direct access to the image data stored in the SMYUVImageRep. When
modifying the data after displaying the image on screen you should send a updateDisplayData
method to force the SMYUVImageRep to rerender the display data.
See also: - getData:, -updateDisplayData
imageSizeFromMode:
- (int)imageSizeFromMode:(int)aMode
Returns the number of bytes that would be required to get data for the current image in mode
aMode. aMode can be one of the modes declared in mode.
See also: - mode, - getData:

Video Developer Kit
init
- init
Initializes the receiver, a newly allocated SMYUVImageRep object. If the object can't be ini-
tialized, this method frees it and returns nil. Otherwise, it returns the object self.
See also: - initTIFFDataFromStream:..., - initDataFromStream:..., -initData:...
initData: pixelsWide: pixelsHigh:YUVMode:

- initData:(unsigned char*)data pixelsWide:(int)width pixelsHigh:(int)height
YUVMode:(int)aMode
Initializes the receiver, a newly allocated SMYUVImageRep object, so that it can render the im-
age as specified in data and the other arguments. If the object can't be initialized, this method
frees it and returns nil. Otherwise it returns the object self.
data points to a buffer containing raw bitmap data. The data component values should be inter-
woven in a single channel (“meshed configuration”). If data is NULL, the SMYUVImageRep
will allocate enough memory to hold YUV data (SM_YUVMODE6) for the image.
All the other arguments to this method are the same as those to initDataFromStream:... See
that method for further descriptions.
initDataFromStream: pixelsWide: pixelsHigh: YUVMode:

- initDataFromStream:(NXStream*)theStream pixelsWide:(int)width
pixelsHigh:(int)height YUVMode:(int)aMode
Initializes the receiver, a newly allocated SMYUVImageRep object, so that it can render the im-
age as specified in theStream and the other arguments. If the object can't be initialized, this
method frees it and returns nil. Otherwise it returns the object self.
The data component values contained in theStream should be interwoven in a single channel
(“meshed configuration”). theStream must be seekable.
Each of the other arguments informs the SMYUVImageRep object about the image. They're ex-
plained below:
• width and height specify the size of the image in pixels.
The size in each direction must be greater than 0.
• aMode indicates how data values are to be interpreted.
It should be one of the following enumerated values:
SM_YUVMODE0 YYVUYY mode
SM_YUVMODE1 Y mode
SM_YUVMODE2 YYVU mode
SM_YUVMODE3 YVU mode
SM_YUVMODE4 YYVUYY (internal mode of Movie Machine)
SM_YUVMODE6 YUYV mode (default mode)
SM_RGBMODE 24 bit RGB mode without a alpha plane
See also: - canLoadFromStream:, - initTIFFDataFromStream:

Video Developer Kit
initFromPasteboard:
- initFromPasteboard:(Pasteboard *)pasteboard
Does nothing and returns nil.
initTIFFDataFromStream:
- initTIFFDataFromStream:(NXStream*)theStream
Initializes the receiver, a newly allocated SMYUVImageRep object, with the TIFF image read
from theStream. If the new object can't be initialized for any reason (for example, theStream
doesn't contain TIFF data), this method frees it and returns nil. Otherwise it returns self.
See also: - initDataFromStream:..., - canLoadFromStream:
isOpaque
- (BOOL)isOpaque
Currently always returns NO.
See also: - hasAlpha, - setOpaque:
mode
- (int)mode
Returns the mode of the currently initialized image. The mode can be one of the following enu-
merated values:
SM_YUVMODE1 Y mode
SM_YUVMODE4 YYVUYY
SM_YUVMODE6 YUYV mode
SM_RGBMODE 24 bit RGB mode
numColors
- (int)numColors
Returns the number of color components in the image. For example, the return value will be 1
for images specified by mode SM_YUVMODE1 or any other one components. It will be 3 for
images specified by SM_RGBMODE, or SM_YUVMODE3 mode.
See also: - setNumColors:, - mode

Video Developer Kit
read:
- read:(NXTypedStream *)stream
Reads the SMYUVImageRep from the typed stream stream. Returns self.
See also: - write:
setAlpha:
- setAlpha:(BOOL)flag
Does nothing and returns self.
See also: - hasAlpha:
setBitsPerSample:
- setBitsPerSample:(int)anInt
See also: - bitsPerSample:
setCompression:andFactor:
- setCompression:(int)compression andFactor:(float)factor
Sets the receiver's compression type and compression factor. compression is one of the support-
ed compression types. factor is a compression factor, usually between 0.0 (no compression) and
255.0 (maximum compression).
When an SMYUVImageRep is created, the instance stores the compression type and factor for
the source data. When the data is subsequently saved, writeTIFF: tries to use the stored com-
pression type and factor. Use this method to change the compression type and factor. Returns
self.
See also: + getTiffCompressionTypes:count:, - getCompression:andFactor:
setNumColors:
- setNumColors:(int)anInt
See also: - numColors:, - mode

Video Developer Kit
setOpaque:
- setOpaque:(BOOL)flag
See also: - isOpaque:
setPixelsHigh:
- setPixelsHigh:(int)anInt
See also: - pixelsHigh (NXImageRep),
setPixelsWide:
- setPixelsWide:(int)anInt
See also: - pixelsWide (NXImageRep)
setSize:
- setSize:(const NXSize *)aSize
See also: - getSize (NXImageRep)
updateDisplayData:
- updateDisplayData
Forces the SMYUVImageRep to rerender its display image data. This is only necessary if you
change the YUV data using the imageDataInMode6 method. Returns self.
See also: - imageDataInMode6
write:
- write:(NXTypedStream *)stream
Writes the SMYUVImageRep to the typed stream stream. Returns self.
See also: - read:

Video Developer Kit
writeTIFF:
- writeTIFF:(NXStream *)stream
Writes a TIFF representation of the image to stream. This method invokes writeTIFF:using-
Compression:andFactor: using the stored compression type and factor retrieved from the in-
itial image data, or is changed using setCompression:andFactor:. If the stored compression
type isn't supported for writing TIFF data
(e.g., NX_TIFF_COMPRESSION_NEXT), the stored compression is changed to NX_TIFF_-
COMPRESSION_NONE and the compression factor to 0.0 before invoking writeTIFF:using-
Compression:andFactor:.
See also: - getCompression:andFactor:, - setCompression:andFactor:
writeTIFF: usingCompression:
- writeTIFF:(NXStream *)stream usingCompression:(int)compression
Writes a TIFF representation of the image to stream, compressing the data according to the
compression scheme. If compression is NX_TIFF_COMPRESSION_JPEG, the default com-
pression factor is used. This and the other compression constants are listed under the following
method.
See also: - writeTIFF:usingCompression:andFactor:
writeTIFF: usingCompression: andFactor:

- writeTIFF:(NXStream *)stream usingCompression:(int)compression
andFactor:(float)factor
Writes a TIFF representation of the image to stream. If the stream isn't currently positioned at
location 0, this method assumes that it contains another TIFF image. It will try to append the
TIFF representation it writes to that image. To do this, it must read the header of the image al-
ready in the stream. Therefore, the stream must be opened with NX_READWRITE permission.
The second argument, compression, indicates the compression scheme to use. It should be one
of the following constants:
NX_TIFF_COMPRESSION_NONE No compression
NX_TIFF_COMPRESSION_LZW LZW compression
NX_TIFF_COMPRESSION_JPEG JPEG compression
The third argument, factor, is used in the JPEG scheme to determine the degree of compression.
If factor is 0.0, the default compression factor of 10.0 will be used. Otherwise, factor should fall
within the range 1.0 - 255.0, with higher values yielding greater compression but also greater
information losses.
See also: - writeTIFF:usingCompression:

Video Developer Kit
SMFLMImageRep
Inherits From: SMYUVImageRep:NXImageRep:Object
Declared In: smkit/SMFLMImageRep.h
Class Description
The SMFLMImageRep is a class to provide compatibility with the FLM data format, which is
used by the DOS and Windows version of the VDOboard's software. It writes and reads the
YUV data, including a special header. FLM data files have a .flm extension. The FLM format
is the ///FASTest way of storing the YUV data from the VDOboard and no conversion between
color spaces has to be made.
Different from the SMYUVImageRep the SMFLMImageRep can store the values for bright-
ness, contrast, saturation, red, green and blue gain. These values take effect when converting
data from the YUV color space to RGB or reloading the image back into the VDOboard. Note
that (as described in the SMControl class) higher values for SMImageHead.nRed, SMImage-
Head.nGreen, SMImageHead.nBlue greater than 32 (SMControl : redGain, blueGain, green-
Gain greater than 1.0) may result in strange image effects when converting to RGB. This is due
to the higher output dynamics of the VDOboard.
When initializing the rep with a initData:, initDataStream: or a initFromPasteboard: and the
pasteboard contains FLM data (SMFLMPboardType) the stReadHeader struct is set and you
can get information about the image with getHeader:. You should not change the stReadHeader
struct when subclassing the SMFLMImageRep, but you can change the header for write opera-
tions with setWriteHeader:. This Header is also used when rendering the data on screen.
In the initialization of a SMFLMImageRep the values of the header and the optional description
text of a FLM-image are copied to the corresponding “write”-variables. You need not to do so
if you don't want to change anything.
The text handled with text:, writeText: and setWriteText: is optional and gives you the pos-
sibility to store with the raw image data a few more information about the image, the date of
creation, the creator and so on.
To get a fast preview of FLM images, you can add an icon of the image to the image data. To
do so you have to setIconEnabled: to YES before using the writeFLM:withMode: method.
The icon is optional and not necessary to display the image through a VDOboard. However,
with getIcon: you can get an idea of the contents of a FLM file before writing it into a VDO-
board or rendering it on screen. The icon is written in the YUV mode SM_YUVMODE0 to use
as less space as possible.
To ensure fast working, converting, and rendering of the FLM data you should only use the
SM_YUVMODE6 mode, because, like the SMYUVImageRep, the SMFLMImageRep class
employs a kind of lazy conversion strategy. That means that the data is internally held in the
SM_YUVMODE6 format and is only converted to other formats when requested. Converting
to other formats can reduce data length, but in most cases with a lack of quality and in every
case a lack of speed.
Reference - SMFLMImageRep 2-93

Video Developer Kit
Instance Variables
struct_SMImageHead {
char sFLMId[5];
char cTextEnd;
long lIconOffs;
short nImageWdt;
short nImageHgt;
char nYUVMode;
short nYUVLeng;
char cYBits;
char cUBits;
char cVBits;
char cCompMod;
char cOldComp;
short nScrMod;
short nIconLenght;
long lTextOffs;
short nTextLen;
short nContr;
short nBright;
short nSatur;
short nHue;
short nRed;
short nGreen;
short nBlue;
char sRes[18];
} SMImageHead;
NXAtom SMFLMPboardType;
SMImageHead stWriteHeader;
SMImageHead stReadHeader;
char *sReadImageText;
char *sWriteImageText;
BOOL bIconEnabled;
NXStream *iconStream;
NXSize iconSize;
SMImageHead.sFLMId[5] Id of FLM files
SMImageHead.cTextEnd Reserved
SMImageHead.lIconOffs Offset of icon data
SMImageHead.nImageWdt Width of the image in pixels
SMImageHead.nImageHgt Height of the image in pixels
SMImageHead.nYUVMode YUV mode of the image data
SMImageHead.nYUVLeng Width of the image in bytes
SMImageHead.cYBits No. of bits of each Y component.
(Currently always 8)
SMImageHead.cUBits No. of bits of each U component.
SMImageHead.cVBits No. of bits of each V component.
2-94 Reference - SMFLMImageRep

Video Developer Kit
SMImageHead.cCompMod Compression mode of the image data.

SMImageHead.cOldComp Reserved
SMImageHead.nScrMod Reserved
SMImageHead.nIconLenght Length of icon data (if 0 no icon available)
SMImageHead.lTextOffs Offset of text data
SMImageHead.nTextLen Length of text data (0 if no text available).
SMImageHead.nContr Contrast value of image when grabbed.
SMImageHead.nBright Brightness value of image when grabbed.
SMImageHead.nSatur Saturation value of image when grabbed.
SMImageHead.nHue Hue value of image when grabbed.
SMImageHead.nRed Red gain of image when grabbed.
SMImageHead.nGreen Green gain of image when grabbed.
SMImageHead.nBlue Blue gain of image when grabbed.
SMImageHead.sRes[18] Reserved
SMFLMPboardType FLM Pasteboard
stWriteHeader FLM-Header for write operations.
stReadHeader Read FLM-Header when initializing.
sReadImageText Read image text when initializing.
sWriteImageText Image text for write operations.
bIconEnabled Write FLM with icon.
iconStream Icon stream
iconSize Size of the icon.
Method Types
Initializing a new SMFLMImageRep object - init
- initData:
- initFromStream:
- initFromPasteboard:
Freeing an SMFLMImageRep - free
Checking data types + canLoadFromStream:
+ imageUnfilteredFileTypes
+ imageUnfilteredPasteboardTypes
Getting information about the image - getHeader:
- getIcon:
- text
- hasIcon
Modifying the write attributes - getWriteHeader:
- writeText
- iconEnabled
- setIconEnabled:
- setWriteHeader:
- setWriteText:
Producing a FLM representation of an image- writeFLM: withMode:

Video Developer Kit
Archiving - read:
- write:
Class Methods
canLoadFromStream:
+ (BOOL)canLoadFromStream:(NXStream *)stream
Tests whether the SMFLMImageRep class can initialize an instance of itself from stream.
See also: - initFromStream:
imageUnfilteredFileTypes
+ (const char *const *)imageUnfilteredFileTypes
Returns a null-terminated array of strings representing all file types (extensions) supported by
the SMFLMImageRep. Supported types are: “tiff” and “flm”. Invoked by NXImage's im-
ageRepForFileType: method to find the NXImageRep subclass capable of handling files with
a particular extension.
imageUnfilteredPasteboardTypes
+ (const NXAtom *)imageUnfilteredPasteboardTypes
Returns a null-terminated array representing all pasteboard types supported by the SMFLMIm-
ageRep. Supported types are :
NXTIFFPboardType
SMFLMPboardType
NXFilenamePboardType
Invoked by NXImage's imageRepForPasteboardType: method to find the NXImageRep sub-
class capable of handling pasteboards containing FLM.
Instance Methods
free
- free
Deallocates the SMFLMImageRep. This method will not free any image data that the object
merely references, such as raw data that was passed to it in a initData:... , initFromStream:...
, initTIFFDataFromStream:..., initDataFromStream:... or initTIFFDataFromStream:...
message. Returns nil.

Video Developer Kit
getHeader:
- getHeader:(SMImageHead *)aHeader
Copies the header of the image to the structure referred to by aHeader, and returns self. The
returned header is the header of the image initialized in a initData:..., initFromStream:... or
initFromPasteboard:... message. For all the other init-methods the returned header contains
default values.
See also: - getWriteHeader:, - setWriteHeader:
getIcon:
- getIcon:(NXImage *)anImage
Copies the icon in the FLM raw data into anImage and returns self. anImage should be allocated
and initialized. If no icon is available (e.g. the FLM data contains no icon or the SMFLMIm-
ageRep is not initialized in a initFromPasteboard:..., initData:... or initFromStream:... mes-
sage) this method returns nil.
See also: - hasIcon
getWriteHeader:
- getWriteHeader:(SMImageHead)aHeader
Copies the header of the image to the structure referred to by aHeader, and returns self. The
returned header is the header that will be written in the writeFLM:... method. If the SM-
FLMImageRep is initialized in a initFromPasteboard:..., initFromStream:... or initData:...
message, the write header is set equal to the read header.
See also: - setWriteHeader:, - getHeader:
hasIcon
- (BOOL)hasIcon
Returns YES if there is an icon in the FLM raw data available. You can extract the icon with
the getIcon: method.
See also: - getIcon:
iconEnabled
- (BOOL)iconEnabled
Returns whether the SMFLMImageRep adds an icon at the end of the FLM data using the write-
FLM:... method. If the SMFLMImageRep is initialized in an initFromPasteboard:..., init-
FromStream:... or initData:... message and contains an icon, iconEnabled is set to YES.
See also - setIconEnabled:

Video Developer Kit
init
- init
Initializes the receiver, a newly allocated SMFLMImageRep object. If the object can't be ini-
tialized, this method frees it and returns nil. Otherwise, it returns the object self.
See also: - initData:, - initFromStream:, - initFromPasteboard:
initData:
- initData:(unsigned char *)data
Initializes the receiver, a newly allocated SMFLMImageRep object, so that it can render the im-
age specified in data. If the object can't be initialized, this method frees it and returns nil. Oth-
erwise, it returns the object self.
See also: - init:, - initFromStream:, - initFromPasteboard:
initFromStream:
- initFromStream:(NXStream *)theStream
age specified in theStream. If the object can't be initialized, this method frees it and returns nil.
Otherwise, it returns the object self.
See also: - init:, - initData:, - initFromPasteboard:
initFromPasteboard:
- initFromPasteboard:(Pasteboard *)thePasteboard
age with data from the given pasteboard thePasteboard. If the object can't be initialized, this
method frees it and returns nil. Otherwise, it returns the object self.
See also: - init:, - initData:, - initFromStream:
read:
- read:(NXTypedStream *)stream
Reads the SMFLMImageRep from the typed stream stream. Returns self.
See also: - write:

Video Developer Kit
setIconEnabled:
- setIconEnabled:(BOOL)flag
Informs the SMFLMImageRep if an icon should be added to the FLM data when using the
writeFLM:... method. If flag is YES an icon is added.
See also: - iconEnabled:
setWriteHeader:
- setWriteHeader:(SMImageHead)aHeader
Informs the SMFLMImageRep that the header aHeader should be used when writing the FLM
data using the writeFLM: method. Negative values in the struct SMImageHead are ignored.
Warning : The setWriteHeader: is overridden by all init methods.
The only values of aHeader used yet are : nContr, nBright, nSatur, nHue, nRed, nGreen and
nBlue.
See also: - getWriteHeader:, - getHeader:
setWriteText:
- setWriteText:(char *)aText
Informs the SMFLMImageRep that the string aText should be added to the FLM data when us-
ing the writeFLM:... method.
Warning : The setWriteText: is overridden by all init methods.
See also: - writeText, - text
text
- (char *)text
Returns the text in the FLM raw data. If no text is available (e.g. the FLM data contains no text
or the SMFLMImageRep is not initialized in a initFromPasteboard:..., initFromStream:... or
initData:... message) this method returns NULL.
See also: - writeText, - setWriteText:
write:
- write:(NXTypedStream *)stream
Writes the SMFLMImageRep to the typed stream stream. Returns self.
See also: - read:

Video Developer Kit
writeFLM: withMode:
- writeFLM:(NXStream *)theStream withMode:(int)aMode
Writes a FLM representation of the image to stream theStream. aMode is the YUV mode of the
FLM representation and should be one of the following enumerated values (also described in
SMYUVImageRep) :
SM_YUVMODE1 Y mode
SM_YUVMODE4 YYVUYY mode (internal mode of Movie Machine)
SM_YUVMODE6 YUYV mode (internal mode of Screen Machine)
The header which is used can be modified with the setWriteHeader:... method.
See also: - setWriteHeader:, - setWriteText:, - setIconEnabled:
writeText
- (char *)writeText
Returns the text that will be written in the writeFLM:... method. If no text is available this
method returns NULL. If the SMFLMImageRep is initialized in an initFromPasteboard:...,
initDataStream:... or initData:... message, the write text is set equal to the read text.
See also: - setWriteText:, - text

Video Developer Kit
MMFLMImageRep
Inherits From: SMFLMImageRep
Declared In: smkit/MMFLMImageRep.h
Class Description
The MMFLMImageRep is a class to provide the keying feature of Movie Machine Pro/II. The
Movie Machine Pro/II has the ability to display images on video out with a transparent color in
it. Use the -setKeyColorFromPoint: method to pick a key color out of an image and activate
keying by using -setKeyingOn: .
- displayImageOnVideoOut:(NXStream*)flmImageStream
{
SMChromaSpace aSpace={0.0,0.0,0.0,0.0,0.0,0.0};
MMFLMImageRep *myRep;
NXImage *myImage;
NXSize imageSize;
NXPoint keyPoint;
// Initialize a new MMFLMImageRep form flmImageStream

myRep=[[MMFLMImageRep alloc] initFromStream:flmImageStream];
if(!myRep)
{
return self;
}
// Get a valid point for keying

[myRep getSize:&imageSize];
keyPoint.x=imageSize.width/2.0;
keyPoint.y=imageSize.height/2.0;
// Pick the color of that point

[myRep setKeyColorFromPoint:&keyPoint]
// Enable keying
[myRep setKeyingOn:YES];
// Make sure that Movie Machine is set to still
if(![mmControl isStill])
{
[mmControl setStill:YES];
}
// Set keying on Movie Machine on

[mmControl setChromaKey:aSpace mode:1];
Reference - MMFLMImageRep 2-101

Video Developer Kit
// Put the rep in an image wrapper

myImage=[[NXImage alloc] initSize:&imageSize];
[myImage useRepresentation:myRep];
// display the image

[mmControl writeImage:myImage];
// Frees also rep inside
[myImage free];
return self;
}
You needn't to use this class just for handling images for Movie Machine Pro/II. Use this class
only if you want to get control of the keying feature. In all other cases the SMFLMImageRep
class is all you need.
Note : Be aware that not every YUV color has a correlating RGB color and the other way round.
To be sure to key the right color use the -setYUVKeyColor: or the -setKeyColorFromPoint:
method.
Instance Variables
SMColor smKeyColor;
NXColor nxKeyColor;
BOOL useKeyColor;
BOOL mm_data_should_update;
typedef struct _SMColor {
unsigned char Y;
char U;
char V;
} SMColor;
smKeyColor Currently set YUV-color for keying in Movie Ma-
chine.
nxKeyColor Currently set RGB-color for keying on screen.
useKeyColor MMFLMImageRep should use key color.
Method Types
- rbgKeyColor
- setKeyColorFromPoint:
- setKeyingOn:
- setRGBKeyColor:
- setYUVKeyColor:
- usesKeying
- yuvKeyColor
2-102 Reference - MMFLMImageRep

Video Developer Kit
Function Types
SMConvertRGBToYUV()
SMConvertYUVToRGB()
Instance Methods
rbgKeyColor
- (NXColor)rbgKeyColor
Returns the currently set RGB key color. If no color was set Black is returned. Use usesKeying
to evaluate whether the MMFLMImageRep uses this color for keying.
See also: -setYUVKeyColor:, -setRGBKeyColor:, -usesKeying
setKeyColorFromPoint:
- setKeyColorFromPoint:(NXPoint*)aPoint
Sets the YUV and the RGB key color to the color aPoint has. aPoint has to be in the image re-
gion. If aPoint is not in the image region keying is set off and nil is returned. Otherwise returns
self. Use this method as your preferred method to set key colors.
See also: -setYUVKeyColor:, -setRGBKeyColor:, -rgbKeyColor, -yuvKeyColor
setKeyingOn:
- setKeyingOn:(BOOL)aBool
If aBool is set to YES color keying is activated. Returns self.
See also: -usesKeying
setRGBKeyColor:
- setRGBKeyColor:(NXColor*)aColor
Sets the RGB key color. aColor is used to set the correct alpha channel in screen representation.
Furthermore the according values for the YUV key color are calculated. Due to fact of conver-
sion loss problems it may happen that a correct RGB color could not be found in the YUV color
space. For that reason it is useful to set the YUV key color separately. Returns self.
See also: -setYUVKeyColor:, -rgbKeyColor
Reference - MMFLMImageRep 2-103

Video Developer Kit
setYUVKeyColor:
- setYUVKeyColor:(SMColor*)aColor
Sets the YUV key color. aColor is used to set a transparent color when displaying an image on
video out. The values for the according RGB key color aren't set. Returns self.
See also: setRGBKeyColor:, -setKeyColorFromPoint:, -yuvKeyColor
usesKeying
- (BOOL)usesKeying
Returns YES if the MMFLMImageRep uses key colors otherwise NO is returned.
yuvKeyColor
- (SMColor)yuvKeyColor
Returns the currently set YUV key color. If no color was set Black is returned. Use usesKeying
to evaluate whether the MMFLMImageRep uses this color for keying.
See also: -setYUVKeyColor:, -setRGBKeyColor:, -usesKeying
Function Calls
SMConvertRGBToYUV
SMColor SMConvertRGBToYUV(NXColor rgb)
Converts the NXColor rgb to SMColor and returns it.
See also: SMConvertYUVToRGB
SMConvertYUVToRGB
NXColor SMConvertYUVToRGB(SMColor yuv)
Converts the SMColor yuv to NXColor and returns it.
See also: SMConvertRGBToYUV
2-104 Reference - MMFLMImageRep

Video Developer Kit
SMTVControl
Declared In: smkit/SMTVControl.h
Class Description
SMTVControl is an object that controls the tuner and the stereo device in the optional TV-Tuner
of ScreenMachineII, or the build in tuner of the other VDOboards. With this class you can tune
channels by setting frequencies and norms, set and request status of the stereo chip, and search
for TV channels. Furthermore, you can store TV channels into memory, save them to and re-
store them from disc.
The initFromControl: method is the designated initializer for the SMTVControl class.
Note: Never use the init method - an initialized SMControl class is needed for most of the meth-
ods included in this object :
- setupNewTVControlForSMControl:sender
{
unsigned short prog_nr;
// check if an SMControl class was received

if(![sender isKindOfClassNamed:"SMControl"])
{
return nil;
}
// initialize a new instance of SMTVControl
tvControl=[[SMTVControl alloc] initFromControl:sender];
if(tvControl)
{
// read file .stations in the users home directory
if([tvControl readProgs:""])
{
// we found a table
// now search the first valid channel
prog_nr=[tvControl nextProgFor:0 searchUp:YES];
// set this channel
[tvControl setProg:prog_nr];
}
return self;
}
return nil;
}
Reference - SMTVControl 2-105

Video Developer Kit
Searching channels
The best way to search for TV channels is to use the afcSearch:withSystem:andNorm: meth-
od. To work, this method needs an AFC-table given in the setAFCTable: methods. If the al-
ready existing tables do not contain enough information, you can add some values or create your
own tables. For further information on AFC-tables see the class description of AFCTables and
AFCTableList. If, for some reason, you can't work with the given tables, you can use the
search:withSystem:andNorm: method. Be aware that this method takes much more time and
it may not give proper results if not configured correctly for your country. You can configure
the search with the setSearch:withWidth: andWeight: method. A third and very difficult way
to search channels is to write your own search algorithm. See the statusVT method for infor-
mation.
Storing channels
There are three ways to use the program table within this class. First, you can use the
setAFCTable: andPresetWithNorm:andSystem: method which presets all channels with the
information given by the AFCTable. The second way is to set every channel manually with the
storeProg:withFrequency:system:andNorm: method. Finally, you can read a table stored
somewhere on your disk by using readProgs:. You should preset the values from an AFCTable
- the user then has the chance to change these settings at any time and save them on the harddisk.
For an example see the MMTV or the VDOtv application.
Instance Variables
id theSMTVView;
theSMTVView VideoText view (only interesting for VideoText).
Method Types
- afcSearch:withSystem:andNorm:
- frequency
- initFromControl:
+ isTunerConnectedAtControl:
- nextProgFor:searchUp:
- norm
- progFrequency:
- progName:
- progNorm:
- progSystem:
- progVisible:
- readProgs:
- search:withSystem:andNorm:
- setAFCTable:
- setAFCTable:andPresetWithNorm:andSystem:
- setChannel:withSystem:andNorm:
- setMonoStereo:
- setProg:
2-106 Reference - SMTVControl

Video Developer Kit
- setSearch:withWidth:andWeight:
- setVisible:theProg:
- statusAudio
- statusTV
- statusVT
- storeProg:withFrequency:system:andNorm:
- storeProg:withFrequency:system
norm:andName:
- tunerType
- writeProgs:
Class Methods
isTunerConnectedAtControl:
+ (BOOL)isTunerConnectedAtControl:(id)aControl
Returns YES if a TV-Tuner is connected to aControl. aControl has to be an initialized instance
of SMControl/MMControl class.
See also: -initFromControl:
Instance Methods
afcSearch:withSystem:andNorm:
- (int)afcSearch:(int)frequenz withSystem:(int)system andNorm:(int)norm
Searches the next valid TV channel on all AFC-frequencies starting at frequenz with system and
norm. The frequency has to be in a range from 4825 to 85525 (48.25 - 855.25 MHz). For values
of system see system (SMControl), for values of norm see norm. If no error occurred the found
frequency is returned, otherwise the error is returned. For a further description of all errors see
Errors at the end of this chapter. Before using this method an AFC-table has to be set. In every
non-error case the returned frequency is set.
See also: -setAFCTable:, -setAFCTable:andPresetWithNorm:andSystem:, -
search:withSystem:andNorm:
frequency
- (int)frequency
Returns the currently set frequency of the TV-Tuner. The frequency is in a range from 4825 to
85525 (48.25 - 855.25 MHz).
See also: -progFrequency:, setChannel:withSystem:andNorm:

Video Developer Kit
initFromControl:
- initFromControl:(id)aControl
Initializes and returns the receiver, a new SMTVControl instance with a set frequency of 48.25
MHz and a norm of SMTV_NORM_BG. aControl has to be an initialized instance of SMCon-
trol class. Before using afcSearch:withSystem:andNorm: you have to set a AFC-Table with
the setAFCTable: method.
See also: -isTunerConnectedAtControl:
nextProgFor:searchUp:
- (int)nextProgFor:(int)aProg searchUp:(BOOL)aBool
Returns the next/previous (aBool equals YES/NO) visible channel starting at aProg. If no chan-
nel is set visible, or aProg is the only visible channel, aProg is returned. In case of overflow
(larger 99/lower 0), the search starts over at the beginning/end of the program table.
See also: -setVisible:theProg, -progVisible:
norm
- (int)norm
Returns the currently set norm of the TV-Tuner. Possible return values (depending on the type
of the TV-Tuner) are :
SMTV_NORM_M M
SMTV_NORM_BG B/G
SMTV_NORM_L L
SMTV_NORM_Li L'
SMTV_NORM_I I
The norm of a TV system describes the frequency distance between picture and sound carrier.
Using wrong norms will result in no sound with a good video picture or good sound with no
picture. If you are not sure about the standard or the norm used in your country ask your local
television/radio dealer or simply experiment until the correct one is found.
See also: -progNorm:, setChannel:withSystem:andNorm:
progFrequency:
- (int)progFrequency:(int)prognr
Returns the frequency of the stored channel prognr. The frequency is in a range from 4825 to
85525 (48.25 - 855.25 MHz). A return value of -1 means that no frequency is currently set for
prognr. All other negative return values indicate that an error occurred. For a further description
of all errors see Errors at the end of this chapter.
See also: -storeProg:withFrequency:system:andNorm:, -frequency

Video Developer Kit
progName:
- (char*)progName:(int)prognr
Returns the name of the stored channel prognr. The name can contain a maximum count of 5
characters. If no name is currently set for prognr an empty string is returned. A return value of
NULL indicates that an error occurred.
See also: -storeProg:withFrequency:system:andNorm:andName:
progNorm:
- (int)progNorm:(int)prognr
Returns the norm of the stored channel prognr. For the possible return values see norm. A re-
turn value of -1 means that no norm is currently set for prognr. All other negative return values
indicate that an error occurred. For a further description of all errors see Errors at the end of
this chapter.
See also: -storeProg:withFrequency:system:andNorm:, -norm
progSystem:
- (int)progSystem:(int)prognr
Returns the system of the stored channel prognr. For the possible return values see system (SM-
Control). A return value of -1 means that no system is currently set for prognr. All other nega-
tive return values indicate that an error occurred. For a further description of all errors see
Errors at the end of this chapter.
See also: -storeProg:withFrequency:system:andNorm:, -system (SMControl)
progVisible:
- (BOOL)progVisible:(int)prognr
Returns YES if the stored channel prognr is set to visible. You needn't use this method when
switching channels with the nextProgFor:searchUp: method. By default (channel prognr has-
n't been stored yet) NO is returned.
See also: -setVisible:theProg:, -storeProg:withFrequency:system:andNorm:

Video Developer Kit
readProgs:
- (int)readProgs:(char*)dateiname
Reads an already stored program table from harddisk. dateiname should contain the complete
filename of the program table you want to read. If dateiname is an empty string (""), or could
not be read from harddisk, this method tries to open the file '.stations' in the users home direc-
tory. A return value of 0 means that an error occurred and no table was read.
See also: -writeProgs:, -storeProg:withFrequency:system:andNorm:
search:withSystem:andNorm:
- (int)search:(int)frequenz withSystem:(int)system andNorm:(int)norm
Searches the next valid TV channel starting at frequenz with system system and norm norm. The
frequency has to be in a range from 4825 to 85525 (48.25 - 855.25 MHz). For values of system
see system (SMControl), for values of norm see norm. If no error occurred the found frequency
is returned, otherwise the error is returned. For a further description of all errors see Errors at
the end of this chapter. If the returned frequency equals frequenz, is larger than frequenz plus
deltafrequenz (setSearch:::),or is larger than 85525, no valid channel has been found. In every
non-error case the returned frequency is set.
Note : This method does not start over at the beginning of the frequency spectrum. You have to
do so by your own.
In comparison to afcSearch:::,this is a distinctly slower method to find valid TV channels be-
cause all frequencies are scanned. You should only use this method if you don't have an AFC-
table for your country or if the channels your searching for are not based on the AFC-frequen-
cies of your country.
Warning : This method can't be interrupted and it can take quite a long time to scan all frequen-
cies.
See also: -setSearch:withWidth:andWeight:, -afcSearch:withSystem:andNorm:
setAFCTable:
- setAFCTable:(id)aTable
Sets an AFC-table used for the afcSearch::: method. aTable has to be an initialized instance of
SMTV_AFC_Table class. aTable will not be freed when freeing the SMTVControl class. Re-
turns self.
See also: -afcSearch:withSystem:andNorm:, -
setAFCTable:andPresetWithNorm:andSystem:

Video Developer Kit
setAFCTable:andPresetWithNorm:andSystem:
- setAFCTable:(id)aTable andPresetWithNorm:(int)aNorm andSystem:(int)aSystem
Sets an AFC-table used for the afcSearch::: method. aTable has to be an initialized instance of
SMTV_AFC_Table class. In comparison to the function above, this method presets the program
table with the values found in the AFC-table. In the AFC-table there is no information concern-
ing the standard or the norm to use. The norm norm and system system will be set for all avail-
able frequencies. The available frequencies are set to visible. aTable will not be freed when
freeing the SMTVControl class. Returns self.
Warning : Only use this method before making changes to the program table. All values of the
program table may be destroyed.
See also: -afcSearch:withSystem:andNorm:, -setAFCTable:, setVisible:theProg:
setChannel:withSystem:andNorm:
- (int)setChannel:(int)frequenz withSystem:(int)system andNorm:(int)norm
With this method you can set the frequenz (frequency), the system and the norm directly. The
frequency has to be in a range from 4825 to 85525 (48.25 - 855.25 MHz). For values of system
see system (SMControl), for values of norm see norm. If no error occurred SMTV_OK is re-
turned, otherwise the error is returned. For a further description of all errors see Errors at the
end of this chapter.
See also: -frequency, -norm, -system (SMControl)
setMonoStereo:
- (int)setMonoStereo:(int)monster
Sets the stereo chip on the TV-Tuner board. This option is not available for all boards. Possible
values for monster are :
SMTV_MONO Sets to mono sound.
SMTV_STEREO Sets to stereo sound if possible.
SMTV_A_CHANNEL Sets to left channel (only possible if twochannel is detected).
SMTV_B_CHANNEL Sets to left channel (only possible if twochannel is detected).
On success SMTV_OK else an error is returned. For a further description of all errors see Er-
rors at the end of this chapter.
Note: The Movie Machine /Pro has only a mono channel in the NTSC version.
See also: -statusAudio

Video Developer Kit
setProg:
- (int)setProg:(int)prognr
Sets the channel that is stored in the program table at position prognr. Be aware that not only
the frequency but also the currently set system and norm may change. If no error occurred
SMTV_OK otherwise the error is returned. For a further description of all errors see Errors at
the end of this chapter.
See also: -progFrequency:, -progNorm:, -progSystem:, -progName:, -readProgs:, -
storeProg:withFrequency:system:andNorm:
setSearch:withWidth:andWeight:
- (int)setSearch:(int)deltafrequenz withWidth:(int)breite andWeight:(int)gewichtung
This methods sets the limit and the adjustments for the search::: method. The default values for
breite and gewichtung (see below) are experienced for german television channels and may not
work in your country. You needn't set values for using search::: the first time.
deltafrequenz has a range from 0 to 80000 (default 80000 relates to 800MHz) and describes a
range of frequency in which should be searched for a valid channel. Keep in mind that a valid
channel could only be detected properly if you scan all over to the end of this channel. The best
picture quality of a channel you get somewhere in between the visible start and end frequency
(see also breite and gewichtung). Use value of 0 to reset to default.
breite is the minimum range of frequency (in MHz * 100) a channel is detected as a valid TV
channel. If you set this minimum to low (default is 150 equals 1.5MHz) you may detect frequen-
cies where you 'can't find' a real TV channel. Use a value of 0 to set to the default.
gewichtung is a percentage value between 0 and 100 (default 75) and represents the frequency
that is set and returned by search::: in between the found visible start and end frequency. So a
value of 100 means to use the frequency at the end of the found channel range. Use a value of
0 to set to the default.
Example: The search::: method detects a valid TV signal between 50000 and 50200 (500 - 502
MHz). Your start frequency was 45000 (450 MHz) and your deltafrequenz has been set to
10000 (100 MHz). So the end of your range hasn't been reached. The width of the channel
(breite) is larger then 150 (50200 - 50000 = 200) which means that a valid channel is detected.
Because of the weight (gewichtung) of 75 a frequency of 50150 (501.5MHz) is set and returned.
In case of error the error is returned (For a further description of all errors see Errors at the end
of this chapter) otherwise SMTV_OK.
See also: -search:withSystem:andNorm:

Video Developer Kit
setVisible:theProg:
- (int)setVisible:(BOOL)aBool theProg:(int)prognr
If aBool is Yes, sets the stored channel prognr visible. This method gives you the opportunity
to decide wether a stored channel should be switched to when using the nextProgFor:sear-
chUp: method. In case of error the error is returned (For a further description of all errors see
Errors at the end of this chapter) otherwise SMTV_OK.
See also: - progVisible, - nextProgFor:searchUp:, -
statusAudio
- (int)statusAudio
Returns the status of the stereo chip. A negative return value means that an error occurred. For
a further description of all errors see Errors at the end of this chapter. Possible non negative
return values are :
SMTV_MONO Mono sound signal detected.
SMTV_STEREO Stereo sound signal detected.
SMTV_TWOCHANNEL Twochannel sound signal detected.
SMTV_NOIDENT The signal couldn't be detected properly.
It takes the stereo chip a maximum time of about half a second after changing frequency to de-
tect a proper audio signal. If waiting less time this method can return a wrong value (in the most
cases not SMTV_NOIDENT).
See also: -setMonoStereo:
statusTV
- (int)statusTV
Returns the status of the TV-Tuner. If a valid TV channel frequency is already set a fine tuning
can be done using this method. The lower three bits of the return value indicate :
<2 Set frequency is to high.
=2 Set frequency is ok.
>2 Set frequency is to low.
You needn't use this method after using the search methods described above in this chapter. The
search methods already do that. This method may be only useful to you if you develop your own
search algorithm.
See also: -statusVT

Video Developer Kit
statusVT
- (int)statusVT
Returns the status of the VideoText chip. A set Bit 0 (SMTV_TV_QUALITY_BIT) indicates
that on the current frequency a valid video signal was detected. That means that this frequency
has a valid picture information. A set Bit 1 (SMTV_VT_QUALITY_BIT) indicates a valid Vid-
eo Text signal. Video Text is not available in most countries. Even if your TV-Tuner has no Vid-
eo Text Chip on board you can use this method to determinate the quality of the currently set
frequency.
You needn't use this method after using the search methods described above in this chapter. The
search methods already do that. This method may be only useful to you if you develop your own
search algorithm.
See also: -statusTV
- (int)storeProg:(int)prognr withFrequency:(int)frequenz system:(int)system
andNorm:(int)norm
Stores the frequency frequenz, the system system and the norm norm in the program table at po-
sition prognr. Positions from 0 to 99 are available. The range of the other parameters are de-
scribed in the methods mentioned below (see also). Like the method setVisible:theProg: this
method sets the position prognr to visible.
Note : Storage of a program channel means only to store it into the memory of the computer. If
you want to store the program table to harddisk you have to use writeProgs: if you loaded your
table with the readProgs: method.
See also: -storeProg:withFrequency:system:norm:andName:,-progFrequency:, -
progNorm:, -progSystem:
storeProg:withFrequency:system:norm:andName:
- (int)storeProg:(int)prognr withFrequency:(int)frequenz system:(int)system
norm:(int)norm andName:(char*)aName
Stores the frequency frequenz the system system the norm norm and the name aName in the pro-
gram table at position prognr. Positions from 0 to 99 are available. aName should only be a
short form of the channel name because only the first 5 characters are taken for storage. The
range of the other parameters are described in the methods mentioned below (see also). Like the
method setVisible:theProg: this method sets the position prognr to visible.

Video Developer Kit
Note : Storage of a program channel means only to store it into the memory of the computer. If
you want to store the program table to harddisk you have to use writeProgs: if you loaded your
table with the readProgs: method.
See also: -storeProg:withFrequency:system:norm:,-progFrequency:, -progNorm:, -
progSystem:, -progName:
tunerType
- (int)tunerType
Returns the type of the tuner. The different tuner types differ on the norm they support. Current-
ly two types of tuners are available :
SMTV_GR_MODUL Supported norms : M, B/G, L
SMTV_UK_MODUL Supported norms : B/G, L, L', I
For a further description of norm see norm.
Note on the specific tuners of the different VDOboards:
The optional Tuner for the Screen Machine II is a multinorm tuner. It supports NTSC, PAL and
SECAM systems.
The onboard tuners of the other VDOboards are single norm tuners, and either support NT-
SC,PAL or SECAM, depending on where you bought the board.
See also: +isTunerConnectedAtControl:
writeProgs:
- (int)writeProgs:(char*)dateiname
Writes an already stored program table from memory to harddisk. dateiname should contain the
complete filename of the program table you want to write. If dateiname is an empty string ("")
this methods tries to write the file '.stations' in the users home directory. A return value of 0
means that an error occurred and no table was written.
You needn't save every change of your program table to harddisk. Usually it is enough to save
the program table when quitting your application.
See also: -readProgs:, -storeProg:withFrequency:system:andNorm:

Video Developer Kit
Errors
SMTV_OK No Error occurred.
SMTV_ERROR General Error.
SMTV_I2CERROR Bus Error (try again and check cables and connections).
SMTV_FREQUENCYERROR Wrong frequency range.
SMTV_SYSTEMERROR Wrong system value.
SMTV_NORMERROR Wrong norm value.
SMTV_PROGRAMERROR Wrong program channel number given.
SMTV_WEIGHTERROR Wrong value for weight.
SMTV_MONOSTEREOERROR Wrong value for the stereo chip given.

Video Developer Kit
SMTV_AFC_Table
Declared In: smkit/SMTV_AFC_Table.h
Class Description
The SMTV_AFC_Table object gives you the opportunity to handle all entries in an AFC-table.
You can get all frequencies of a county an the corresponding channel descriptors stored in an
AFC-Table file. The initTableFromFile: method is the designated initializer for the
SMTV_AFC_Table class. Normally you needn't initialize this class by your own. This is done
by an initialized SMTV_AFC_TableList. You can get a specific table with a getAFCTable-
WithName:(SMTV_AFC_TableList) call. The argument for the call has to be one of the names
returned in a tableNamesForCountry:(SMTV_AFC_TableList) method.
Creating your own AFC-Table

If you want to create your own AFC-Table here's a list of the supported arguments :
%%!AFC-Table ver Has to be in the first line of an AFC-Table. ver is
an optional argument to keep track of table ver-
sions.
%% Name aName Name of the AFC-Table. aName has to be the
same as the name placed in the Countries.afcTable
file (See SMTV_AFC_TableList for further de-
scription). aName has to be quoted.
%% Entries num num has to be the number of entries in the AFC-
Table.
%% Begin Between a Begin and End statement all table en-
tries are placed. A table entry is a [TAB] separated
pair of channel descriptor and channel frequency.
The channel descriptor is the name of a channel.
The frequency is in MHz. The entries should be
sorted by frequency.
%% End Marks the end of the table entries.
All other arguments followed by %% are ignored.
Reference - SMTV_AFC_Table 2-117

Video Developer Kit
Instance Variables
int numEntries;
char tableName[1024];
numEntries Number of frequencies that have been found.
tableName[1024] Name of the AFC-table.
Method Types
- channelNameForFreq:
- channelNameForVal:
- count
- freqForVal:
- initTableFromFile:
- nextValForFreq:
- tableName
Instance Methods
channelNameForFreq:
- (char*)channelNameForFreq:(int)aFreq
Returns the channel Descriptor for the frequency aFreq. If there is no known channel at the fre-
quency aFreq, "?" is returned. The search range for this method is –3 MHz. The frequency has
to be in a range from 4825 to 85525 (48.25 - 855.25 MHz).
See also: -channelNameForVal:
channelNameForVal:
- (char*)channelNameForVal:(int)aVal
Returns the channel Descriptor for the table entry aVal. If aVal is not in the table "" is returned.
See also: -channelNameForFreq:
count
- (unsigned)count
Returns the number of entries in the AFC-Table. The first entry is number 0.
See also: -initTableFromFile:
2-118 Reference - SMTV_AFC_Table

Video Developer Kit
freqForVal:
- (int)freqForVal:(int)aVal
Return the frequency of table entry aVal. If aVal is not in the table -1 is returned.
See also: -nextValForFreq:
initTableFromFile:
- initTableFromFile:(const char*)filename
This method is the designated initializer for the SMTV_AFC_Table class. filename has to be
the full path to an AFC-Table file. If the file can't be read or no valid entries are found the class
will be freed and nil will be returned. Otherwise self is returned.
See also: -initForPath: (SMTV_AFC_TableList)
nextValForFreq:
- (int)nextValForFreq:(int)aFreq
Returns the next table entry whose frequency is larger than aFreq. If no larger frequency was
found 0 is returned.
See also: -freqForVal:
tableName
- (char*)tableName
Returns the name of the AFC-Table stored in the AFC-Table file.
See also: -initTableFromFile:
Reference - SMTV_AFC_Table 2-119

Video Developer Kit
SMTV_AFC_TableList
Inherits From: List
Declared In: smkit/SMTV_AFC_TableList.h
Class Description
The SMTV_AFC_TableList object gives you the oportunerty to handle all the countries and all
the AFC-table corresponding to this countries. The initForPath: method is the designated ini-
tializer for the SMTV_AFC_TableList class. The argument for this method is the path where all
the tables should be. This class contains a list of all AFC-tables and has methods to handle them
(tableNamesForCountry:, getAFCTableWithName:). So you do not have to know anything
about the SMTV_AFC_Table class to use the afcSearch: methods of SMTVControl class.
Creating your own AFC-Tables

If you want to use your own AFC-table (see SMTV_AFC_Table) you have to append the name
of your table to the names already written at the end of the line of your country name ('Coun-
tries.afcTable' file). In the 'Countries.afcTable' file you can also add a new country but be aware
that you have to increase the count of entries (%% Entries). The format of the 'Countries.afcTa-
ble' file and also of the other *.afcTable files are as we hope quite easy and self explanatory.
The values of norm and system are explained in SMControl (system) and SMTVControl
(norm).
To get an idea what you can do with this class examine the SMTV or the VDOtv application.
Instance Variables
int numEntries;
numEntries Number of countries that have been found.
Method Types
- countries
- getAFCTableWithName:
- initForPath:
- normForCountry:
- nrOfCountries
- systemForCountry:
- tableNamesForCountry:
2-120 Reference - SMTV_AFC_TableList

Video Developer Kit
Instance Methods
countries
- (char**)countries
Returns a NULL-terminated list of country names found in the 'Countries.afcTable' file.
See also: -nrOfCountries
getAFCTableWithName:
- getAFCTableWithName:(const char*)aName
Returns an initialized SMTV_AFC_Table class named aName. If the table named aName
couldn't be found nil is returned.
Warning : Do not free the returned table. It will be freed when freeing the SMTV_AFC_Ta-
bleList class.
See also: -tableName (SMTV_AFC_Table), -initTableFromFile: (SMTV_AFC_Table)
initForPath:
- initForPath:(const char*)my_path
This method is the designated initializer for the SMTV_AFC_TableList class. my_path should
contain the full path to the AFC-table files and the 'Countries.afcTable' file. If either no tables
or the no 'Countries.afcTable' files are found the class will be freed and nil will be returned. Oth-
erwise self is returned.
See also: -initTableFromFile: (SMTV_AFC_Table)
normForCountry:
- (int)normForCountry:(const char*)aCountry
Returns the norm that is usually in use in the country aCountry. If aCountry wasn't found -1 is
returned. For a description of all possible norms see the description of SMTVControl class.
See also: -norm (SMTVControl), -countries
nrOfCountries
- (int)nrOfCountries
Returns the number of countries that have been found in the 'Countries.afcTable' file.
See also: -countries
Reference - SMTV_AFC_TableList 2-121

Video Developer Kit
systemForCountry:
- (int)systemForCountry:(const char*)aCountry
Returns the system that is usually in use in the country aCountry. If aCountry wasn't found -1
is returned. For a description of all possible systems see the description of SMControl class.
See also: system (SMControl), -countries
tableNamesForCountry:
- (char**)tableNamesForCountry:(const char*)aCountry
Returns a NULL-terminated list of all the AFC-table names that are usually in use in the country
aCountry. Every country has at least one corresponding AFC-table. But some have more (e.g.
the USA has USA and USA Cable).
See also: -tableName (SMTV_AFC_Table), -countries
2-122 Reference - SMTV_AFC_TableList

Chapter 3
SMPalette
Video Developer Kit
InterfaceBuilder
The InterfaceBuilder is a development tool where the most used objects of user interfaces are
provided as instantiated objects. The available objects are grouped by their range of use and are
linked into InterfaceBuilder through so called palettes. These objects can be used by dragging
them from the palettes window into the appropriate working area of an InterfaceBuilder docu-
ment and simply dropping them. An object added to a user interface is fully instantiated, which
means they have their own set of object variables. For this reason, settings made for these ob-
jects are saved through the object and restored when the user interface is loaded by an applica-
tion.
An object is able to have outlets and actions. Outlets are object variables of type <id> and allow
the object to establish connections to other objects. Actions are object-methods which can be
called directly or by events of other objects.
After the initialization of an object the value of every outlet is normally NULL, so there has to
be a possibility to make other objects known to an object by assigning values to its outlets. To
do this, simply select an object and drag the mouse over the target object while pressing the con-
trol key and the left mouse button. Release the buttons over the target object and InterfaceBuild-
er will offer the possibility to make the target object an outlet of the source object through the
Connection Inspector. InterfaceBuilder is drawing a line (connection), between the object
where you started to drag the mouse and the object found under the current mouse location to
visualize the connection you are setting. Outlets can only be assigned if the source object is not
a subclass of Control, because these objects do not normally have outlets but are able to call
methods in other objects. To trigger an action, set a connection between two objects as de-
scribed above, where the starting object is a subclass of Control, e.g. a Button. You are then able
to set the action method to be called in the activated Connection Inspector.
Through objects provided in InterfaceBuilder and objects made known to InterfaceBuilder you
are able to give the designed user interface functionality just by dragging and dropping objects
and by 'drawing' connections between objects.
The outlets and action methods set in an InterfaceBuilder session are saved in special objects
embedded in the saved InterfaceBuilder files and are restored after these files are loaded in an
application.
3-124 SMPalette - InterfaceBuilder

Video Developer Kit
Palettes
Objects having related characteristics or specialized objects derived from the same class are
grouped and provided in so called palettes shown in the InterfaceBuilders palettes window. If
you are activating InterfaceBuilder for the first time there are four palettes available to the user.
In InterfaceBuilder's Tool Menu you will find a submenu entry called Load Palette which offers
the possibility to add palettes to the palettes window. Adding palettes to InterfaceBuilder offers
new sets of objects to the user and allows an extended and more detailed, easy to use way of
designing user interfaces with InterfaceBuilder. The ability to add a palette to InterfaceBuilder
was used to provide SMPalette, a set of three customized objects to build simple applications
using VDOboards.
The SMPalette provides three basic objects SMControl, SMWindow and SMPaletteView (a
subclass of SMView). These three objects are the minimum requirement to build a simple
VDOboard application.
The SMWindow is a specialized window class designed to manage VDOboard videos through
SMPaletteViews. A special window class is needed to manage the visibility of the live video
while the window is moved, rearranged or hidden.
The SMControl class is the fundamental object to control the VDOboard settings and to manage
the views assigned to one VDOboard.
SMPalette - Palettes 3-125

Video Developer Kit
The SMPaletteView is a subclass of SMView where action methods for InterfaceBuilder use
were added.
The action methods of the SMPaletteView are the following:
Method: accepted Value:
Mode
- setVideoOnOff:sender (BOOL) [sender state]
- setStill:sender; (BOOL) [sender state]
- setColorOn:sender; (BOOL) [sender state]
- setMixerOn:sender (BOOL) [sender state]
- setChromaInvert:sender; (BOOL) [sender state]
- setLumaInvert:sender; (BOOL) [sender state]
- setFlipped:sender; (BOOL) [sender state]
- setVideoInput:sender; (unsigned char) [[sender selectedCell] tag]
- setVideoFields:sender; (unsigned char) [[sender selectedCell] tag]
Quality
- setLumaIntensity:sender; (float) [sender floatValue]
- setChromaIntensity:sender; (float) [sender floatValue]
- setRedGain:sender; (float) [sender floatValue]
- setGreenGain:sender; (float) [sender floatValue]
- setBlueGain:sender; (float) [sender floatValue]
- setBrightness:sender; (float) [sender floatValue]
- setSaturation:sender; (float) [sender floatValue]
- setHue:sender; (float) [sender floatValue]
- setContrast:sender; (float) [sender floatValue]
- setSharpness:sender; (float) [sender floatValue]
- setLumaSharpness:sender; (float) [sender floatValue]
Audio
- setAudioMute:sender; (BOOL) [sender state]
- setAudioVolume:sender; (float) [sender floatValue]
- setAudioBalance:sender; (float) [sender floatValue]
- setAudioFader:sender; (float) [sender floatValue]
- setAudioTreble:sender; (float) [sender floatValue]
- setAudioBass:sender; (float) [sender floatValue]
Image
- saveImageAs:sender;
All these methods are made available through the SMControl class known to the SMPalette-
View.
If the column Value taken: says (BOOL) [sender state], the control object connected to this ac-
tion has to be a Button or any other Control which can be in the state representing on and off.
If the column Value taken: says (float) [sender floatValue], the control object connected to this
action has to be a Control offering the action floatValue which represents the state of the control
3-126 SMPalette - Palettes

Video Developer Kit
in a continuous range (e.g. the Slider provided by InterfaceBuilder). The range of the control
object should be set to a minimum of 0.0 and a maximum of 1.0.
If the column Value taken: says (unsigned char) [[sender selectedCell] tag], the control object
connected to this action has to be a Matrix of radio buttons. The buttons in this matrix should
have continuous tags.
e.g. setVideoInput:tags from 0 to 3
A tag value of 0 represents SM_INPUT_BLACK, 1 = SM_INPUT_RED, 2 = SM_INPU-

T_YELLOW and a tag value of 3 stands for SM_INPUT_SVHS.
e.g. setVideoFields: tags from 1 to 3
A tag value of 1 represents SM_FRAME_ODD, 2 = SM_FRAME_EVEN, and 3 = SM_-

FRAME_BOTH.
The method saveImageAs: does not use its argument sender. When called, this method allows
you to save the current frame as a TIFF-image. This method can be called through any control
(e.g. Button, MenuItem … ).
SMPalette - Palettes 3-127

Video Developer Kit
Features
SMControl
The SMControl has a customized inspector to indicate what type of board should be used. En-
abling a switch means the board has to support this feature. Disabling it means 'I don't care about
this feature'. If the selected features don't match the installed board types on startup time of your
application, a panel appears to indicate that no suitable board was found.
The different features are described below :

Inputs : Select the number and types of inputs the desired board has to have.
If you don't care just select no types and set the number of inputs to 0.
Outputs : Select the number and types of outputs the desired board has to have.
If you don't care just select no types and set the number of outputs to 0.
The desired board has to have the Features :

• Overlay It can display live overlay . Currently all boards support this feature.
• Frames The board has enough memory do display both video fields at once ( 1 MB).
• 4:2:2 YUV-Mode 6 is supported, otherwise 4:1:1.
• Tuner It has an on board TV-Tuner or an external TV-Tuner is connected.
• Audio It supports audio features.
• VT It has an on board or external TV-Tuner with a VideoText chip on it.
Check this tabel to find out about the features of the boards.
Overlay Full Frames YUV 4 2 2 YUV 4 1 1 Tuner Audio VT Input Output S-VHS
Screen Machine II X X X O O O 3 X
Screen Machine II Rev. 2 X X X O O O 3 X
Movie Machine X X X X (tv out) X 2+1
Movie Machine PRO X X X X (tv out) 2+1 X
Movie Machine II X X X X (tv out) X 2+2 X X (in & out)
FPS 60 X X O (mpeg) 2+2 X
X = supported O = optional
3-128 SMPalette - Features

Video Developer Kit
Note:
Through SMPalette you are only able to make the SMPaletteView do something, but there is no
possibility to get a direct response. This implies that you are not able to query the SMPalette-
View about the states or values which you can set using the SMPaletteViews action methods.
For this you have to be sure about the options installed or the methods you are using.
For example a button connected to the setAudioMute:sender method in the state of being select-
ed would cause the SMPaletteView turning the sound on through the SMControl class. Howev-
er sound can only be turned on if there is a sound option installed to your VDOboard.
SMPaletteView
The SMPaletteView has a customized inspector to indicate and change its behaviour.
• scaled video If the SMPaletteView is resized, the video will be adjusted.
• grab on stop The last frame of the video will be grabbed and displayed if the video
display is stopped.
• enable dragTo enables dragging of .flm and .tiff files into the SMPaletteView.
The dragged files will be displayed through the VDOboard.
• enable dragFrom Enables dragging of stopped images out of the SMPaletteView
if grab on stop has been set.
Default handling:
• Video Input allows the user to select a default video input for this SMPaletteView.
• VideoFields allows a default setting of the input frame type used by the SMControl
object to display live video in this SMPaletteView.
• Audio mute indicates whether audio will be turned on or off for this SMPaletteView.
• Volume allows setting of a default volume for this SMPaletteView.
The audio defaults will only take effect if an audio option is installed.
Settings of the default handling will be set when the SMPaletteView receives a start: message
or if the SMPaletteView is activated again when you are using multiple SMPaletteViews.
Every action method changing these values will update the corresponding default setting used
while the application is running.
SMPalette - Features 3-129

Video Developer Kit
Building an Application
To build a simple VDOboard application you need at least one of every object provided in the
SMPalette. There is no special order required to add these objects to your InterfaceBuilder doc-
ument. For the following example it is just the way we are using the objects.
• First we need a SMWindow

There are two ways to get a SMWindow. The first way is to change the class of a window to
SMWindow in the Window Inspector and the second way is to drag the SMWindow symbol
from the SMPalette into your working area, exactly the same way you would open a new Win-
dow from the standard palettes provided by InterfaceBuilder.
• The next step is to add a SMPaletteView into the SMWindow

To get a SMPaletteView, drag the symbol from the SMPalette into the content area of a
SMWindow. InterfaceBuilder won't prevent you from dragging a SMPaletteView into a normal
3-130 SMPalette - Building an Application

Video Developer Kit
Window, so be careful to change the class of the Window to SMWindow, otherwise the two ob-
jects won't cooperate.
• The last object needed is a SMControl object.

A SMControl object has to be dragged into the file window of the InterfaceBuilder document
you are working on. You should not use more SMControl objects than VDOboards available on
your computer.
• Create connections
If you have a SMWindow including a SMPaletteView and a SMControl object, you have to set
one connection to enable the new objects to communicate and to cooperate. Connect the SM-
PaletteView to the SMControl object and make the SMControl object the smControl outlet of
the SMPaletteView.
Minimum needed connections:

SMPaletteView smControl: ----> SMControl
If these connections are made, the SMWindow is able to control the video displayed in the SM-
PaletteView through the SMControl known by the SMPaletteView.
• Start Button
If you add a button to your interface which is connected to the start: action of the SMPalette-
View and a video source is connected to the "black input" of the VDOboard you could go into
InterfaceBuilders test mode and press the button to start the video display.
Obviously it is not much work to create a simple application using a VDOboard with the classes
provided by the SMPalette.
(See the examples for more information on building applications using InterfaceBuilder)
SMPalette - Building an Application 3-131

Video Developer Kit
MAKE
IT
VIDEO
3-132 SMPalette - Building an Application

Chapter 4
Appendix
Video Developer Kit
Variable Types and Constants

SMChromaSpace
DECLARED IN smkit/SMControl.h
SYNOPSIS
typedef struct _SMChromaSpace {

float Umin;
float Umax;
float Vmin;
float Vmax;
float Ymin;
float Ymax;
} SMChromaSpace;
DESCRIPTION
SMChromaspace is used to describe a color space for chroma and luma keying. Each of the col-
orspaces components limits can be specified. The range is between 0.0 and 1.0.
SMColor
DECLARED IN smkit/SMFLMImageRep.h
SYNOPSIS
typedef struct _SMColor{

unsigned char Y;
char U;
char V;
} SMColor;
DESCRIPTION
SMColor is used to describe a color for keying images on video out.
4-134 Appendix - Variable Types and Constants

Video Developer Kit
Television system
SYNOPSIS
System Value
SM_SYSTEM_NTSC 0
SM_SYSTEM_PAL 1
SM_SYSTEM_SECAM 2
DESCRIPTION
These constants define the television system used to decode the video signal.
Frame type
SYNOPSIS
Fields Value
SM_FRAME_ODD 1
SM_FRAME_EVEN 2
SM_FRAME_BOTH 3
DESCRIPTION
These constants define the number of fields a digitized frame consists of.
Video Input
SYNOPSIS
Input Value
SM_INPUT_BLACK 0
SM_INPUT_RED 1
SM_INPUT_YELLOW 2
SM_INPUT_SVHS 3
DESCRIPTION
These constants define the input of the VDOboard from which the video is digitized.
Appendix - Variable Types and Constants 4-135

Video Developer Kit
Video-Out Input
DECLARED IN smkit/MMControl.h
SYNOPSIS
Input Value
VIDEO_OUT_INPUT_NONE 0
VIDEO_OUT_INPUT_IMAGE 1
VIDEO_OUT_INPUT_A 2
VIDEO_OUT_INPUT_B 3
VIDEO_OUT_INPUT_TV 3
DESCRIPTION
These constants define the video out input of theVDOboards that support video out.
Memory mode
SYNOPSIS
Mode Value
SM_MEM_NORM 1
SM_MEM_SPLIT 2
DESCRIPTION
These constants define the memory mode the VDOboard digitizes the images in.
Type of image representation

SYNOPSIS
ImageRep Value
SM_IMAGE_YUV 1
SM_IMAGE_FLM 2
DESCRIPTION
These constants define the image representation used when calling the readImage: method.

Video Developer Kit
Mode for YUV and FLM Imagereps

DECLARED IN smkit/SMYUVImageRep.h
SYNOPSIS
Mode Value
SM_YUVMODE0 0
SM_YUVMODE1 1
SM_YUVMODE2 2
SM_YUVMODE3 3
SM_YUVMODE4 4
SM_YUVMODE6 6
SM_RGBMODE 8
DESCRIPTION
These constants define the mode in which the data is present in the YUV and FLM image rep-
resentations.
Audio Input
SYNOPSIS
Input Value
SM_AUDIO_INPUT_BLACK 0
SM_AUDIO_INPUT_RED 1
SM_AUDIO_INPUT_YELLOW 2
SM_AUDIO_INPUT_SVHS 0
DESCRIPTION
These constants define the input of the optional audio-on-bracket or TV-tuner used .

Video Developer Kit
Audio Signal Type

DECLARED IN smkit/SMTVControl.h
SYNOPSIS
Input Value
SMTV_MONO 2
SMTV_STEREO 3
SMTV_A_CHANNEL 4
SMTV_B_CHANNEL 5
SMTV_TWOCHANNEL 6
SMTV_NOIDENT 7
DESCRIPTION
These constants define the audio signal type of the TV-tuner used .
Television norm (audio standard)

SYNOPSIS
System Value
SMTV_NORM_M 0
SMTV_NORM_BG 1
SMTV_NORM_L 2
SMTV_NORM_Li 3
SMTV_NORM_I 4
DESCRIPTION
These constants define the television norm used to decode the audio signal.
Tuner Type
SYNOPSIS
System Value
SMTV_GR_MODUL 0
SMTV_UK_MODUL 1
DESCRIPTION
These constants define the TV-tuner type.

Video Developer Kit
Tuner Return Values

SYNOPSIS
System Value
SMTV_OK 1
SMTV_ERROR 0
SMTV_I2CERROR -1
SMTV_FREQUENCYERROR -2
SMTV_SYSTEMERROR -3
SMTV_NORMERROR -4
SMTV_PROGRAMERROR -5
SMTV_WEIGHTERROR -6
SMTV_MONOSTEREOERROR -7
DESCRIPTION
These constants define the possible return values of the SMTVContol class.
Board types
DECLARED IN smkit/VDOboards.h
SYNOPSIS
Board Value
VMCBOARD_MMPRO 0x00000001
VMCBOARD_SMII8 0x00000002
VMCBOARD_MM 0x00000008
VMCBOARD_MMII 0x00000010
VMCBOARD_FPS60 0x00000020
VMCBOARD_ALL_BOARDS 0x000000FF
DESCRIPTION
Specifies the VDOboard type.

Video Developer Kit
Board features
DECLARED IN smkit/VDOboards.h
SYNOPSIS
System Value
VMCBOARD_INPUTS 0x00000700
VMCBOARD_INPUT_SVHS 0x00000800
VMCBOARD_INPUT_NTSC 0x00001000
VMCBOARD_INPUT_PAL 0x00002000
VMCBOARD_INPUT_SECAM 0x00004000
VMCBOARD_OUTPUTS 0x00070000
VMCBOARD_OUTPUT_SVHS 0x00080000
VMCBOARD_OUTPUT_NTSC 0x00100000
VMCBOARD_OUTPUT_PAL 0x00200000
VMCBOARD_OUTPUT_SECAM 0x00400000
VMCBOARD_AUDIO 0x01000000
VMCBOARD_TUNER 0x02000000
VMCBOARD_VT 0x04000000
VMCBOARD_OVERLAY 0x08000000
VMCBOARD_422 0x10000000
VMCBOARD_SQP 0x20000000
VMCBOARD_FRAMES 0x40000000
VMCBOARD_LEILA 0x80000000
DESCRIPTION
These constants define the features of the individual VDOboards. The first and second set de-
scribe the number of inputs or outputs the board has, as well what characteristics the in or out-
outs have. e.g if they support NTSC or PAL. The third set describes the general features of a
board.

Video Developer Kit
The FLM file format

File Structure
Header
Image Data Line
•
•
•
•
•
•
•
End of Image
Icon Data Line
•
•
•
End of Icon
Additional Text
Every FLM file consist of the following main elements :

- header
- YUV image data with different ordering
- icon of the image
- additional text info
Header
The header is described in the previos figure.
Image Data
The image data consists of a certain number of rows depending on the YUV mode
and the same number of lines like the original video image. The modes and rows are
described in the next figure.
Icon
Each file also has an icon of its image at the end. The Icon always has same format
and size. (60 pixels height, 80 pixels width, YUV mode 0)
Text
Flm files may contain additional text of any length. The offset and length are set in the
header.
Appendix - The FLM file format 4-141

Video Developer Kit
File Header
0 FLM Image Header [LSB] 53 S Header of the FLM File Format

1 • 4D M
2 • 31 1
3 • 2E . Nativ PC byte ordering.
4 FLM Image Header [MSB] 30 0 64 Bytes, the first five always
5 Text End 1A contains the string "SM1.0".
6 Icon Offset [LSB] 80
7 • A0 This figure shows :
8 • 01 - offset
9 Icon Offset [MSB] 00 - meaning
10 Image Width [LSB] 20 - example hex data
11 Image Width [MSB] 01 - example interpretation
12 Image Height [LSB] B9
13 Image Height [MSB] 00
14 YUV Mode 06
15 YUV Length [LSB] 40
16 YUV Length [MSB] 02
17 Y Bits 08
18 U Bits 08
19 V Bits 08
20 Compression Mode 00
21 Old Compression 00
22 Source Mode [LSB] 01
23 Source Mode [MSB] 00
24 Icon Length [LSB] 00
25 Icon Length [MSB] 00
26 Text Offset [LSB] 00
27 • 00
28 • 00
29 Text Offset [MSB] 00
30 Text Length [LSB] 00
31 Text Length [MSB] 00
32 Contrast [LSB] 21
33 Contrast [MSB] 00
34 Brightness [LSB] 24
35 Brightness [MSB] 00
36 Saturation [LSB] 24
37 Saturation [MSB] 00
38 Hue [LSB] 3F
39 Hue [MSB] 00
40 Red [LSB] 1F
41 Red [MSB] 00
42 Green [LSB] 1F
43 Green [MSB] 00
44 Blue [LSB] 1F
45 Blue [MSB] 00
46 Reserved [LSB] 00
47 • 00
00
00
62 • 00
63 Reseverd 00
4-142 Appendix - The FLM file format

Video Developer Kit
Data Format
FLM Image Data Format

The YUV Ordering
A FBAS signal (Video and TV) consists of two "channels", one for the luminance Y
and one for the two chrominance Ur and Vb signals. The green component is calculated
out of the other two chrominance information. This was made because the human eye
has more receptors for the luminance than for color.
Most video systems use the YUV format, so Screen Machine does. Therefore the fastes
way to save an image is to store it in the original YUV format.
1 2 3 4 Video Pixel SM_YUVMODE0

YYVUYY (4 pixels; 4*Y,1*V,1*U)
Y1 Y2 V1234 U1234 Y3 Y4 File Bytes Rarely used.
mode factor : 1.5
RGB RGB RGB RGB RGB Image Pixel
1 Video Pixel SM_YUVMODE1

Y (1 pixel; 1*Y)
Y File Bytes Shortes mode, greyscale image.
mode factor : 1.0
RGB RGB Image Pixel
1 2 Video Pixel SM_YUVMODE2

YYVU (2 pixels; 2*Y,1*V,1*U)
Y1 Y2 V12 U12 File Bytes Same like mode 6, but different
byte ordering.
RGB RGB RGB Image Pixel mode factor : 2.0
1 Video Pixel SM_YUVMODE3

YVU (1 pixel; 1*Y,1*U,1*V)
Y V U File Bytes Largest but exactest format.
mode factor : 3.0
RGB RGB Image Pixel
1 2 Video Pixel SM_YUVMODE6

YUYV (2 pixels; 2*Y,1*V,1*U)
Y1 U12 Y2 V12 File Bytes Screen Machine internal format.
Commonly used.
RGB RGB RGB Image Pixel mode factor : 2.0
YUV Image Data Lengt : ( video pixels x mode factor ) x lines = bytes
FLM file length : Header 64 bytes

Image vp x mf x l bytes
Text text length bytes
+ Icon 7 680 bytes
= absolute sum bytes
Appendix - The FLM file format 4-143

Video Developer Kit
The YUV Color Model

Explanation and Mapping
The YUV format predominates in the TV technology. As a result, the brightness information
and color information are separated. To save bandwidth and also, therefore, data amount, the
color information will be transmitted with less depth in relation to the brightness information.
The reason for this is that the human eye can distinguish slighter variations in brightness than it
can in color. Screen Machine also uses this format by fully digitizing the brightness for every
picture point, but including more points per value in the color information. In this context, the
format is 4:2:2.
The YUV color space is more “extensive” as that from computers which employ RGB color
space. This means that color values can only be shown in approximation. The color space is
therefore not completely conveyed, which is, for example, distinct in the chroma keying. While
the visible RGB color space is a cube in the Cartesian coordinate system, the YUV (in English
also the YIC) color space becomes a convex polyhedron.
R = Y + 1.370705 V
G = Y - 0.698001 V - 0.337633 U
B=Y + 1.732446 U
R 1 1.37 0 Y
G = 1 $ 0.698 $ 0.338 × U
B 1 0 1.732 V
(1|1|1)
R (0|0|0)
4-144 Appendix - The YUV Color Model

Video Developer Kit
Index count
countries
118
121
D
A defaultAudioMute 76
afcSearch:withSystem:andNorm: 107 defaultFields 76
Audio Input 137 defaultInput 77
Audio Signal Type 138 defaultVolume 77
audioInput 59 doesGrabOnStop 71
B E
balance 59 enabledMovingInSMWindow: 65
bandpassFilter 36 enabledMovingInVideoView: 65
bass 59
bitsPerSample 86 F
blueGain 36 fader 60
Board 139 Fields 135
Board features 140 FLM Data Format 143
Board types 139 FLM file format 141
boardAvailable: 26 FLM Header 142
boardsPresentOfType:withFeatures: 27 Frame type 135
brightness 36 free 37, 86, 96
freqForVal: 119
C frequency 107
cancel: 63
canDragFrom 83 G
canDragTo 83 getAFCTableWithName: 121
canLoadFromStream: 86, 96 getCompression:andFactor: 87
channelNameForFreq: 118 getData: withMode: 87
channelNameForVal: 118 getFeaturesOfBoard: 28
choose: 63 getHeader: 97
chooseSM:status:pids: 63 getIcon: 97
chromaIntensity 37 getImageSize: 37
closeAllBoards 28 getKernelVersion: minor: 37, 29
contrast 37 getSplitModeGrabSize: 38
control 71 getTIFFCompressionTypes:count: 86
controlWillFree: 71 getVideoFrame: 38
Index 145
Video Developer Kit
getVideoSize: 71 initTIFFDataFromStream: 89
getWindowFrame: 38 Input 135, 136, 137, 138
getWriteHeader: 97 inputFilter 39
getZoomFrame: 38 inputFrameType 39
grab 71 inputType 40
greenGain 38 isAudioMute 60
isChromaInverted 40
H isClippingOn 72
hasAlpha 87 isColor 40
hasAudio 60 isFlipped 40
hasControl 72 isHorizontalScaled 40
hasIcon 97 isInterlaced 41
hasVideo 66 isLumaInverted 41
horizontalOffset 38 isMixed 41
hue 39 isOpaque 89
isPreFiltered 41
isScaledVideo 72
I
isStill 41
iconEnabled 97
isTunerConnectedAtControl: 107
image representation 136
isVCRTimebase 41
imageDataInMode6 87
isVideoOn 42
ImageRep 136
imageSizeFromMode: 87
imageType 39 L
imageUnfilteredFileTypes 96 lumaIntensity 42
imageUnfilteredPasteboardTypes 96
init 39, 88, 98 M
initData: 98 Memory mode136
initData: pixelsWide: pixelsHigh: memoryMode 42
YUVMode: 88 Mode 136,137
initDataFromStream: pixelsWide: mode 89
pixelsHigh: YUVMode: 88 mosaicWidth 42
initForPath: 121
initFrame:control 72 N
initFromControl: 108 newForBoardNum: 29
initFromPasteboard: 89, 98 newForBoardNum:fromZone: 29
initFromStream: 98 newWithSelectionAndBoards
initTableFromFile: 119 :withFeatures: 29
146 Index
Video Developer Kit
newWithSelectionAndBoards: S
withFeatures:fromZone: 29 saturation 44
nextProgFor:searchUp: 108 saveImageAs: 77
nextValForFreq: 119 search:withSystem:andNorm: 110
noiseFilter 42, 43 setAFCTable: 110
norm 108 setAFCTable:andPresetWithNorm:
normForCountry: 121 andSystem: 111
nrOfCountries 121, 42 setAlpha: 90
numberOfOutputs 43 setAudioBalance: 60, 77
numColors 89 setAudioBass: 60, 77
setAudioFader: 60, 77
O setAudioInput: 61
outputType: 43 setAudioMute: 78
setAudioTreble: 61, 78
setAudioVolume: 61, 78
P
setBandpassFilter: 45
pll 43
setBitsPerSample: 90
posterization 43
setBlueGain: 45, 78
progFrequency: 108
setBrightness: 45, 78
progName: 109
setChannel:withSystem:andNorm: 111
progNorm: 109
setChromaIntensity: 45, 78
progSystem: 109
setChromaInvert: 45, 78
progVisible: 109
setClippingOn: 73
setColorOn: 46, 79
R setCompression:andFactor: 90
rbgKeyColor 103
setContrast: 46, 79
read: 90
setControl: 73
read: 98
setDefaultAudioMute: 79
readImage 44
setDefaultFields: 79
readImageSelection: 44
setDefaultInput: 79
readProgs: 110
setDefaultVolume: 80
redGain 44
setDragFrom 83
redrawClips 72
setDragTo: 83
registerVideoView: 66
setFlipped: 46, 80
reset 44
setGrabOnStop 73
resetClips 73
setGreenGain: 46, 80
setHorizontalOffset: 47
setHorizontalScaled: 47
Index 147
Video Developer Kit
setHue: 47, 80 setVideoFields: 82

setIconEnabled: 99 setVideoFrame: 52
setImageType: 47 setVideoInput: 53, 82
setInputFilter: 47 setVideoOn: 53
setInputFrameType: 48 setVideoOnOff: 82
setInterlaced: 48 setVideoOut: 56
setKeyColorFromPoint: 103 setVideoOutInput: 56
setKeyingOn: 103 setVideoOutInvert: 56
setLumaIntensity: 48, 80 setVideoOutScaled: 56
setLumaInvert: 48, 80 setVideoOutSystem: 56
setLumaSharpness: 81 setVideoOutXofs: 57
setMemoryMode: 49 setVideoOutYofs: 57
setMixerOn: 49, 81 setVideoSize: 74
setMonoStereo: 111 setVisible:theProg: 113
setMosaicWidth: 49 setWindowFrame: 53
setMovingInSMWindow: 66 setWriteHeader: 99
setMovingInVideoView: 66 setWriteText: 99
setNoiseFilter: 50 setYUVKeyColor: 104
setNumColors: 90 setZoomFrame: 53
setOpaque: 91 sharpness 54
setPixelsHigh: 91 showVideoInMiniWindow 66, 67
setPixelsWide: 91 SMChromaSpace 134
setPll: 50 SMColor 134
setPosterization: 50 SMConvertRGBToYUV 104
setPreFilter: 50 SMConvertYUVToRGB 104
setProg: 112 smNum 54
setRedGain: 50, 81 start: 74
setRGBKeyColor: 103 statusAudio 113
setSaturation: 51, 81 statusTV 113
setScaledVideo: 74 statusVT 114
setSearch:withWidth:andWeight: 112 stop: 74
setSharpness: 51, 81 storeProg:withFrequency:system:
setSize: 91 andNorm: 114
setSplitMemGrabSize: 51 storeProg:withFrequency:system:
setStill: 51, 81 norm:andName: 114
setSystem: 52 suspendClipping: 74
setVCRTimebase: 52 System 135,138, 139, 140
setVerticalOffset: 52 system 54
148 Index
Video Developer Kit
systemForCountry: 122 writeTIFF: usingCompression: 92

writeTIFF: usingCompression: andFactor:92
T
tableName 119 Y
tableNamesForCountry: 122 YUV Color Model144
Television norm 138 YUV Modes 137
Television system 135 yuvKeyColor 104
text 99
treble 61
Tuner Return Values 139
Tuner Type 138
tunerType 115
U
unregisterVideoView: 67
updateDisplayData: 91
usesKeying 104
V
verticalOffset 54
Video Input 135
Video-Out Input 136
videoInput 54
videoOutInput 57
videoOutSystem 57
videoOutXofs 57
videoOutYofs 58
videoViews 67
volume 61
W
write: 91, 99
writeFLM: withMode: 100
writeImage: 54, 58
writeProgs: 115
writeText 100
writeTIFF: 92
Index 149
Video Developer Kit
150 Index

SM-II Developer Kit

Uploaded by

Copyright:

Available Formats

SM-II Developer Kit

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SM-II Developer Kit

Uploaded by

Copyright:

Available Formats

VIDEO Developer Kit

NeXT Version 2.0

Screen Machine II! is a registered trademark of FAST Multimedia AG, München.

1-6 Concepts - Introduction

Board Special Features

How the VDOboards work

Concepts - The VDOboards 1-7

// We need an instance of NXApp for the control class

// Get instance of control class

1-8 Concepts - The VDOboards

// Set output window of 360 by 240 at 20.0,20.0

// Wait for keypress

// Turn video off

// Free control & App

With a simple loop you could animate the window:

// Grab the current live picture as image

Concepts - The VDOboards 1-9

Here is some code to display an image in the video window:

myImageRep = [SMYUVImageRep alloc];

1-10 Concepts - The VDOboards

The Display Details

Concepts - The Display Details 1-11

1-12 Concepts - The Display Details

Things you should always do

Concepts - Things you should always do 1-13

Split Memory Mode

The call to do this is setMemoryMode:mode offset:offset width:width.

1-14 Concepts - Split Memory Mode

Example: Split memory mode for a PAL frame

// Set split mem bounds

// Set size of grabed image

You can also try this with the SMParser:

smp> mem 2 406272 736

Concepts - Split Memory Mode 1-15

Integration Into NEXTSTEP

1-16 Concepts - Integration Into NEXTSTEP

• Uses SMWindow, SMView and SMControl

Concepts - Examples 1-17

This is what we are set out to create.

Start a new application in InterfaceBuilder. (Command-n)

1. Delete the window supplied.

1-18 Concepts - Examples

2. Drag a SMWindow onto the screen.

Concepts - Examples 1-19

SplitIt is a application with the following features :

• Uses split memory mode

In split memory mode the memory is divided into two regions:

1-20 Concepts - Examples

Calculate the offset :

To retain the NTSC ratio = 640/480=4/3 we get :

Concepts - Examples 1-21

1-22 Concepts - Animations

Object Responder Window SMWindow

NXImageRep SMYUVImageRep SMFLMImageRep

The Kernel / Driver

The SMControl class

2-24 Reference - Classes Overview

SMYUVImageRep, SMFLMImageRep, MMFLMImageRep

Reference - Classes Overview 2-25

Declared In: <smkit/VDOboards.h>