e
Yes, it works out to a ton of power and money, but on the other hand, 2x the computation could be like a few percent better in results. so it’s often a thing of orders of magnitude, because that’s what is needed for a sufficiently noticeable difference in use.
basing things on theoretical tops is also not particularly equivalent to performance in actual use, it just gives a very general idea of a perfect workload.
They’re also the second largest battery manufacturer in the world, behind CATL, another Chinese company.
.008%, you forgot to convert to a percentage
Still very small though
At the datacenter scale Gaudi 3 was pretty good, at least when it came out.
Intel GPU support?
ZLUDA previously supported Intel GPUs, but not currently. It is possible to revive the Intel backend. The development team is focusing on high‑quality AMD GPU support and welcomes contributions.
Anyways, no actual AI company is going to buy $100M of AI cards just to run all of their software through an unfinished community made translation layer, no matter how good it becomes.
OneAPI is decent, but apparently usually fairly cumbersome to work with and people prefer to write software in cuda as it’s the industry standard (and the standard in academia)
Intel’s Gaudi 3 datacenter GPU from late 2024 advertises about 1800 tops in fp8, at 3.1 tops/w. Google’s mid 2025 TPU v7 advertises 4600 tops fp8, at 4.7 tops/w. Which is a difference, but not that dramatic of one. The reason it is so small is that GPUs are basically TPUs already; almost as much die space as is allocated to actual shader units is allocated to matrix accelerators. I have heard anecdotally.
It’s not even a pivot. They’ve been focusing on AI already. I’m sure they want it to seem like a pivot (and build up hype); the times before apparently just having the hardware and software wasn’t enough. nobody cared when the gaudi cards came out, nobody uses sycl or onednn, etc
i remember this from the paper (pg. 14)
i love godot but it seems like such overkill for a clock lol
although I like a lot of what Valve does (I have a lot of Steam games, valve games, have a steam deck oled, use steamvr, etc) they are a fairly flawed company. sweeney is so great at shooting himself in the foot though that any opinion he has people will by default believe the opposite of (and probably should)
I doubt people favor the men’s teams because they perform better, they’re not even playing against women’s leagues so viewers don’t have an automatic comparison of performance. If every basketball player’s free throw percentage went down by half, I don’t think people would enjoy the sport that much less, because I think people care more about the rankings etc than the absolute value. So I think the problem is mostly that the men’s teams came first and already built up a following, and it’s more fun to follow a team with a bigger following, so everyone just follows the men’s teams. There probably isn’t that much of a solution to this, because even with some meta-classification thing like i’m proposing people might still mostly follow the top absolute performance league which would be mostly men (depending on how much of an advantage gender gives in a specific sport). in the ideal society where gender stops mattering as much, then it could become ‘just another factor’ like height and weight that the people in the leagues with less potential performance would have
i don’t know, i don’t think there is a perfect solution or even a very good solution
I think it was specifically when the subject of the sentence was not the most recent name that it seemed the most ambiguous.
As in "bob disliked joe because he … " and if the context doesn’t make it explicit who ‘he’ is then you just can’t tell.
yeah, you could just use names when it might be ambiguous, which isn’t all that often anyways, but i don’t think pronouns really need to be that limited
just idle conversation tho because obviously replacing a language on demand requires a huge amount of effort which isn’t going to happen anytime soon for such relatively little gain
motors have been getting better but having both the max speed and max strength of a human arm at the same time in that small a space with a reasonable cost and power consumption remains difficult afaik
but it would be so much fun to design new arm parts, little gizmos etc
i would absolutely make my arm play doom, have a built in flashlight, battery bank, reactive rgb lights, modular customization, etc. since I have to be wearing a bulky electronic thing anyways…
there’s a guy on youtube who lost most of his hand who’s machined a very incredible purely mechanical replacement (no electronics)
Note that this was a 10% cut. Meta has a lot of staff. They are also slightly pivoting to AR glasses, as they say Ray-Ban Display sales are much higher than anticipated. I think they see their chance of cornering the VR market as gone (probably because of android xr) (see https://www.uploadvr.com/meta-pauses-third-party-horizon-os-headsets/ ) but think they have a chance to dominate AR glasses / wearables.
i have played halo infinite a lot (~200 hrs, not that much but my highest fps game) and the way its designed it always pairs me with people who are pretty evenly matched, with occasional ones outside of the norm. which is fine for moment-to-moment gameplay but takes away any sense of how good I actually am compared to the average (and any sense of progression that could come with that)
I have ADHD but not autism and I do this. It’s basically normal ADHD executive dysfunction I think, for me at least
i pronounce it like chalk-oh talk-oh
there are a lot more comments on their profile on piefed, some must be in communities that aren’t federated here or smth