paulmd

34 ms·

271.
by paulmd 2y ago
This stuff is the actual reason nvidia is under antitrust investigation. boo boo, a GTX 670 that cost you $399 in 2012 now costs $599 - grow up, do the inflation calculation, and realize you’re being a child. gamers get the best deal on bul
272.
by paulmd 2y ago
Columbo: “Say, you aren’t posting on a throwaway because your main got flagged for disinformation, are ya?” https://dl.acm.org/doi/10.1145/3501247.3531573
273.
by paulmd 2y ago
They were also the first to usb-c nirvana - they were shipping laptops with thunderbolt in 2011, and moved to usb-c in 2016, giving you four full-capability outputs at a time when most laptops had one , sometimes, and it took another 5 yea
274.
by paulmd 2y ago
elsevier isn’t paying authors a dime lmao, the authors are the ones paying here. A publisher would never deign to pay an academic . They get “paid in exposure”, and again, actually they have to pay for that exposure in the first place. El
275.
by paulmd 2y ago
Nvidia. Pretty much the only founder-led megacompany, still disrupting the world. Pushing forward at a time when the competition decided to pull the brake on dGPU development, pushing forward on gpgpu long before it was cool (many years of
276.
by paulmd 2y ago
Wow, you still have an expense budget for routine employee lunches such that you’re doing work out at restaurants and stuff!? /s
277.
by paulmd 2y ago
simply imagine a person as a unitary data cube (all dimensions between (0,1)) with all personality traits and disorders expressed as dimensions … the question is, do you have the right embeddings? ;) (/s, mostly)
278.
by paulmd 2y ago
I thought this was going to be the xkcd ‘int rand() {return 6; //chosen by fair dice roll}’, which is more or less what they did lol.
279.
by paulmd 2y ago
https://en.wikipedia.org/wiki/Cross-device_tracking#Ultrason...
280.
by paulmd 2y ago
https://www.youtube.com/watch?v=AaFkXsdr0mk (incidentally, this flow over the base of the dam is new, looks like since 2018 (it normally goes through culverts at the base) and the dam likely won't survive much longer,
281.
by paulmd 2y ago
When the proverbial one-horse town/single-industry company town has the single-industry leave town, what exactly do you want anyone to do about it? We already shovel in far more tax dollars maintaining infrastructure for a handful of r
282.
by paulmd 2y ago
Sounds like a good application for a neural model to sieve out the stuff you’re interested in. (And that stuff actually existed for a long time, CRM-114 is an example but not the first, I was playing with this in the early 00s with stuff th
283.
by paulmd 2y ago
why can fundamentally stochastic things not be modeled? monte carlo simulations were literally some of the first computer programs.
284.
by paulmd 2y ago
Obese people generally have lower lifetime healthcare spending. The third knee replacement and a year in assisted living is expensive, and in the US that senior-care cost is primarily borne by the government as opposed to the health insuran
285.
by paulmd 2y ago
Link seems to have gone dead since earlier, lol, but: https://en.wikipedia.org/wiki/AMD_APP_SDK
286.
by paulmd 2y ago
You seem to be tangentially attempting to refute a claim I didn't make / responding to a comment I didn't make? Are you lost? Yes, it's probably true that in 2009 or whatever it wasn't on NVIDIA's radar as bei
287.
by paulmd 2y ago
> It’s questionable whether or not he anticipated this specific market vs he just positioned the company to eat up sales to high energy particle physics & similar super-computer applications and then got lucky it's actually spec
288.
by paulmd 2y ago
It's not anticompetitive to try and foster fields of software in which your competitors are underinvested/underperforming/disadvantaged. AMD is trying to do the same thing right now - AI/NPU is an area where they think
289.
by paulmd 2y ago
AMD actually had more resources than NVIDIA for a good long while (although they did have more places to spread them over, that was their choice to do so), and the balance of power didn't really shift until the 2014-era when AMD real
290.
by paulmd 2y ago
The biggest problem with SyCL is that AMD doesn’t want to back a horse that they don’t control (same reason they opposed streamline) so they won’t support it. When the #2 player in a 2-player market won’t play ball, you don’t have a standar
291.
by paulmd 2y ago
AMD’s first attempt at displacing CUDA-as-a-runtime was called AMD APP (advanced parallel processing) so this layer did indeed exist. It just sucked as badly as the rest of AMD’s slop. https://john.cs.olemiss.edu/heroes&#x2F
292.
by paulmd 2y ago
GPL is also part of the problem. Linux will simply never support HDMI 2.1 properly because the license precludes it. It will never support OSS like ZFS properly because the license precludes it. It is a restrictive/closed copyleft ecos
293.
by paulmd 2y ago
Well, people don’t like the “other” analogy - the “iPhone gold rush” basically never ended. https://cdn.statcdn.com/Infographic/images/normal/7800.jpeg When you invent a whole new class of device, of which yo
294.
by paulmd 2y ago
Really? What hardware was llama3 trained on? now realize there’s 15 other models you haven’t heard of trained on the same brand’s hardware that didn’t work out / weren’t impressive enough to release. you are literally describing Jevons
295.
by paulmd 2y ago
> I've seen so much ridiculous hype about AI becoming godlike super-geniuses before the decade is out I’ve actually seen literally none of that from proponents, like I’ve probably seen 100x as much strawmanning and uncharitable summ
296.
by paulmd 2y ago
Well, this one does, so what’s the point of this observation in this context?
297.
by paulmd 2y ago
I actually think the CPU and GPU meeting at the idea of SIMT would be very apropos. AVX-512/AVX10 has mask registers which work just like CUDA lanes in the sense of allowing lockstep iteration while masking off lanes where it “doesn’t
298.
by paulmd 2y ago
> This is more Intel than AMD. AMD will generally stick to the rated TDP unless you manually reconfigure something. no, it's generally the opposite. Intel tends to exceed TDP substantially due to boost, and AMD actually exceeds it
299.
by paulmd 2y ago
yes, but if the 7945HX3D is running at 112W (75W cTDP + AMD allows unlimited-duration boost at 50% above this) and the macbook is running at 50W cpu-only (70W TDP) then the macbook is actually substantially more efficient in this comparison
300.
by paulmd 2y ago
i mean, macos is just unix, and you will do a lot of tinkering given how many bad defaults it's got, and how many weird foibles etc. I have a script which fixes all my file assignments after a brew upgrade because macos grabs them bac

More ›