Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
rrss
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
37 ms
·
391.
▲
by
rrss
6y ago
Note the results you quoted are for a test case in which all instructions are back-to-back dependent, which is good for testing the hardware and developing performance models, but not representative of well-tuned GPU codes. > doesn'
392.
▲
by
rrss
6y ago
I just watched a couple of his videos, and it sounds like he got pretty close to landing with solid propellant ('echo'), but ultimately decided that it was too finicky and ended that project to move onto a liquid biprop rocket.
393.
▲
by
rrss
6y ago
It seems like it would be really hard to land the boosters with no throttle.
394.
▲
by
rrss
6y ago
I don't think it's useful to generalize about whether people 'should' 'stay and fight' or not - it depends so much on the individuals and on the circumstances, but anyway: Ultra was able to successfully decrypt
395.
▲
by
rrss
6y ago
yeah, even without considering who is likely and able to emigrate, HK median income is higher than the median income of some U.S. states.
396.
▲
by
rrss
6y ago
I agree for cars and planes, but for spacecraft, the crew doesn't really need to "keep their eyes on the road," so to speak (and it's not like the spacecraft controls can be done without looking, anyway, like turning up
397.
▲
by
rrss
6y ago
No. People who are concerned about 5G deployment for non-conspiracy theory or ignorance reasons are concerned about unknown effects of increased exposure to nonionizing radiation. X-rays are very much different from the sub-100 GHz bands us
398.
▲
by
rrss
6y ago
Einstein sent his papers to Planck, and Planck, as an editor of Annalen der Physik, decided to publish them in the journal. Here's what Einstein had to say about the modern peer-review process in which a draft is reviewed by anonymous
399.
▲
by
rrss
6y ago
There are several reasons why people would do this, but the decision is often not made by the person writing the code, so it may be a red flag to you, but it doesn't point to inexperience of the author. In addition to the MSVC issue me
400.
▲
by
rrss
6y ago
it can also just indicate that when it was written it also had to compile as C++ for whatever reason.
401.
▲
by
rrss
6y ago
FP16 tensorcore = 312 tflops x 2 (counting sparse as dense) = 624 tflops x 8 GPUs = 5 "pflops" The missing 8x you are looking for is just because tensorcore math is much faster than their normal fma path.
402.
▲
by
rrss
6y ago
Looks like yes: https://news.ycombinator.com/item?id=23240076
403.
▲
by
rrss
6y ago
I doubt Microsoft really cares. Microsoft will ship it in the linux kernel in WSL2, and people will use it, and if the upstream community doesn't want it I don't see why Microsoft would have a problem with that.
404.
▲
by
rrss
6y ago
Yeah, reasonable, as in: the united states put full suppression measures in place piecemeal sometime between 0.2 and 1.6 deaths / 100k / week (depending on jurisdiction), and the actual number of deaths is trending to end up betwe
405.
▲
by
rrss
6y ago
This article is not very substantive, so: what's actually wrong with the model (in terms of predictions), and what actions should the UK government have taken instead? For the U.S., the Imperial College team predicted 2.6M deaths if no
406.
▲
by
rrss
6y ago
> which it did - given Neil rolled back his predictions after a while Can you please source this? Sorry to have to ask, but last time someone said this it turned out they were comparing predictions in dissimilar scenarios from two differ
407.
▲
Rendering Millions of Dynamic Lights in Real-Time
(news.developer.nvidia.com)
2 points
by
rrss
6y ago
|
0 comments
408.
▲
by
rrss
6y ago
> 8x from somewhere else that I have no idea. 8 GPUs in the box.
409.
▲
by
rrss
6y ago
technically the first customer is Argonne National Labs, which is DoE, not DoD. https://www.globenewswire.com/news-release/2020/05/14/203353...
410.
▲
by
rrss
6y ago
Note that the wafertech/TSMC fab11 only does processes down to 0.16 um. https://www.wafertech.com/en/foundry/technology.html
411.
▲
by
rrss
6y ago
Speaking of vapor, still waiting for rex computing to ship...
412.
▲
by
rrss
6y ago
'brain floating point' is also bad, but no one cares because it's just bfloat16. If this becomes popular, it will just be tf32 or tfloat32 or something.
413.
▲
by
rrss
6y ago
> they may not actually know which permissions need to be changed How can they not know? They decide whether the update is accepted or rejected, and there's somebody or something at google that makes that decision, so google has to
414.
▲
by
rrss
6y ago
From a certain point of view, modern deep neural networks for audio are 'just' nonlinear adaptive filters on steriods. Linear adaptive filters have been around for a long long time, and nowadays are everywhere. They can't cap
415.
▲
by
rrss
6y ago
My thinking is that the exemption allows for counties to have the opportunity to allow those businesses in critical infrastructure sectors to stay open. The order says that employees "may continue their work," and Tesla's law
416.
▲
by
rrss
6y ago
Is this legal argument actually any good? It seems like the major claim is that Alameda County's order "overrides an express order of the Governor of California," which doesn't make much sense to me. Newsom's order
417.
▲
by
rrss
6y ago
That presentation is pretty old - must be over a decade since the example uses G80. Nvidia has shipped 5 or 6 new architectures since then. Kepler introduced a 'shuffle' intrinsic to accelerate the kind of 'horizontal' r
418.
▲
by
rrss
6y ago
This was really fun to read, thanks fsanglard. > This is correlated with the warning nvcc issued. Because the raytracer uses recursion, it uses a lot of stacks. So much actually that the SM cannot keep more than a few alive. Stack frame
419.
▲
by
rrss
6y ago
Eh, YMMV. I tried cmder/conemu for a while, and gave up after I fighting with one strange behavior after another for a few days. IMO, the best terminal on windows is a linux terminal running in WSL displaying to a windows x server. I u
420.
▲
by
rrss
6y ago
> it's Bloomberg citing one person only, it's not a scientific paper. Not sure I follow, I see a lot more citations than one person. FTA: https://ashpublications.org/blood/article/doi/10.1182/
More ›