rrss

67 ms·

661.
by rrss 7y ago
What more does it take?
662.
by rrss 7y ago
Some of the early tube computers had some really interesting circuits. One that comes to mind is the use of "kirchoff adders," which are more-or-less analog adders with easy one-bit DACs and ADCs on the inputs/outputs. I can&
663.
by rrss 7y ago
yeah, electrical power propagates way faster than electrons drift (and the power flux is primarily outside the wire, but that's a fun story for a different day). One way to think about it is as a newton's cradle sorta deal. It&#
664.
by rrss 7y ago
"You guys have got a lot of nice words! Any way we can get some of those words over here?"
665.
by rrss 7y ago
Right, the exhaust is electrically neutral, but only the positive ions are accelerated and generate thrust. So if the ions return eventually to their original potential (as 'Steuard described), there cannot be any net change in momentu
666.
by rrss 7y ago
ELI5: Given this, how does an ion drive work?
667.
by rrss 7y ago
If the reasoning there is sound (and it makes sense to me), the closer analogy is trying to propel a canoe by rolling a cannonball along the bottom. It will work for a while, but eventually the cannonball hits the back of the canoe. No dice
668.
by rrss 7y ago
What's the battle there? How do physicists define impedance?
669.
by rrss 7y ago
> why won't everyone migrate to Vulkan compute shaders? 1. No libraries. cudnn, cublas, cufft, are all the fastest available (except maybe magma sometimes), plus no one writing an actual application wants to reinvent a fast gemm. Al
670.
by rrss 7y ago
(This is entirely unrelated to the content of this submission, but I'm curious: I'm pretty sure I submitted this more than 24 hours before the time that HN now shows, and it didn't get much attention. But then it somehow got
671.
by rrss 7y ago
> CUDA no longer supports arm outside of tegra. ARM support is coming (back), apparently: https://www.nextplatform.com/2019/06/17/nvidia-makes-arm-a-p...
672.
by rrss 7y ago
> Keep talking about the software. Make it about the software, so no one knows the airframe is fundamentally flawed. It's pretty clear that the author of the comment I originally responded to was not considering software.
673.
GPUDirect Storage: A Direct Path Between Storage and GPU Memory(devblogs.nvidia.com)
69 points by rrss 7y ago|25 comments
674.
Discovering the Structure of RDNA(gpuopen.com)
2 points by rrss 7y ago|0 comments
675.
by rrss 7y ago
> OpenCL since it's portable Even if you choose OpenCL, the tools (profiler, debugger, etc) are usually platform specific. In addition, my experience with opencl across platforms was that each of the vendors' compilers had dist
676.
by rrss 7y ago
FWIW ROCm, which is what AMD has indicated they will be investing in moving forward, doesn't support OpenCL 2.0 either.
677.
by rrss 7y ago
I'm genuinely curious what field you work in and what problems you work on that IMO level algebra and combinatorics skills are frequently useful (IMO = the olympiad, not my opinion). In the fields I have experience with, the usually ap
678.
by rrss 7y ago
My pages apparently don't align with yours (I see page 46 has only a single exercise), but I don't see where Rudin says anything is "good enough." He states the definition of convergence, meaning that if a sequence satis
679.
by rrss 7y ago
I agree that that is typically true, but in the particular case of the 2018 tariffs on China, the cost pretty much fell entirely on consumers. > Overall, using standard economic methods, we find that the full incidence of the tariff fall
680.
by rrss 7y ago
The United States is not homogeneous, and the current government does not reflect the will of 100% of the people. It is entirely possible for one fraction of the population to be concerned about spreading extremism while another fraction do
681.
by rrss 7y ago
Anyone remember when rust had ~ and @?
682.
by rrss 7y ago
"Peer review" nowadays refers almost exclusively to prepublication review by people other than the editor. This was still relatively rare at the beginning of the 20th century. If you name a famous publication prior to 1900, it alm
683.
by rrss 7y ago
That would be a very unusual performance distribution. I've seen a lot of distributions, and they included a load of right-skewed distributions, some normal distributions, but I honestly can't think of any case with a left-skewed
684.
by rrss 7y ago
> using minimum values is almost never a good idea Care to explain why? I find the argument in http://blog.kevmod.com/2016/06/benchmarking-minimum-vs-avera... for using the minimum to be pretty compelling. Also
685.
by rrss 7y ago
> That's an interesting use of the word 'only' A ~thousand to several thousand dollar product with a market penetration of 24% (low) becomes completely free , and the market penetration shifts to 34% (still low). I think
686.
by rrss 7y ago
> I think the S in SMT is for symmetric It's definitely for simultaneous. You are probably thinking of SMP (symmetric multiprocessing).
687.
by rrss 7y ago
> That's simple economics. Sure, it's simple economics, but in this case it is too simple. It turns out that the hearing aid market has inelastic demand [1][2]. In 2010, market penetration for hearing aids in the U.S. was aroun
688.
by rrss 7y ago
> numbers are large enough How many people have hearing aids, though? Seems like the market for this would be much smaller than other things and therefore the price higher.
689.
by rrss 7y ago
> This stands out to me since doing that on a battery powered device is unusual This is absolutely not true. I would bet that > 50% of all devices shipping with a digital signal processor are doing FFT/IFFT at some point. > us
690.
by rrss 7y ago
> Many bullets (SPECTRE, etc.) have been dodged Rail control systems run untrusted code? Why not dodge Spectre by not running random code in the control system?

More ›