Comments by "Siana Gearz" (@SianaGearz) on "Nvidia's Computational Lithography Breakthrough" video.

  1.  @Ritternkreis  Yes he does it alone. But he also has an educational and work background in semiconductor manufacturing, which is how he's able to do it.
    16
  2. NVidia's weight in HPC has become so enormous, this little niche of semiconductor design doesn't really move the needle much one way or another. AMD could get their shit together but doesn't; and of course there's Intel with Xeon Phi but they think that whoever hasn't run away to NVidia cannot move, that they have a captive customer base. I don't know that this sort of complacency is a good way to go about things, it usually doesn't end well. If anything this here goes to show why NV wins, because they keep pushing the field forward, they are not complacent not even for a second. Perhaps they deserve this monopoly, who knows, even though we all know it's not good.
    3
  3. The progress is growing slower harder and more expensive, but we appear to be far enough from the hard limits to be busy for a couple decades yet. Asymptotically approaching?
    2
  4. We've been at this stage for a long time. None of the advanced chips of today are possible without advanced computational optics.
    1
  5. Or indeed fortunate.
    1
  6. Yeah but you spin up a chip at an enormous expense, it's super niche single use case, and then after a short time it's technically obsolete such that the next general purpose chip can outpace it, what's the point? This is why niche tasks even heavyweight ones need to be done in software with off the shelf hardware. After you're able to get everything out of the off the shelf hardware such that it really can't get better and have gained a crucial insight about the problem field, maybe you can design custom. It's funny how much things changed. Back in the 80s throwing together a custom chip was that much more par for the course. Of course it was lower complexity and cheaper and then off the shelf options were so limited in flexibility.
    1
  7. It makes me sad that we don't have chinatowns in Europe. Well i know there are one or two but in none of the large cities that i tend to be in. Or even not neessarily chinatowns but some sort of cute weird enclaves that straddle the line between a theme park and a lived-in livable space.
    1
  8.  @jannegrey  HPC OGs are Intel with AMD and in the past the likes of Fujitsu with their DSPs, one or another odd entrant. Nvidia is sort of the new guy, sweeping up from under both categories, both microprocessor and DSP based. Though honestly DSP as they used to be were on borrowed time anyway ever since CPUs gained SIMD. And GPGPUs have more SIMD density.
    1
  9. I wonder, does it matter that the chips that are targeted are huge? They do after all consist of repeat elements, so doesn't it suffice to run computationla lithography on smaller building blocks and just copy it over?
    1