4 ms·

Fierce competition in this space is a good thing, we sorely need faster matrix multiplication for everything from graphics processing to the grammar checker in

by dsign 5y ago

Fierce competition in this space is a good thing, we sorely need faster matrix multiplication for everything from graphics processing to the grammar checker in [insert name of your word-processing software]. So yes, more transistors please.

throwaway4good 5y ago
Do we really? (Serious question - what are the end-user applications that cannot be done today because of the lack of faster matrix multiplication?)
- pfortuny 5y ago
  VR headsets, games running at 4K 120fps, etc… Anything using massive linear algebra.
  - throwaway4good 5y ago
    4K 120fps - is that "sorely needed"?
    - belval 5y ago
      Of course not and "640K software is all the memory anybody would ever need on a computer".
      - dmitriid 5y ago
        False equivalence
      - throwaway4good 5y ago
        No. I am asking if there are genuine new use cases that will be enabled by say a ten-fold increase in processing power? Other than it is nice going from 2K at 30 Hz to 4K at 120 Hz (which is an 8-fold increase).
        belval 5y ago
        Your question is difficult to answer in good faith because "everything" would benefit from a ten-fold increase in processing power. Super computers could process protein foldings faster, any kind of simulation really could benefit from it. Deep learning could eat a ten-fold increase overnight (especially if accompanied by faster buses and memory) gaming at 30Hz is really not great, gaming at 60/120/144Hz is much better. VR was already mentioned in this very thread. I answered your question with a joke because your question is a joke, what doesn't benefit from a 10-fold increase in processing power? Why would no new use case arise from a widely available 10-fold increase in processing power when the last 40 years have shown that new tech always materializes? Or maybe you just wanted to make a cynical statement, that all this tech from the last 20 years was pointless, that social media are a net negative for the world, that better video game graphics are pointless because only gameplay matter, that smartphones (only possible because of a previous wave of 10-fold increases) are only addictive little screens and not actually useful in everyday life? Yes there are genuine new use cases that would appear with a ten-fold increase in processing power. It's a certainty.
        Dylan16807 5y ago
        > I answered your question with a joke because your question is a joke, what doesn't benefit from a 10-fold increase in processing power? You misread the question pretty badly. They asked for new use cases, not things that would benefit. We can fold proteins pretty well now, 10x wouldn't fundamentally change things. There was a point where 10x in power made VR feasible. We're past that point, and can already do 144Hz VR with pretty good graphics. More power would increase the resolution but that's not a new use case. Deep learning, hmm. A 10x increase in GPU RAM would be pretty great for using things like GPT-3, but the limiting factor there isn't really processing power. Remember that you said "sorely need". That's a much stronger statement than benefiting. > Why would no new use case arise from a widely available 10-fold increase in processing power when the last 40 years have shown that new tech always materializes? Well, I can look back about a decade for the last 10-fold increase. Everyone has an SSD now, that's major but not based on processing power. VR is the only significant new use case I can name. That's cool but it's not exactly a big impressive list.
    - shock-value 5y ago
      The equivalent of that (or more) is needed for a good VR experience.
    - rowanG077 5y ago
      In that sense no it's not needed. In fact computers aren't needed at all. We made due without them for thousands of years.
- thrashh 5y ago
  Imagine being able to have the top of the line graphics performance that you can get now but on a chip the size of your fingernail in the future and the kind of industries that would be created once that became possible.
- coolspot 5y ago
  Lightweight AR glasses that can last whole day on a 1000mAh battery.
melff 5y ago
I think the opposite is true. Modern Hardware is more than fast enough for pretty much all end-user applications(except maybe high-end gaming, but that's not sorely needed), we're just too wasteful with all the performance we've already got.

by dsign 5y ago

throwaway4good 5y ago

Do we really? (Serious question - what are the end-user applications that cannot be done today because of the lack of faster matrix multiplication?)

pfortuny 5y ago
VR headsets, games running at 4K 120fps, etc… Anything using massive linear algebra.
- throwaway4good 5y ago
  4K 120fps - is that "sorely needed"?
  - belval 5y ago
    Of course not and "640K software is all the memory anybody would ever need on a computer".
    - dmitriid 5y ago
      False equivalence
    - throwaway4good 5y ago
      No. I am asking if there are genuine new use cases that will be enabled by say a ten-fold increase in processing power? Other than it is nice going from 2K at 30 Hz to 4K at 120 Hz (which is an 8-fold increase).
      - belval 5y ago
        Your question is difficult to answer in good faith because "everything" would benefit from a ten-fold increase in processing power. Super computers could process protein foldings faster, any kind of simulation really could benefit from it. Deep learning could eat a ten-fold increase overnight (especially if accompanied by faster buses and memory) gaming at 30Hz is really not great, gaming at 60/120/144Hz is much better. VR was already mentioned in this very thread. I answered your question with a joke because your question is a joke, what doesn't benefit from a 10-fold increase in processing power? Why would no new use case arise from a widely available 10-fold increase in processing power when the last 40 years have shown that new tech always materializes? Or maybe you just wanted to make a cynical statement, that all this tech from the last 20 years was pointless, that social media are a net negative for the world, that better video game graphics are pointless because only gameplay matter, that smartphones (only possible because of a previous wave of 10-fold increases) are only addictive little screens and not actually useful in everyday life? Yes there are genuine new use cases that would appear with a ten-fold increase in processing power. It's a certainty.
        Dylan16807 5y ago
        > I answered your question with a joke because your question is a joke, what doesn't benefit from a 10-fold increase in processing power? You misread the question pretty badly. They asked for new use cases, not things that would benefit. We can fold proteins pretty well now, 10x wouldn't fundamentally change things. There was a point where 10x in power made VR feasible. We're past that point, and can already do 144Hz VR with pretty good graphics. More power would increase the resolution but that's not a new use case. Deep learning, hmm. A 10x increase in GPU RAM would be pretty great for using things like GPT-3, but the limiting factor there isn't really processing power. Remember that you said "sorely need". That's a much stronger statement than benefiting. > Why would no new use case arise from a widely available 10-fold increase in processing power when the last 40 years have shown that new tech always materializes? Well, I can look back about a decade for the last 10-fold increase. Everyone has an SSD now, that's major but not based on processing power. VR is the only significant new use case I can name. That's cool but it's not exactly a big impressive list.
  - shock-value 5y ago
    The equivalent of that (or more) is needed for a good VR experience.
  - rowanG077 5y ago
    In that sense no it's not needed. In fact computers aren't needed at all. We made due without them for thousands of years.
thrashh 5y ago
Imagine being able to have the top of the line graphics performance that you can get now but on a chip the size of your fingernail in the future and the kind of industries that would be created once that became possible.
coolspot 5y ago
Lightweight AR glasses that can last whole day on a 1000mAh battery.

melff 5y ago

I think the opposite is true. Modern Hardware is more than fast enough for pretty much all end-user applications(except maybe high-end gaming, but that's not sorely needed), we're just too wasteful with all the performance we've already got.