4 ms·

These are steady, predictable, iterative improvements. We have nothing groundbreaking like 1970s-1990s. Of course I'm not saying modern hardware is slow, but it

These are steady, predictable, iterative improvements. We have nothing groundbreaking like 1970s-1990s. Of course I'm not saying modern hardware is slow, but it's not progressing like it used to.

CyberDildonics 6y ago
That is both not true and not what you said at first. Transistor density has continued to rise but there is only so much you can do when people mostly run javascript on a single thread. There were a lot more break throughs in the early days of electronic components too because everything was new. CPUs are a world away from where they were two decades ago, but people don't notice because typing into a facebook window still lags. It is an incredibly superficial nonsense way to look at what has really happened at the hardware level.
- astrange 6y ago
  If you only have a single core (common on battery powered devices) then you should only run a single thread. Excess parallelism doesn't make things go faster and uses more memory. The important thing is avoiding unnecessary waits on non-CPU resources like I/O.
  - CyberDildonics 6y ago
    This reads like a reply to a different comment. The person above was saying there haven't been any advancements in CPUs or GPUs which is obviously not true. No one said anything about battery powered devices and even those have had multiple cores for many years now - as a result of transistors still shrinking.
    - astrange 6y ago
      They're multicore devices but often only one core is free. Depending on how important you are, the rest are busy doing other things and you're not going to get scheduled.
      - CyberDildonics 6y ago
        This thread was someone saying there haven't been any advances in CPUs and GPUs in general. You seem to be in a completely different goal post shifting context of some mythical mobile device that has only one core, but actually has multiple cores, but all them are pegged at 100% all the time. This doesn't make any sense in any context since it isn't even true, but it has nothing to do with what you are replying to.

These are steady, predictable, iterative improvements. We have nothing groundbreaking like 1970s-1990s. Of course I'm not saying modern hardware is slow, but it's not progressing like it used to.

CyberDildonics 6y ago

That is both not true and not what you said at first. Transistor density has continued to rise but there is only so much you can do when people mostly run javascript on a single thread. There were a lot more break throughs in the early days of electronic components too because everything was new. CPUs are a world away from where they were two decades ago, but people don't notice because typing into a facebook window still lags. It is an incredibly superficial nonsense way to look at what has really happened at the hardware level.

astrange 6y ago
If you only have a single core (common on battery powered devices) then you should only run a single thread. Excess parallelism doesn't make things go faster and uses more memory. The important thing is avoiding unnecessary waits on non-CPU resources like I/O.
- CyberDildonics 6y ago
  This reads like a reply to a different comment. The person above was saying there haven't been any advancements in CPUs or GPUs which is obviously not true. No one said anything about battery powered devices and even those have had multiple cores for many years now - as a result of transistors still shrinking.
  - astrange 6y ago
    They're multicore devices but often only one core is free. Depending on how important you are, the rest are busy doing other things and you're not going to get scheduled.
    - CyberDildonics 6y ago
      This thread was someone saying there haven't been any advances in CPUs and GPUs in general. You seem to be in a completely different goal post shifting context of some mythical mobile device that has only one core, but actually has multiple cores, but all them are pegged at 100% all the time. This doesn't make any sense in any context since it isn't even true, but it has nothing to do with what you are replying to.