5 ms·

I laughed at this quote (tl;dr) from the article: "about a 4K movie every two seconds". Hope that doesn't qualify as a spoiler.

noir_lord 7y ago
If his answer is on the right order magnitude it's kinda interesting that our signalling technology is getting up there in bandwidth.
Ididntdothis 7y ago
Latency is quite high though as far as I know.
- mhh__ 7y ago
  Speaking of latency, one of the most bizarre experiences I can think of is when (on a certain cough substance beginning whose name is a type of broadband spelt backwards) one's reaction times double to about 500-600 ms and trying to do a reaction test. Really fascinating but also hilarious to watch someone else try.
  - lagadu 7y ago
    Interesting, do you have a source for that? I'd love to read it. I'm interested because my experience with that differs strongly (though it's with shrooms, not LSD): while tripping on low doses I was a god* at guitar hero, significantly better than I was at playing sober, the same was true for another friend of mine as well. Maybe the task being rhythm-related offsets the increased delay. *sober I mastered Hard but had some trouble on the harder songs on expert, under the influence was the first time I managed to do several of them in expert.
    - mhh__ 7y ago
      I don't have a source other than a ... friends experience (wink wink, "which means I am lying"). It's the only area of medicine I'd do research in, Psychedelics seem to have dramatically different effects while still having some constants - across different people. I could type 10 WPM faster than I usually could but my companion couldn't get above 15 WPM (He's usually bad not that bad). Once you get going it seems to be manageable, as if you have thought-momentum, i.e. Maybe it's a movement issue rather than literally reaction time directly. > Maybe the task being rhythm-related offsets the increased delay. I think that may have an effect. I can play actual guitar fairly well (I can still play pretty fast Yngwie stuff) under the influence but only once I've "warmed up" but then I'm effectively just reciting motion from memory rather than reacting to anything There are papers/research on this but they're all behind paywalls, which I can't be bothered to get around rn.
      - lagadu 7y ago
        The feeling of playing was very different; it felt like my hands were playing the controller by themselves without any real input or reactions on part of "me".
- gingabriska 7y ago
  But I read somewhere that the brain is able to adjust heart beats and breathing at a very fine frequency and and keeping it in sync requires good latency and all this is done without concurrency locks. So does brain have massively parallel architecture where key things have their own core?
  - inciampati 7y ago
    In biology every single protein (or other biochemical entity) can be thought of as it's own core. It's parallel in an amazing and beautiful way. Or the most boring way possible. It just depends on if you are a computer scientist or biologist how you would see it.
    - bdamm 7y ago
      There are to this day computer science researchers having conniptions about the impossibility of massively parallel lock-free architectures based on eventual consistency. Yet literally all they must do is open their eyes and see the truth!
      - jplayer01 7y ago
        It helps that our brains and bodies are analog. It doesn't matter if a particular variable is changed by different factors at the same time. The total magnitude of all the changes is far more important. Thus, lock-free.
        farazbabar 7y ago
        Kind of like CRDT (conflict free replicated data types for the biologists amongst us).
    - whatshisface 7y ago
      Protiens? Every particle in every atom is solving exponential-time problems constantly, in parallel. That's the quantum chemistry perspective. :)
alevskaya 7y ago
This is an excessive overestimate by not carefully considering the known types of afferent (incoming) fibers and their average firing statistics (most neurons fire sparsely). See the comment on that thread by "JohnShaft" who noodles a better estimate of around 200MB/s (incoming) by taking anatomy into account. The motor efferent signals are going to be a lot lower bandwidth still...

noir_lord 7y ago

If his answer is on the right order magnitude it's kinda interesting that our signalling technology is getting up there in bandwidth.

Ididntdothis 7y ago

Latency is quite high though as far as I know.

mhh__ 7y ago
Speaking of latency, one of the most bizarre experiences I can think of is when (on a certain cough substance beginning whose name is a type of broadband spelt backwards) one's reaction times double to about 500-600 ms and trying to do a reaction test. Really fascinating but also hilarious to watch someone else try.
- lagadu 7y ago
  Interesting, do you have a source for that? I'd love to read it. I'm interested because my experience with that differs strongly (though it's with shrooms, not LSD): while tripping on low doses I was a god* at guitar hero, significantly better than I was at playing sober, the same was true for another friend of mine as well. Maybe the task being rhythm-related offsets the increased delay. *sober I mastered Hard but had some trouble on the harder songs on expert, under the influence was the first time I managed to do several of them in expert.
  - mhh__ 7y ago
    I don't have a source other than a ... friends experience (wink wink, "which means I am lying"). It's the only area of medicine I'd do research in, Psychedelics seem to have dramatically different effects while still having some constants - across different people. I could type 10 WPM faster than I usually could but my companion couldn't get above 15 WPM (He's usually bad not that bad). Once you get going it seems to be manageable, as if you have thought-momentum, i.e. Maybe it's a movement issue rather than literally reaction time directly. > Maybe the task being rhythm-related offsets the increased delay. I think that may have an effect. I can play actual guitar fairly well (I can still play pretty fast Yngwie stuff) under the influence but only once I've "warmed up" but then I'm effectively just reciting motion from memory rather than reacting to anything There are papers/research on this but they're all behind paywalls, which I can't be bothered to get around rn.
    - lagadu 7y ago
      The feeling of playing was very different; it felt like my hands were playing the controller by themselves without any real input or reactions on part of "me".
gingabriska 7y ago
But I read somewhere that the brain is able to adjust heart beats and breathing at a very fine frequency and and keeping it in sync requires good latency and all this is done without concurrency locks. So does brain have massively parallel architecture where key things have their own core?
- inciampati 7y ago
  In biology every single protein (or other biochemical entity) can be thought of as it's own core. It's parallel in an amazing and beautiful way. Or the most boring way possible. It just depends on if you are a computer scientist or biologist how you would see it.
  - bdamm 7y ago
    There are to this day computer science researchers having conniptions about the impossibility of massively parallel lock-free architectures based on eventual consistency. Yet literally all they must do is open their eyes and see the truth!
    - jplayer01 7y ago
      It helps that our brains and bodies are analog. It doesn't matter if a particular variable is changed by different factors at the same time. The total magnitude of all the changes is far more important. Thus, lock-free.
      - farazbabar 7y ago
        Kind of like CRDT (conflict free replicated data types for the biologists amongst us).
  - whatshisface 7y ago
    Protiens? Every particle in every atom is solving exponential-time problems constantly, in parallel. That's the quantum chemistry perspective. :)

alevskaya 7y ago

This is an excessive overestimate by not carefully considering the known types of afferent (incoming) fibers and their average firing statistics (most neurons fire sparsely). See the comment on that thread by "JohnShaft" who noodles a better estimate of around 200MB/s (incoming) by taking anatomy into account. The motor efferent signals are going to be a lot lower bandwidth still...