General > General Technical Chat
Whats the clock speed/flops per second of our brains?
CatalinaWOW:
--- Quote from: BitsnBytes on February 13, 2021, 11:58:21 am ---
--- Quote ---If our brains were computers what would the clock speed be?
--- End quote ---
I think the time taken by nerve signals would be the quantitative representation of brain speed.
Regards
--- End quote ---
Trying to reduce computer or brain speeds to a single number is going to lead to confusion and arguments.
Nerve speed or other measures can give a rough idea of clocking rate, or minimum response time or something like that. But these measures give answers in the five to ten Hertz range. If that were the whole answer no one would be bothered by screen or fluorescent light flicker in the 20-30 Hz range. So some processes operate somewhat faster and some are slower.
But clock rate isn't the be all, end all answer for speed, for either computers or people. Supercomputers don't achieve their multi-Teraflop rates through multi-teraflop clock rates, they are also "massively" parallel. I put the "massively" in quotes. The term is widely used in describing supercomputers. But brains are many orders of magnitude more parallel than supercomputers.
It would be nice to compare brain and computer capabilities for many reasons. One of the most compelling is to estimate when they might take over from us. But we really don't even have the tools yet to properly ask the question, let alone find the answers.
james_s:
I don't think clock speed is really a useful concept when discussing the brain. It is not a digital computer, it is not synchronous, it has no clock, it takes time for chemical and electrical signals to travel around but that's more of a propagation and response time than a clock. Analog computers didn't have a clock either.
helius:
The brain is more synchronous than you might expect. Instead of a clock tree it uses timed feedback loops, but that still keeps all the parts synchronized. Of course, the type of synchronization is not the D-flipflop kind where you sample an input at a precise time. It is more a matter of having a billion voltage-to-frequency converters that pull each other into quasistable regimes.
cgroen:
--- Quote from: Beamin on February 10, 2021, 01:45:09 am ---.
.
I think my brain has an 8086 ( at 3.4 MHz, the switch to 7 is broken) in it with a broken math co-processor, and 64k ram; 60 of which is used up to keep me breathing and standing upright. And the real time clock battery is leaking.
--- End quote ---
I think you are very correct...
NiHaoMike:
--- Quote from: helius on February 13, 2021, 10:30:33 pm ---The brain is more synchronous than you might expect. Instead of a clock tree it uses timed feedback loops, but that still keeps all the parts synchronized. Of course, the type of synchronization is not the D-flipflop kind where you sample an input at a precise time. It is more a matter of having a billion voltage-to-frequency converters that pull each other into quasistable regimes.
--- End quote ---
Surprisingly similar to the time when someone tried to see if evolution could build a useful circuit on a FPGA.
http://www.netscrap.com/netscrap_detail.cfm?scrap_id=73
--- Quote from: Beamin on February 10, 2021, 01:45:09 am ---I think my brain has an 8086 ( at 3.4 MHz, the switch to 7 is broken) in it with a broken math co-processor, and 64k ram; 60 of which is used up to keep me breathing and standing upright. And the real time clock battery is leaking.
--- End quote ---
The software side must be very impressive (relatively speaking) to do what it can with so little. Kind of like the Amigas that were known for being so capable with comparatively little hardware.
Navigation
[0] Message Index
[*] Previous page
Go to full version