## When will computers be faster than brains?

Saw this:

By most estimates, there are 100 billion neurons in the brain. Some neurons are known to have more than 1,000 dendrites, and up to about 1,000 different branchings of their axons. There are some 50 known neurotransmitters, and who knows how many other neuromodulators may exist (hormones, neural growth factors, neurosteroids). There are also many different receptor types for each of the neurotransmitters. A conservative estimate of the number of interactions you’d have to model to be biologically accurate is somewhere around 225,000,000,000,000,000 (225 million billion).

On average there can be about 100 action potentials - or nerve impulses per second.

This site has an interesting way to put it:

A human being has about 100 billion brain cells. Although different neurons fire at different speeds, as a rough estimate it is reasonable to estimate that a neuron can fire about once every 5 milliseconds, or about 200 times a second. The number of cells each neuron is connected to also varies, but as a rough estimate it is reasonable to say that each neuron connects to 1000 other neurons- so every time a neuron fires, about 1000 other neurons get information about that firing. If we multiply all this out we get 100 billion neurons X 200 firings per second X 1000 connections per firing = 20 million billion calculations per second.

This site adds:

It takes robot vision programs about 100 computer instructions to derive single edge or motion detections from comparable video images. 100 million instructions are needed to do a million detections, and 1,000 MIPS to repeat them ten times per second to match the retina.

Computer chess bolsters this yardstick. Deep Blue, the chess machine that bested world chess champion Garry Kasparov in 1997, used specialized chips to process chess moves at a the speed equivalent to a 3 million MIPS universal computer (see Figure 3-4). This is 1/30 of the estimate for total human performance. Since it is plausible that Kasparov, probably the best human player ever, can apply his brainpower to the strange problems of chess with an efficiency of 1/30, Deep Blue’s near parity with Kasparov’s chess skill supports the retina-based extrapolation.

The 1,500 cubic centimeter human brain is about 100,000 times as large as the retina, suggesting that matching overall human behavior will take about 100 million MIPS of computer power.

This assumes that all vision calculation is done in the retina which is not really a fair assessment.

So each neuron can gauge 50+ neurotransmitters as well as neuro-modulators - and each can finely measure those quantities. So I suppose we can call that one calculation - although that may well be called more than one calculation depending on what computers are capable of.

So mips = 1 million instructions per second. Flops is perhaps a better measure - FLops = floating point operations per second - any type of mathematical calculation. To make the calculations conservative, let’s assume there are 100 neurotransmitters and neuro-such that each neuron requires 100 floating point calculations in order to fire. So if there are 20 million billion impulses, each requiring 100 floating point calculations, we have 2 billion billion - 10^18 or about 2 exaflops.

As of 2010, intel is at about 100 Gigaflops. If computing power continues to to double every year, it’s about 10 years per 1000 fold leap, meaning we need about 25 years to get to brain speed. Of course it will take longer if progress is a bit slower - and it could come more quickly as this estimate is conservative in some ways.

Other people have their opinions. Here’s Kurzweil who says we get there in 2020, but don’t have the software to take hold of all this until 2030.

Source: develintel.blogspot.com

## 50 Notes/ Hide

- m-elowz likes this
- broadenme posted this
- Show more notes