• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
  • The Politics forum has been nuked. Please do not bring political discussion to the rest of the site, or you will be removed. Thanks.

Simulating one second of real brain activity : 40 minutes, 83K processors, 1PB of RAM

Status
Not open for further replies.

DieH@rd

Banned
Dec 9, 2006
35,045
4
0
Goddamn, one petabyte of ram.... thats ONE MILLION gigabytes.

Yeah, this can run Crysis. :D
 

Zaptruder

Banned
Jun 7, 2004
27,818
3
0
Because your brain sucks at it. Most of your brain is specialist hardware for doing one thing or another. It's fantastic at language and facial recognition but abysmal at mathematical operations. High level reasoning, planning, maths etc are not as computationally intensive as you might think, while getting something to stand upright or parse a sentence is much harder than you might think. Yet we are designed to do the latter, so it seems trivial to us, while we are not designed to do the former so it seems really hard or complicated. This is known as "Moravec's paradox".

Hmmm... the idea that the brain has 'specialist hardware' is a misnomer I feel. It characterizes the brain as having these prebuilt innate areas like a visual area, a language area, a face area, etc, etc.

The reality is that the neocortex is more like a cloud computing center. General purpose massively parallel computing network.

The reason that humans so consistently develop visual centers, language, centers, facial recognition areas, is because overall, the brain is wired up very consistently to its sensory modules and the various connections to the body (similar to how the make up of the body itself is very consistent from human to human). And that the environment that it interfaces with is very consistent as well (certainly has many more fine grained similarities than we'd intuitively feel - access to family, access to language, access to edges of high contrast and low contrast and colours, etc).

With that said, as remarkable as the brain is, it's not an idealized system for processing all types of information.

Especially information that requires massive operations and iteration - it's far better in processing patterns that can be heirachically catergorized.

As a result, it does things like facial recognition well enough... but iterating in mathematical operations poorly.

Additionally, mathematics tends to require abstraction that ends up at the higher levels of the heirachy... and for most people require way more stacks of the neural pyramid (and thus processing power) than is optimal for the problem.

But remarkably, even numbers and maths can be somehow catergorized for efficient sorting and traversal by this neural structure - as evidenced by savants and other high functioning autistics and people with conditions that nonetheless express a remarkable degree of intelligence in some field.

But despite this, I have a feeling that the brain is still not an efficient number crunching system; even in brains that have through whatever reason and circumstance been wired up for such a task... at least not for standard mathematical operations.
 

AndyD

aka andydumi
Jan 24, 2007
19,098
0
0
Nashville, USA
...and here we are on Neogaf making fantastic use of our built-in super compluters.

For some reason I laughed at this. Maybe because of the intentional mis-spelling.

As to OP, I wonder how much of our brain function is dedicated/spent on internal/staying alive work (regulating various organs) and how much to processing "other" stuff like say watching TV.
 
Jul 18, 2008
14,274
0
0
37
Philadelphia
Someone who has more freetime than I have should calculate about how long it would take for the growth in RAM and computing capacity to reach the point where such calculations can be done in real time, or at least near real time. They hint at a decade in the article.
 

heidern

Junior Member
Jun 7, 2004
2,288
0
1,375
The reason that humans so consistently develop visual centers, language, centers, facial recognition areas, is because overall, the brain is wired up very consistently to its sensory modules and the various connections to the body (similar to how the make up of the body itself is very consistent from human to human).

Does this mean for example a blind person would not develop a visual centre and instead those brain cells would develop to be used for something else(such as an auditory centre)?
 

Zaptruder

Banned
Jun 7, 2004
27,818
3
0
Does this mean for example a blind person would not develop a visual centre and instead those brain cells would develop to be used for something else(such as an auditory centre)?

Absolutely. If there's no visual sensory information to create a visual information processing area, it allows those neurons to be dedicated to other functions - and would result in a rearrangment of typical brain areas as a result.
 
Status
Not open for further replies.