May be if you use 100+ of these (~14400 'computers') you can build a machine wit...

sliverstorm · on Nov 23, 2011

If 15,000 computers was all you needed for AI, we'd have it by now. Major datacenters have way more machines than that, which are far more powerful than a milliwatt microcontroller to boot.

ypcx · on Nov 22, 2011

I knew it. The name "Moore" sounded suspicious to me immediately. Now I know his real name is Chuck Testa!!!

On a more serious note, what we need much more of is not the processing speed. What we need much more of is what I call "memory processability", which roughly means "how many times per second can you process the whole memory" - or something like that. Basically how much CPU is there per RAM. Indexing is a great hack, but a hack it still is. Memory and processing cells must be merged into a single, massively parallel chip.

thisrod · on Nov 22, 2011

That was done in the 80's: http://en.wikipedia.org/wiki/Connection_Machine . Richard Feynman had to help the designers with the hard bits.

Users recall them fondly. In the end, clusters became too cheap for custom hardware like this to compete.

itsameta4 · on Nov 23, 2011

>What we need much more of is what I call "memory processability", which roughly means "how many times per second can you process the whole memory"

I might be misinterpreting, but I think the phrase you're looking for is "memory bandwidth".