Live data from Hacker News

An inside look at the custom CPUs in Tesla's Dojo Supercomputer

semianalysis.com

1–10 of 132 posts

Re: An inside look at the custom CPUs in Tesla's Dojo Supercomputer

#2
Now assume that Moore's law stays healthy and in a few decades we're carrying around dojo-equivalent smart phones. Or implants. What do our apps do with all that compute? Detailed convincing AR, accurate real-time translation, identify almost any object in sight, turing-test passing answering machine. We'll be carrying around neural-net training hardware that is different than the stuff between our ears, often worse but sometimes much better. A cyber brain. Scary af.

Re: An inside look at the custom CPUs in Tesla's Dojo Supercomputer

#3
post #2

Now assume that Moore's law stays healthy and in a few decades we're carrying around dojo-equivalent smart phones. Or implants. What do our apps do with all that compute? Detailed convincing AR, accurate real-time translation, identify almost any object in sight, turing-test passing answering machine. We'll be carrying around neural-net training hardware that is different than the stuff between our ears, often worse…

I wish I had your optimism. All I expect are better ways to deliver targeted advertising.

Re: An inside look at the custom CPUs in Tesla's Dojo Supercomputer

#4
post #2

Now assume that Moore's law stays healthy and in a few decades we're carrying around dojo-equivalent smart phones. Or implants. What do our apps do with all that compute? Detailed convincing AR, accurate real-time translation, identify almost any object in sight, turing-test passing answering machine. We'll be carrying around neural-net training hardware that is different than the stuff between our ears, often worse…

We will have 7 more layers of UI abstraction to solve the problem of not adequately warming the users hands as they use it during extreme winter weather caused by climate change.

Re: An inside look at the custom CPUs in Tesla's Dojo Supercomputer

#6
post #2

Now assume that Moore's law stays healthy and in a few decades we're carrying around dojo-equivalent smart phones. Or implants. What do our apps do with all that compute? Detailed convincing AR, accurate real-time translation, identify almost any object in sight, turing-test passing answering machine. We'll be carrying around neural-net training hardware that is different than the stuff between our ears, often worse…

Performance per watt is not improving that much, nor is feature size. You'll probably never be carrying around a supercomputer with this many FLOPS.

Re: An inside look at the custom CPUs in Tesla's Dojo Supercomputer

#7
post #2

Now assume that Moore's law stays healthy and in a few decades we're carrying around dojo-equivalent smart phones. Or implants. What do our apps do with all that compute? Detailed convincing AR, accurate real-time translation, identify almost any object in sight, turing-test passing answering machine. We'll be carrying around neural-net training hardware that is different than the stuff between our ears, often worse…

Moore’s law is quite dead. Transistor density isn’t doubling. We’re still getting speed increases due to architectural improvements, but transistor density is capping out because thermals are a problem. That’s why you see these massive wafer designs and they’re not shrinking.

Re: An inside look at the custom CPUs in Tesla's Dojo Supercomputer

#8
post #2

Now assume that Moore's law stays healthy and in a few decades we're carrying around dojo-equivalent smart phones. Or implants. What do our apps do with all that compute? Detailed convincing AR, accurate real-time translation, identify almost any object in sight, turing-test passing answering machine. We'll be carrying around neural-net training hardware that is different than the stuff between our ears, often worse…

That's not a world I want to live in. Not even a little bit.

Re: An inside look at the custom CPUs in Tesla's Dojo Supercomputer

#10
post #2

Now assume that Moore's law stays healthy and in a few decades we're carrying around dojo-equivalent smart phones. Or implants. What do our apps do with all that compute? Detailed convincing AR, accurate real-time translation, identify almost any object in sight, turing-test passing answering machine. We'll be carrying around neural-net training hardware that is different than the stuff between our ears, often worse…

Moore’s law is quite dead. Transistor density isn’t doubling. We’re still getting speed increases due to architectural improvements, but transistor density is capping out because thermals are a problem. That’s why you see these massive wafer designs and they’re not shrinking.

> Transistor density isn’t doubling.

It pretty much continues to double: https://en.m.wikipedia.org/wiki/Transistor_count#/media/File...

Post reply on HN