Hacker News new | ask | show | jobs
by jakobov 822 days ago
They are claiming a 25x reduction in power consumption. That can't be right. Anyone understand where this number is coming from?
2 comments

Comes from here [1]. Basically 100 racks of H vs 8 racks of B.

I think there may be a typo though, I assume this also includes liquid-cooled vs air-cooled.

[1] https://nvdam.widen.net/s/xqt56dflgh/nvidia-blackwell-archit...

Did you read that in the linked article? I couldn’t find it. But maybe due to the better efficiency with regard to the performance boost (5x) and the ability to now use 27 trillion parameters versus 1.7 Trillion, one can presumably finish the same amount of work in 1/25th of the time and bam, reduction in power consumption. As you say, I’m skeptical the max power draw itself is 25x lower.
I think Jensen said something like needing 25x fewer GPUs (vs. A100) to get the same performance, which amounts to essentially the same thing.
It doesn't imply a full 25x reduction in power consumption though, that might "only" go down by 10x.