What is B200?

Halving bit precision should yield more than double the FLOP count due to quadratic scaling in matrix multiply hardware, meaning the historical 2x rule understates the true gain.

“Historically, up until B100 or B200, every time you halved the bit precision, you doubled the FLOP count. For the reason you said, because of this quadratic scaling, that ratio is actually slightly wrong. You should get an even bigger speedup than you might otherwise think.”

Space-based data centers (per Elon Musk's proposal) would require roughly $100 per B200 GPU-hour to be financially viable, versus $2 to $3 spot pricing on Earth.

“Elon is focused on moving the chips to moving the data centers to space and I think on the numbers that he has I think they are looking at like a $100 an hour per GPU hour for a B200 for it to make sense.”

Philip Johnston on StarCloud's unique radiation-failure data for flagship NVIDIA GPUs.

“I think we're the only people in the world now that know where both an H100, a B200, H200 will fail if you blast it with high velocity protons and heavy ions.”

Kernel-level optimization yields only single-digit percentage gains on H100s (already heavily optimized) but 20 to 40 percent improvements on B40/B200, and 2x or more on Mac, AMD, and Intel hardware.

“If you start looking at think is like a B40, or a B200, you'll start seeing like much more significant gains to be had over there. you know, we have seen 20 30 40% improvements in performance in many cases.”

Jack Farley on GPU scarcity: B200 availability has already collapsed, H100s are close behind.

“GPU availability for B200s has collapsed H100s close behind.”