Qesa

People don't know/don't care that RDNA2 achieved near-parity to Ampere through a higher bill of materials on a more advanced node, and...

The production bottleneck for these is advanced packaging and HBM, not actual chip production. 5nm wafer throughput itself (wafer, not...

Why would you expect a N4 chip to have perf/xtor more in line with N6 than N5? All of which are on the same node. Notably, while the...

4070Ti Super is a cut down 379mm² GPU. And you think AMD is gonna actually match that in performance with just 240mm² on more or less...

H100 is already pretty much a bundle of matrix multiply units though. They can strip out the vector and high precision stuff - and...

This perceived risk of "evil intelligence" is just feeding our egotistic desire to play god (perhaps because of how powerless we feel...

Back in 2023 they disclosed Sony was 16% of their revenue in 2022, which came to $3.8B for the year. With Xbox as well, easily over a...

Do we actually want to load the CPU less? Generally in games the GPU is loaded 100% while the CPU has idle cores. And decompressing...

Sorry, not their white paper, their slide deck Specifically this one: https://regmedia.co.uk/2023/12/06/amd_mi300_wire_rates.jpg

So you're saying AMD are just lying in their white paper?

MI300 has 4.8 TB/s bidirectional bisection bandwidth, that's less than half B100

Assuming high level arch is largely unchanged compared to Hopper, it seems like a huge layout improvement. In a similarly sized die on...

I assume he's counting the 7900 XTX as 12288 cores rather than 6144. The issue isn't so much scheduling but register file bandwidth...

They're well overdue for an architecture overhaul. Hopefully that will yield some perf/mm^2 and perf/W benefits independent of node...

For RDNA3 to be an "outlier" it must land into a position which isn't characteristic of where AMD GPUs usually are in comparison to...