Next gen GPU with 512MB of SRAM

Mining
1

Quoting from https://www.tweaktown.com/news/80740/amds-next-gen-rdna-3-gpu-up-to-2-5x-faster-than-geforce-rtx-3080-ti

“Infinity Cache is expected to be greatly improved, with AMD’s next-gen Navi 31-based Radeon GPU packing up to 512MB of Infinity Cache… this is a 4x increase over the 128MB of Infinity Cache on the Radeon RX 6800 XT and Radeon RX 6900 XT graphics cards under the RDNA 2 GPU architecture.”

That has the potential to run a single pass lean miner. It still might not be very efficient if changing a single bit in cache requires writing a long cache line.

Note that the Ipollo G1 ASIC has only 64MB of SRAM and thus needs to make 8 passes in every edge trimming round, limiting its efficiency.

12 Likes
2

If a 6800/6800 XT is doing 0.9-1 GPS @150w, then I wonder what a Navi31-based Radeon could achieve?

It seems like next year we could end up with a scenario where C32 is more efficient on high-end GPUs vs current Gen ASICs.

5 Likes