[jiehong]

> -benchmarked Radeon VII and RTX 3080, shows that FFT is extremely bandwidth limited on modern GPUs

Great to see that!

I expect huge improvements in that area with AMD's new RX series with SAM activated [0].

[0]: https://www.amd.com/en/technologies/smart-access-memory

[DTolm]

Actually, it is still best to aim at zero transfers between GPU and CPU during the execution. The GPU is limited by VRAM-chip bandwidth which is much bigger than the PCI-E bandwidth. And it should not be affected by SAM.