|
|
|
|
|
|
by microtonal
1461 days ago
|
|
|
The opposite has been true in desktops. I have a Ryzen 3900X box and there still isn’t anything from Apple that I would replace it with. Not even to run macOS on it (which I do on the AMD box, using GPU passthrough). My MacBook Pro 14" is pretty much on-par with my Ryzen 5900X CPU-wise (the M1 Ultra would surpass it by a wide margin). The GPU of the M1 Max is nothing to sneeze at, but I'd love it if they'd bring eGPU support to the M1 line. (And if one can dream, if NVIDIA would also make CUDA available.) |
|
|
Well, in which tasks?
Comparing e.g. geekbench results for the M1 Ultra https://browser.geekbench.com/v5/cpu/14664498 and a decent 5950X (PBO) result https://browser.geekbench.com/v5/cpu/14665931 it looks like a lot of Apple's multicore score is attributable to AES (which uses special fixed-function HW). And with other tests, different CPUs win different benchmarks. 5950X notably wins in clang :)
> bring eGPU support to the M1 line
Probably not M1.
marcan tweeted recently that Apple's PCIe integration has been is broken in the same way as on Broadcom's RPi4 SoC — mapping PCIe BARs as normal memory (which allows unaligned access) doesn't work. I can't find the tweet (deleted??) but here's the RPi thread: https://github.com/geerlingguy/raspberry-pi-pcie-devices/iss... Basically there's no quick kernel level workaround for this, if you really want a GPU to work on such a broken platform you need to patch every single thing in userspace to avoid unaligned access.
We'll see soon if that's fixed in M2, but I suspect they don't care…