Join the discussion
Write your take first — we'll ask for email only when you're ready to publish.
- Hacker News
- surely someone in apple would have run some common neural net architectures through the rtl simulator before tapeout and checked that the performance meets expectations??
- I guess any processes affected by this will solve this in software as time goes on?by msephton
- This goes way over my head and I don't understand most of it lol. I noticed you're still in the middle of getting your B.S. degree and you're already writing things like this...amazing.
- This is kinda nuts.
Can this mostly be fixed in software, or are M3 users basically stuck unless apps work around it themselves?
Going from 10 to 24 tok/s just from a workaround is a pretty crazy jump.
by imnotr0b0t - Just checking here- this systemverilog is a hypothetical telling of what you think is going on? Or do you have the actual source of the RTL?by Neywiny
- Not all systems affected M1 and M5MAX are OKby anemll
- "Apparently the memory controller's throughput has a dominant harmonic with wavelength 2048 in tensor-dimension space."
That got a laugh out of me.
by thenewwazoo - Nice investigation.
It is always surprising to me when a nice round number like 1MiB results in the “bad performance” configuration (although it happens).
Are you sure erratum is the right word in this context? I usually see it used to describe the notice that a document has an error in it.
by bee_rider