2 ms·
Wouldn’t most of the compute be on the GPU anyway? So the specifics of the ARM instruction set implemented on the CPU cores don’t matter very much. Maybe the M
by pavlov 3y ago
Wouldn’t most of the compute be on the GPU anyway? So the specifics of the ARM instruction set implemented on the CPU cores don’t matter very much.
Maybe the M2 GPU is also better at bfloat16, but the author admits to not having researched this. So I wouldn’t make any far-reaching conclusions based on this one difference in the CPU instruction set.
- miohtama 3y agoMacs have Neural Engine, a separate chip from GPU, for AI workloads https://www.apple.com/newsroom/2023/10/apple-unveils-m3-m3-pro-and-m3-max-the-most-advanced-chips-for-a-personal-computer/ https://www.apple.com/newsroom/2023/10/apple-unveils-m3-m3-p...
- adrian_b 3y agobfloat16 is not used for graphics, where FP16 is appropriate, so all GPUs that are new enough to support 16-bit floating-point numbers support FP16. For bfloat16, there exists only one application, ML/AI, because it is the only application that can tolerate a so low resolution in numbers. AFAIK, the first device that has provided hardware support for bfloat16 was the Intel Cooper Lake CPU, which was launched in Q2 2020. AMD has added bfloat16 support in Zen 4, in 2022. If any GPU has added hardware support for bfloat16, it must be a model launched in 2020 or later. Support for bfloat16 is much more likely to be provided by dedicated ML/AI accelerators.
- corsix 3y agoSome compute might be on the AMX units (dedicated matrix multiplication coprocessor, closely attached to the CPU, distinct from both ANE and GPU). They gained bf16 support in M2.