From a keynote perspective, I guess showing software optimizations is less cinematic and not everyone can appreciate them as opposed to new or beautiful looking hardware.
they are mostly likely using a higher density battery in the air, at least that's what the rumors suggest
Raw gemm computation was never the real bottleneck, especially on the newer GPUs. Feeding the matmuls i.e memory bandwidth is where it’s at, especially in the newer GPUs.
Why should it matter? They can't link it to me since they don't have my genetic profile.
From a keynote perspective, I guess showing software optimizations is less cinematic and not everyone can appreciate them as opposed to new or beautiful looking hardware.
they are mostly likely using a higher density battery in the air, at least that's what the rumors suggest
Raw gemm computation was never the real bottleneck, especially on the newer GPUs. Feeding the matmuls i.e memory bandwidth is where it’s at, especially in the newer GPUs.
Why should it matter? They can't link it to me since they don't have my genetic profile.