| ▲ | radarsat1 5 days ago | ||||||||||||||||
Why haven't Nvidia developed a TPU yet? | |||||||||||||||||
| ▲ | dist-epoch 5 days ago | parent | next [-] | ||||||||||||||||
This article suggests they sort of did: 90% of the flops is in matrix multiplication units. They leave some performance on the table, but they gain flexible compilers. | |||||||||||||||||
| ▲ | Philpax 5 days ago | parent | prev | next [-] | ||||||||||||||||
They don't need to. Their hardware and programming model are already dominant, and TPUs are harder to program for. | |||||||||||||||||
| ▲ | HarHarVeryFunny 5 days ago | parent | prev [-] | ||||||||||||||||
Meaning what? Something less flexible? Less CUDA cores and more Tensor Cores? The majority of NVidia's profits (almost 90%) do come from data center, most of which is going to be neural net acceleration, and I'd have to assume that they have optimized their data center products to maximize performance for typical customer workloads. I'm sure that Microsoft would provide feedback to Nvidia if they felt changes were needed to better compete with Google in the cloud compute market. | |||||||||||||||||
| |||||||||||||||||