Viewing profile — zepmck

NVIDIA is showing real performance through using independent MLPerf benchmark.

Does this hide any problem with AMD Frontier?

Software developers, especially for certain type of components, e.g. compilers, are NOT easy to find out.

The main reasons of "monopoly" are long-term vision and enormous effort spent in software development. AMD is lacking behind because they have not yet seriously invested on their o…

The next challenge is software stack.

People pretending to do professional work with low budget/quality workstation. If you cannot afford to operate a 4090 according to specs, you cannot blame NVIDIA if things go wrong…

servethehome.com/arm-neoverse-v…

You blame CUDA because you have never played AMD + ROCm. Furthermore OpenCL is on dead-end street. With standard programming languages natively supporting CUDA, it has no sense to …

The most powerful and unfortunately unusable supercomputer of the world. AMD's approach to GPUs is on a failing track since its inception. The only software stack available is supe…

If you have GPUs, NVTabular outperforms most of the frameworks out there: https://github.com/NVIDIA/NVTabular

This not completely correct because while a carrier transport anything indescrminately, social networks manipulate and control information exchange, for instance giving more priori…

I think this work is quite ridiculous. As far as I could see, they trained the net with a batch size equal to 1.

In terms of PCI lanes efficiency, there is no competition between Intel and AMD. Intel is much ahead wrt AMD. Don't not be impressed about number of lanes available on the board.

When? Where? How can it be miles ahead if the hardware has not been released yet?

One-API is not completely open source. Support for Ponte Vecchio will be not released open source for many reasons.