zepmck
Re:
on Thursday December 14, 2023 @05:17AM (#38639749)
NVIDIA is showing real performance through using independent MLPerf benchmark.
Re:Operation of IBM Summit system gets extended to 2024
on Wednesday September 20, 2023 @11:21AM (#37585344)
Does this hide any problem with AMD Frontier?
Re:Nvidia's CUDA Monopoly
on Monday August 07, 2023 @04:22AM (#37031797)
Software developers, especially for certain type of components, e.g. compilers, are NOT easy to find out.
Re:Nvidia's CUDA Monopoly
on Monday August 07, 2023 @04:20AM (#37031785)
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 own software stack.
Re:AMD CEO: The Next Challenge Is Energy Efficiency
on Wednesday February 22, 2023 @12:54PM (#34899698)
The next challenge is software stack.
Re:The horror has a face – Nvidia’s hot 12VHPWR adapter
on Friday October 28, 2022 @01:10PM (#33374260)
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.
ARM Neoverse V2 Nvidia Grace CPU
servethehome.com/arm-neoverse-v…
Re:Run your own DALL-E-like image generator
on Monday August 08, 2022 @04:22PM (#32390537)
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 add extra layers.
Re:AMD-powered Frontier supercomputer breaks the exascale barrier
on Tuesday May 31, 2022 @02:09PM (#31572813)
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 super fragile, buggy and barely supported. Rather than building a HPL machine I would have preferred see public money spent in a different way.
Re:ETL Pipelines with Airflow: The Good, the Bad and the Ugly
on Friday October 08, 2021 @12:21PM (#28801033)
If you have GPUs, NVTabular outperforms most of the frameworks out there: https://github.com/NVIDIA/NVTabular
Re:Facebook is now claiming official CDC.gov links are “False Informat...
on Thursday July 29, 2021 @05:16PM (#28001600)
This not completely correct because while a carrier transport anything indescrminately, social networks manipulate and control information exchange, for instance giving more priority to a content wrt to another.
Re:CPU-based algorithm trains deep neural nets up to 15 times faster t...
on Friday April 09, 2021 @04:12AM (#26747912)
I think this work is quite ridiculous. As far as I could see, they trained the net with a batch size equal to 1.
Re:Intel 3rd gen Xeon Scalable (Ice Lake): generationally big, competi...
on Tuesday April 06, 2021 @05:28PM (#26717617)
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.
Re:Intel's Ponte Vecchio Xe-HPC GPU Boasts 100B Transistors
on Saturday March 27, 2021 @04:28PM (#26605349)
When? Where? How can it be miles ahead if the hardware has not been released yet?
Re:Intel's Ponte Vecchio Xe-HPC GPU Boasts 100B Transistors
on Saturday March 27, 2021 @04:26PM (#26605336)
One-API is not completely open source. Support for Ponte Vecchio will be not released open source for many reasons.