Part III: The evolution to AI GPUs
CUDA, Tensor Cores, and NPUs rewrote computing history.
Tsavorite’s Omni Processing Unit
A custom GPU with an Arm Neoverse CPU.
GPUs vs. ASICs—the threat from within
Some of AMD’s and Nvidia’s biggest customers are building their own AIPs.
Lisuan G100 GPU does well on OpenCL
Who cares about OpenCL performance?
Nvidia brings CUDA to RISC-V, opening new front in AI compute
In China, Nvidia shows how CUDA can run on a RISC-V host—and signals it’s ready for a more open AI future.
AMD ‘Scales’ up its CUDA capabilities
Spectral Compute’s Scale lets AMD run native CUDA programs.
Chaos V-Ray 6 Benchmark
Is your system’s rendering performance as fast as you think it is?
Moore Threads commits to AI GPUs and training center
Blocked from US suppliers, the company seeks a fab.
Can anyone catch up with Nvidia?
Are there any playbooks AI processor companies can use to catch up with Nvidia? A lot of people are asking that question.
Everybody wants CUDA, right?
Intel offers an escape route.
Nvidia’s massive Ampere GPU
Nvidia has been pushing the envelope on graphics chips since it integrated the geometry processor and pixel shader into one chip and called it a GPU, 21 years ago. The potential for GPUs then vastly expanded in 2003 when a branch of development spiked out of GPU applications that took advantage of the GPUs parallel processing for pure computation using … Read more
AI accelerators and open software transform the computing landscape
Three years ago, we had maybe six or less AI accelerators, today there’s over two dozen, and more are coming. One of the first commercially available AI training accelerators was the GPU, and the undisputed leader of that segment was Nvidia. Nvidia was already preeminent in machine learning (ML) and deep learning (DL) applications and adding neural net acceleration was … Read more