CUDA

Razer, LG, and Lenovo have presented early glimpses at the upcoming RTX 5060 and 5050 laptop graphics cards. The mainstream and entry-level products are the last members...

According to VideoCardz, Nvidia's board partners have received the final specifications for the upcoming RTX 5060 Ti. Additionally, prior reports indicate that Nvidia has pushed the launches...

Article URL: https://www.tedtanner.org/nvidia-gtc-2025-the-time-has-come-the-valley-said/ Comments URL: https://news.ycombinator.com/item?id=43455287 Points: 1 # Comments: 0

AMD is back in the fray but has to improve its GPUs to truly challenge NVIDIA.

Nvidia announced that Nvidia Blackwell hardware will accelerate by 50 times the big computer-aided software engineering firms' software for digital twins.

Nearly RTX 4080 gaming performance for under $600? Why not!

The RTX 5070 excels at 1440p gaming, and it does a decent job at 4K. But it's not a major upgrade over the RTX 4070, and it's even less impressive in games...

32-bit PhysX is officially dead.

Nvidia's RTX 5080 for laptops may turn out be as powerful as last-gen's RTX 4090 laptop GPU

Take a hard look at the specs, and you'll soon realize the RTX 5080 is just an RTX 5070 in disguise. Nvidia's greed may be at play, and we'll prove it by analyzing historical GPU configurations.

A few days ago, we reported on AMD's newest expansion plans for Serbia. The company opened two new engineering design centers with offices in Belgrade and Nis. We were invited to join the opening ceremony and got an exclusive interview with one of AMD's top executives, Andrej Zdravkovic, who is the ...

The Nvidia RTX 5070 is almost here, and Nvidia claims it can match the RTX 4090 frame for frame. But how much frame generation does that need? Let's find out.

Even setting aside Frame Generation, this is a fast, power-hungry $2,000 GPU.

Nvidia's GeForce RTX 5090 is the most brutally fast graphics card ever introduced, augmented by new DLSS 4 technology that feels like magic. But you pay dearly for it.

Asus has added enough of its own innovation to its NUC-branded mini-PCs for them to finally feel like Asus products.

In this blog post, we'll walk through an implementation of the SGEMM (Single-precision GEneral Matrix Multiply) operation defined as C := alpha*A*B + beta*C. We will review three different kernels, each optimized for specific matrix size problems. Our final implementation is optimized for Ampere architecture and outperforms cuBLAS on wide range of matrix size problems.

The Nvidia RTX 5090 is going to be the fastest card in the world when it debuts, but how does it compare to the RTX 4090? Let's find out.

Nvidia today unveiled Nvidia Project Digits, a personal AI supercomputer that serves AI researchers, data scientists and students worldwide.

The new generation of PC gaming is here...