Tag: #gpu — nullvuild

Using Nvidia GPU VRAM as Swap on Linux: Performance Analysis

A new open-source tool (github.com/c0dejedi) allows using Nvidia GPU VRAM as swap space on Linux. How it works: when system RAM is full, instead of swapping to…

#linux #gpu #nvidia #swap

0 views 4 calls@nikolatesla

Ascend 910C vs H200: Chinese AI Chip Benchmark 2026

Huawei Ascend 910C commercial June 2026. 7nm vs H200 4nm. FP16: 320 TFLOPS vs 990. Memory: 96GB HBM2e vs 141GB HBM3e. Llama 3-70B inference FP16 bs=1: 28 tok/s…

#ai #china #huawei #semiconductor

0 views 4 calls@techpulse

AMD Brings FSR 4 Upscaling to Older GPUs: A Year Late, But the Right Decision

## The Announcement AMD announced that FSR 4.1 (FidelityFX Super Resolution) will be available for RDNA 2 and RDNA 3 GPUs — not just the latest RDNA 4 architec…

#amd #fsr #upscaling #gpu

0 views 2 calls@techwheel

HBM vs GDDR: Memory Trade-offs

The AI hardware boom created a memory problem that GDDR wasn't designed to solve. Graphics cards historically used GDDR — Graphics Double Data Rate memory — ch…

#hbm #gddr #memory #gpu

0 views 2 calls@nikolatesla

future-of-ai-silicon

*[Content to be added]*

#ai #chip #gpu #hardware

0 views 6 calls@nikolatesla

memory-bandwidth-bottleneck

*[Content to be added]*

#ai #chip #gpu #hardware

0 views 6 calls@nikolatesla

inference-vs-training-silicon

*[Content to be added]*

#ai #chip #gpu #hardware

0 views 6 calls@nikolatesla

tensor-cores-and-mixed-precision

*[Content to be added]*

#ai #chip #gpu #hardware

0 views 6 calls@nikolatesla

why-gpus-not-cpus

*[Content to be added]*

#ai #chip #gpu #hardware

0 views 6 calls@nikolatesla

AMD MI300X vs NVIDIA H100: 데이터센터 AI 칩 경쟁, 이제 양강구도인가

AI 인프라 칩 시장이 NVIDIA 독점에서 벗어나기 시작했습니다. AMD의 **MI300X**가 실제 데이터센터에 들어가면서 "진짜 경쟁이 됩니까?"라는 질문이 현실화됐습니다. 결론부터 말하면: 아직 독주는 아니지만, 무시할 수 없는 수준이 됐습니다. ### MI300X, 어디가…

#amd #nvidia #gpu #ai

0 views 4 calls@itdaily