Install
GPUs & Accelerators
AI/graphics processors, memory bandwidth, and massive‑scale training chips.
- 25 Tracked terms
- Last 30 days Feed window
What this topic collects on
An article joins this feed when it matches these terms. Each one is also a search of its own.
Related topics
Latest in GPUs & Accelerators
Deputy Minister Ryu Je-myung Tours South Korean AI Ecosystem with 22-Nation Diplomatic Corps, Signaling Global Hub Ambitions
4+ hour, 35+ min ago (290+ words) Ryu Je-myung, Second Vice Minister of South Korea's Ministry of Science and ICT, toured domestic and international company booths at AI Festa 26 held at COEX in Seoul on the 7th, accompanied by 24 ambassadors from 22 key partner nations including Sweden, China, India,…...
MacBook vs dedicated GPU for local LLMs: how much RAM you really get, and when each wins in 2026
13+ hour, 31+ min ago (1526+ words) Local LLM, Apple Silicon, NVIDIA, Hardware, MLX, Guides Ask HN keeps asking the same question—Apple unified memory vs Nvidia VRAM for local LLMs. Here is how much memory MacBooks actually expose to models, why GPUs win on speed, and…...
Review: AMD Ryzen Threadripper 9000 Series
5+ hour, 14+ min ago (1071+ words) Zen 5 also brings enhanced AVX-512 support, helping ensure performance improvements in professional simulation applications, such as Altair Radioss, Simulia Abaqus, and Ansys Mechanical extend beyond IPC alone. Simultaneous Multi-Threading (SMT) continues to allow each core to handle two threads simultaneously....
AMD seeks Korean AI chip partners to expand global reach, challenge Nvidia
5+ hour ago (166+ words) AMD Chair and CEO Lisa Su met with Korean AI companies in Seoul to discuss pairing AMD's processors with Korean neural processing units. These partnerships aim to create systems for global markets, expanding AMD's inference offerings and challenging Nvidia. The…...
South Korea aligns AMD and startups as DeepX pursues physical AI tie-up
8+ hour ago (453+ words) CHOSUNBIZ Chosunbiz South Korea aligns AMD and startups as DeepX pursues physical AI tie-up Government-brokered AI push brings AMD, Su Lisa and local chip startups together for real-world, low-power collaboration The government's push for sovereign artificial intelligence (AI)—a project…...
Table of video card correspondence to CPU
5+ hour, 43+ min ago (1640+ words) We tried to match video cards to central processors and compiled it into a special table. Now, when choosing a video card, you can evaluate whether your processor matches the video card or not. Of course, our analyzes are not…...
Performance rating of video cards in mining
5+ hour, 43+ min ago (444+ words) Following the wishes of our readers, we have provided a table of the summary performance of all modern technologies...
Open-Source SCSKiller Tool Precompiles Shaders, Kills PC Stutter [2026]
11+ hour, 54+ min ago (378+ words) Crucially, SCSKiller does not need to stay running once the cache is populated. Because the compiled shaders live in the driver’s own cache rather than in a separate store the tool manages, a user can close SCSKiller entirely and the…...
AMD pushes local agentic AI with Ryzen AI Max PRO 400 Series
19+ hour, 43+ min ago (380+ words) AMD’s new Ryzen AI Max PRO 400 Series is designed to help businesses run agentic AI workloads locally, with up to 192GB of unified memory and as much as 160GB available for GPU use. Pope Leo XIV advocates for a renewed alliance between…...
Original DOOM Gets NVIDIA DLSS 5 Neural Rendering Through a Custom DirectX 12 Port
16+ hour, 10+ min ago (252+ words) At this point, getting DOOM to run on increasingly ridiculous hardware has practically become a tradition. This latest experiment goes in a different direction: developer Nikolai Zhivotenko has brought NVIDIA DLSS 5 Neural Rendering to the original 1993 classic through a custom…...