<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>The Machine Herald — Hardware &amp; Semiconductors / GPUs</title><description>GPUs articles in Hardware &amp; Semiconductors from The Machine Herald.</description><link>https://machineherald.io/</link><language>en-us</language><copyright>The Machine Herald. AI-generated content with verifiable provenance.</copyright><generator>Astro + Machine Herald Pipeline</generator><item><title>AMD Unveils MI400 Series on TSMC 2nm and Helios Rack-Scale AI Platform as Delay Rumors Swirl</title><link>https://machineherald.io/article/2026-03/26-amd-unveils-mi400-series-on-tsmc-2nm-and-helios-rack-scale-ai-platform-as-delay-rumors-swirl/</link><guid isPermaLink="true">https://machineherald.io/article/2026-03/26-amd-unveils-mi400-series-on-tsmc-2nm-and-helios-rack-scale-ai-platform-as-delay-rumors-swirl/</guid><description>AMD&apos;s Instinct MI400 family debuts three CDNA 5 accelerators on TSMC 2nm alongside the 72-GPU Helios rack, but a production timeline dispute clouds the second-half 2026 launch.</description><pubDate>Thu, 26 Mar 2026 11:19:17 GMT</pubDate><source>4 verified sources</source><category>amd</category><category>ai-accelerators</category><category>gpu</category><category>data-center</category><category>hbm4</category><category>nvidia</category><category>semiconductors</category></item><item><title>AI&apos;s Insatiable Memory Appetite Forces NVIDIA and AMD to Retreat From 16 GB Gaming GPUs as PC Hardware Makers Warn of the Most Challenging Year Ever</title><link>https://machineherald.io/article/2026-03/20-ais-insatiable-memory-appetite-forces-nvidia-and-amd-to-retreat-from-16-gb-gaming-gpus-as-pc-hardware-makers-warn-of-the-most-challenging-year-ever/</link><guid isPermaLink="true">https://machineherald.io/article/2026-03/20-ais-insatiable-memory-appetite-forces-nvidia-and-amd-to-retreat-from-16-gb-gaming-gpus-as-pc-hardware-makers-warn-of-the-most-challenging-year-ever/</guid><description>Both GPU makers are shifting mainstream production to 8 GB models as DRAM prices surge 180 percent, MSI calls 2026 its hardest year since founding, and IDC warns the sub-$500 PC market may vanish by 2028.</description><pubDate>Fri, 20 Mar 2026 09:34:50 GMT</pubDate><source>7 verified sources</source><category>NVIDIA</category><category>AMD</category><category>GPU</category><category>DRAM</category><category>memory shortage</category><category>PC gaming</category><category>RTX 5060</category><category>RDNA 4</category><category>HBM</category><category>semiconductors</category><category>MSI</category><category>gaming hardware</category></item><item><title>AMD Launches Radeon RX 9070 XT With RDNA 4 Architecture, Matching RTX 5070 Ti at $599 Before Supply Collapses</title><link>https://machineherald.io/article/2026-03/04-amd-launches-radeon-rx-9070-xt-with-rdna-4-architecture-matching-rtx-5070-ti-at-599-before-supply-collapses/</link><guid isPermaLink="true">https://machineherald.io/article/2026-03/04-amd-launches-radeon-rx-9070-xt-with-rdna-4-architecture-matching-rtx-5070-ti-at-599-before-supply-collapses/</guid><description>AMD&apos;s first RDNA 4 graphics cards, the Radeon RX 9070 and RX 9070 XT, launched on March 6 at $549 and $599 with competitive rasterization performance, a machine-learning-powered FSR 4 upscaler, and 16 GB of VRAM — but sold out within hours as AMD described demand as &apos;unprecedented&apos;.</description><pubDate>Wed, 04 Mar 2026 14:50:12 GMT</pubDate><source>3 verified sources</source><category>AMD</category><category>GPU</category><category>RDNA 4</category><category>Radeon</category><category>Semiconductors</category><category>Gaming Hardware</category><category>FSR 4</category></item><item><title>Intel Announces Data-Center GPU Push to Challenge Nvidia&apos;s AI Chip Dominance</title><link>https://machineherald.io/article/2026-02/05-intel-announces-data-center-gpu-push-to-challenge-nvidias-ai-chip-dominance/</link><guid isPermaLink="true">https://machineherald.io/article/2026-02/05-intel-announces-data-center-gpu-push-to-challenge-nvidias-ai-chip-dominance/</guid><description>Intel CEO Lip-Bu Tan reveals plans to build data-center GPUs and hires former Qualcomm architect Eric Demers to lead the effort.</description><pubDate>Thu, 05 Feb 2026 12:09:29 GMT</pubDate><source>5 verified sources</source><category>Intel</category><category>Nvidia</category><category>GPU</category><category>AI</category><category>semiconductors</category><category>data-center</category></item><item><title>NVIDIA Unveils Rubin: A Six-Chip Platform Promising 10x Cost Reduction for AI Inference</title><link>https://machineherald.io/article/2026-02/04-nvidia-unveils-rubin-a-six-chip-platform-promising-10x-cost-reduction-for-ai-inference/</link><guid isPermaLink="true">https://machineherald.io/article/2026-02/04-nvidia-unveils-rubin-a-six-chip-platform-promising-10x-cost-reduction-for-ai-inference/</guid><description>NVIDIA announces its most ambitious AI platform yet at CES 2026, integrating six new chips designed to dramatically reduce the cost of running AI models.</description><pubDate>Wed, 04 Feb 2026 21:56:32 GMT</pubDate><source>3 verified sources</source><category>nvidia</category><category>rubin</category><category>ai-hardware</category><category>ces-2026</category><category>gpu</category><category>inference</category><category>data-centers</category></item></channel></rss>