Tag
A technical blog post detailing the reverse-engineering of Intel's undocumented next-page prefetcher (NPP) in Skylake-X, tracing how the mechanism evolved from Ivy Bridge through Redwood Cove's LLC Page Prefetcher.
A detailed history of the x86 CPU lineage from Intel's 4004 (1971) through the 80486 (1989), explaining how seemingly arbitrary architectural quirks like little-endian ordering, register naming, and segmentation accumulated through decades of backward-compatibility decisions.
Shareholders sued Intel, CEO Lip-Bu Tan, and the CFO after the company's stock plunged 28.6% in a single day following 'disastrous' Q3 earnings, alleging that statements about Intel's AI-infrastructure prospects were materially false and misleading.
A researcher shares leaked details about Intel's upcoming LGA 1954 platform, highlighting PCIe 5.0 x8/x8/x8 slot configurations, Thunderbolt support, and the return of AVX512 on desktop CPUs.
Intel's stock decline deepens as a broader sell-off worth $3.5 trillion spreads across chip stocks, pressuring the semiconductor sector.
Intel's stock crashed around 28% after a weaker-than-expected Q4 outlook, wiping $12 billion from the U.S. government's 10% stake amid AI-driven memory shortages pressuring PC sales.
Intel在其最新财报电话会议上表示,预算级Wildcat Lake笔记本芯片的需求已超出供应,目前处于售罄状态。同时Intel警告内存短缺短期内不会改善,预计将持续到2027年甚至更久。
A pull request to llama.cpp refactors Intel XMX matrix engine support, delivering massive improvements to speculative decoding performance on Intel hardware.
Intel and AMD jointly announced they will collaborate on x86 CPU software optimizations for PyTorch via oneDNN, with AMD contributing its ZenDNN improvements upstream and joining the UXL Foundation as a steering member.
Ken Shirriff 的深度技术文章,通过开盖显微成像与微码逆向分析,详解 Intel 8087 协处理器如何实现浮点加法与减法,包括 1648 条微指令、整数加法器、移位器与指数转换器等关键电路模块及其处理无穷大等特殊值的方式。
Intel SVP Cho Min-sung confirmed that Razor Lake-AX SoCs will launch in 2028 with 700 TOPS of AI compute, a 3.88x increase over Panther Lake, targeting premium devices and the Physical AI market against NVIDIA Jetson and AMD Ryzen Halo platforms.
Intel released LLM-Scaler-vLLM PV 1.5, a major update to its containerized vLLM stack for Arc Graphics hardware, jumping from vLLM 0.14 to 0.26, adding new Qwen and Gemma model support, FP8/INT4 quantization, and performance improvements.
Intel CEO 吴恩达级别的公开表态,称中国前沿AI模型的采用速度'非常快',反映出中美AI竞争格局的快速演变以及对芯片供应链的潜在影响。
Intel 的 "Crescent Island" GPU 获得了 D 系列命名,并支持最高 480 GB 的 LPDDR5X 内存,主要面向数据中心 AI 加速场景。
Intel 发布基于 Xe3P 架构、专为 agentic AI 优化的 GPU D 系列,这款风冷 PCIe 5.0 显卡支持 vLLM 与开源 PyTorch 生态中的主流开放模型。
Intel 发布面向推理优化的数据中心 GPU D270,配备最高 480GB LPDDR5X 内存、32 个 Xe3P 核心与 256 个 XMX 引擎,采用 350W 风冷 PCIe 5.0 标准形态,专为 agentic AI 和长上下文推理工作负载设计。
一篇博客调查 Intel SDM 手册中记载、但实际 CPU 上会触发 #UD 的幽灵指令 vfmaddrnd231pd 的来源之谜,推测 Intel 在最后一刻移除了该指令但未同步更新文档。
Due to soaring RAM prices driven by the DDR5 shortage, Intel and AMD are reviving support for decade-old DDR4 memory, with new LGA 1700 and AM4 motherboards expected as DDR4 prices remain far lower than DDR5.
Cascadia 论文展示了在11台 Intel Core Ultra X7 358H AI PC 上分布式常驻推理 975B 总参数(41B 激活参数)的 Inkling MoE 模型,通过自定义 MoE 引擎与 OpenVINO 融合算子实现 60.29 聚合 decode tokens/s 的性能。
Google acknowledged that some Android apps may not run smoothly on its new Intel-based Googlebook laptops because many apps were originally designed for Arm chips, saying a 'small handful' of complex apps need developer tuning.