Hackaday – 25 May 26 Z386: An Open-Source 80386 Built Around Original Microcode There are many ways you can implement an Intel i386 CPU on an FPGA, with the use of original microcode probably being one of the most interesting approaches. This is what [nand2mario]’s z386 … [!quote]+ 在 FPGA 上实现英特尔 i386 CPU 的方法有很多,其中使用原始微代码可能是最有趣的方法之一。这就是 [nand2mario] 的 z386 项目所做的事情,最近的一篇博文总结了这个 FPGA 项目迄今为止的发展情况。 该项目与之前开发的 z8086 项目类似,大家可能都猜到了,除了英特尔 8086 CPU 之外,该项目也做了类似的事情。通过执行原始微代码,你基本上可以保证与原始硬件的紧密兼容性,当然,8086 和 80386 之间的微代码规模是完全不同的。 它的指令集更大,需要跟踪的内部状态也相应复杂得多,包括内存管理、分页和寄存器调试等所有新奇功能,以及从 i286 开始的保护模式扩展。 目前,z386 可在多种 FPGA(包括 Altera Cyclone V 和 Gowin GW5A)上运行,其性能相当于约 70 MHz 的 i386,但周期效率略低,部分原因可能是 16 kB 缓存有限,而最快的 i386 CPU 拥有 32 kB 以上的缓存。无论如何,它足以运行各种软件,包括 DOOM 等游戏。 需要注意的是,这里的目标并不是要比 ao486 等内核的性能更强,而是要对原始硬件及其与上述微码的交互进行考古重建。 github.com GitHub - nand2mario/z386: Compact 80386 CPU in SystemVerilog Compact 80386 CPU in SystemVerilog 1 个帖子 - 1 位参与者 阅读完整话题
cursor.com Cursor - The best way to code with AI Built to make you extraordinarily productive, Cursor is the best way to code with AI. cursor属于是AI IDE里做的做好的产品,自带的composer模型量大管饱,然后你就算用万人骑机场也能用opus 4.7、gpt5.5 。当然解决了网络问题性价比最高的还是走官方订阅。 1 个帖子 - 1 位参与者 阅读完整话题
最近看到很多人用 Lovable / Base44 / Bolt / Replit / Cursor 很快做出了产品。 我在做 TraceMind ,想解决产品上线后的下一步问题:不只是把产品做出来,而是让 AI 帮你看懂真实用户行为。 核心思路比较简单: 1. AI 自动埋点 不需要先设计复杂埋点,也不需要维护一套看板。 2. AI 分析数据 可以直接问 AI:用户卡在哪、为什么没转化、哪个功能没人用。 3. AI 迭代闭环 改完产品后,再验证用户行为有没有真的变好。 现在想找 5 个已经上线、或者正在找前 10 个真实用户的 AI-built 产品试点。我可以免费协助接入,并在 24 小时内给第一份用户行为诊断。 如果你有类似产品,或者正在用 AI 做产品,欢迎交流。我也想听听这个方向是否真能解决你们上线后的问题。 介绍页: https://tracemind.sandbox.galaxycloud.app/
I've been working on an AI video generation tool for the past few months. It's called Gemini Omni — you describe what you want, upload a reference image, and it generates a short video clip. No editing software, no timeline, just a prompt. What it does: Text-to-video: describe a scene, mood, camera move, and format Image-to-video: animate album art, product photos, or brand stills Outputs landscape (16:9), portrait (9:16), and square (1:1) for social platforms Who's it for: Creators, marketers, and small teams who need video content for ads, launches, reels, and social posts — but don't have a full production setup. There's a free tier to try it out. I'm still actively building and improving it. 🔗 Main site: https://gemini-omni.online 🔗 Also live at: https://omnivideo.video
A short-form video generator where video and audio come out of a single generation pass, instead of needing a separate sound step. Text-to-video and Image-to-video Omni Reference: up to 9 reference images, 3 reference videos, 3 reference audio clips for style/character/voice anchoring Synced audio generated together with video — dialogue, ambience, foley line up on the first export 5s / 8s / 10s clips, 16:9 / 9:16 / 1:1 480P and 720P, Fast and Pro modes Typical generation time 30–90s Free credits on signup: https://geminiomni.org Happy to take questions about how the reference inputs work or what the output looks like in practice.
Notion Notion | Where teams and agents work together A collaborative AI workspace, built on your company context. Build and orchestrate agents right alongside your team's projects, meetings, and connected apps. 审查比GPT官网的少 但是应该不是GPT image 2 4 个帖子 - 2 位参与者 阅读完整话题
Cerebral Valley Built with Opus 4.7: a Claude Code hackathon Join us again for Built with Opus 4.7: a Claude Code virtual hackathon! Opus 4.7 is our most capable model yet, and we’re excited for you to push the limits ... | Tuesday, April 21, 2026 We’ll select 500 participants and give each one $500 in Claude API credits to build for a week. Compete to win $100k in Claude API credits. https://x.com/claudeai/status/2045248224659644654 1 个帖子 - 1 位参与者 阅读完整话题