Building an FPGA 3dfx Voodoo with Modern RTL Tools

What appears to be happening is this: someone has taken on the quietly audacious project of recreating a 3dfx Voodoo graphics card in FPGA hardware using modern RTL design tools, and posted about it on Hacker News on March 22nd to a modest but genuine response. The 3dfx Voodoo, for anyone who wasn't haunting LAN parties in the late 1990s, was effectively the GPU that invented consumer 3D gaming — a piece of silicon so beloved that its discontinuation in 2000 still generates something resembling actual grief in certain corners of the internet. The project sits at an interesting crossroads: vintage hardware preservation meeting contemporary tooling, with RTL design tools that are substantially more accessible than anything that existed when the original Voodoo was being fabbed. What makes this particularly interesting as a cultural artifact is the implicit argument it contains — that the gap between "enthusiast with modern tools" and "1990s GPU engineering team" has narrowed enough for one person to attempt a serious recreation. That's a claim worth examining carefully. The signal scoring suggests the Hacker News audience found it interesting without quite knowing what to make of it yet, which is often the right reaction to something genuinely novel.

If confirmed, here is what this means. The obvious implication is for hardware preservation: an FPGA recreation of the Voodoo could run original 3dfx software — Glide API and all — on modern hardware indefinitely, solving the slow degradation problem that haunts vintage GPU collectors. But the more interesting second-order effect is what this says about the democratisation of chip design. Modern RTL tools, combined with the AI-assisted code generation that's increasingly part of the vibe coding ecosystem, have apparently lowered the barrier enough that passion projects at this level of complexity are now feasible for individuals. If one person can meaningfully reproduce a historically significant GPU architecture, the implications for open-source hardware development are considerable — and slightly unsettling for anyone who thought chip design remained safely in the domain of well-funded teams.

Watch for whether this project surfaces additional signals from hardware preservation communities or open-source silicon forums — any engagement from those quarters would indicate the work has substance beyond the initial posting. A follow-up with working Glide output would change this story entirely.

NewsHive monitors these sources continuously. All signal titles above link to the original reporting.

Intelligence by NewsHive. Need help navigating what this means for your business? Contact GeekyBee →