Hey all,
I've been working on a little project called ReMac simply because I love the look of the old iMacs and didn't want to throw one away just because it could no longer run a modern browser and watch YouTube videos.
The idea is simple: instead of asking the old computer to do all the work, let a more powerful machine do the rendering and turn the old Mac into a thin client.
The display machine is a 2007-era Intel iMac with a Core 2 Duo T7300 @ 2.0 GHz, running a patched version of NetBSD 10.1.
The server is a Debian 13.7 Trixie machine with an Intel Core i7-7700 @ 3.6 GHz, 4 cores / 8 threads, 16 GB RAM and Intel HD Graphics 630.
The Debian machine runs the desktop, browser, web content and state. The iMac receives the resulting frame buffer, decompresses it, displays it, and sends mouse and keyboard input back to the server via Ethernet.
Although just a POC at this stage the interesting part is that it actually works surprisingly well. To my old eyes anyway.
I'm running a minimal desktop with Qt/WebEngine and Chromium-based browsing, including YouTube, on the server while the old iMac handles the display and input. The Core 2 Duo doesn't need to run the browser or do the web rendering itself — it mainly needs to receive, decompress and present the resulting pixels.
It's not intended to compete with established remote-desktop protocols.
Next up is getting NetBSD running on a G3 iMac, porting the ReMac client to it, and then seeing if I can get the entire NetBSD system running from RAM, with no local disk required — turning either machine into a completely diskless ReMac client.
I've been working on a little project called ReMac simply because I love the look of the old iMacs and didn't want to throw one away just because it could no longer run a modern browser and watch YouTube videos.
The idea is simple: instead of asking the old computer to do all the work, let a more powerful machine do the rendering and turn the old Mac into a thin client.
The display machine is a 2007-era Intel iMac with a Core 2 Duo T7300 @ 2.0 GHz, running a patched version of NetBSD 10.1.
The server is a Debian 13.7 Trixie machine with an Intel Core i7-7700 @ 3.6 GHz, 4 cores / 8 threads, 16 GB RAM and Intel HD Graphics 630.
The Debian machine runs the desktop, browser, web content and state. The iMac receives the resulting frame buffer, decompresses it, displays it, and sends mouse and keyboard input back to the server via Ethernet.
Although just a POC at this stage the interesting part is that it actually works surprisingly well. To my old eyes anyway.
I'm running a minimal desktop with Qt/WebEngine and Chromium-based browsing, including YouTube, on the server while the old iMac handles the display and input. The Core 2 Duo doesn't need to run the browser or do the web rendering itself — it mainly needs to receive, decompress and present the resulting pixels.
It's not intended to compete with established remote-desktop protocols.
Next up is getting NetBSD running on a G3 iMac, porting the ReMac client to it, and then seeing if I can get the entire NetBSD system running from RAM, with no local disk required — turning either machine into a completely diskless ReMac client.