Very neat, this reminds me of the organic shapes of passive demultiplexers in photonics such as https://pubs.acs.org/doi/10.1021/acsphotonics.7b00987

This looks like quite a huge breakthrough, unless I'm missing something?

~25x faster performance than Flux-dev, while offering comparable quality in benchmarks. And visually the examples (surely cherry-picked, but still) look great!

Especially since with GenAI the best way to get good results is to just generate a large amount of them and pick the best (imo). Performance like this will make that much easier/faster/cheaper.

Code is unfortunately "(Coming soon)" for now. Can't wait to play with it!

You can download Windows, Mac, and Linux builds from the GitHub releases page. Note that you'll also need the graphics and .DAT files from the original game, which you can find all over the internet. (Hint: try the Internet Archive)

To get the Mac build running:

   - Drag the "SpaceCadetPinball" out from the DMG file to somewhere local (e.g. /Applications )
   - Right-click on the executable and choose "Show Package Contents"
   - Navigate to Contents/Resources/
   - Unzip the file you found with all the game data
   - Move all the unzipped game files into Contents/Resources