jv8 VS Ejecta

I owe a lot of the most informative programming work I’ve done to Impact.

Impact was so ahead of its time. Proud to say I was one of the 3000 license owners. One of the best purchases I’ve ever made. The only game I’ve ever really properly finished was made in Impact.

I loved that the source code was part of the license, and even modified the engine and the editor to suit my needs.

I was so inspired that I worked on my own JS game engine (instead of finishing games - ha!) for years after. I never released it, but I learned a ton in the process and made a lot of fun mini web games with it.

I was also inspired by Impact’s native iOS support, but frustrated that it didn’t run on Android (at the time at least), so I fumbled my way through writing JVM bindings for V8 and implemented a subset of WebGL to run my game engine on Android without web views.[0] I made the repo for V8 bindings public and to my surprise it ended up being used in commercial software.

I won’t bore you with the startup I tried to bootstrap for selling access to private GitHub repos, which was inspired by Impact’s business model…

Anyway, it warms my heart and makes me laugh to see Impact getting an update for the “modern” web with a C port!

I’d say these are strange times for the web, but I can’t remember a time when things were anything but strange. Cheers!

[0]: https://github.com/namuol/jv8

It does, but the main speedup comes from using WebGL instead of Canvas2D. Sadly, Canvas2D is still as slow as it ever was and I really wonder why.

Years back I wrote a standalone Canvas2D implementation[1] that outperforms browsers by a lot. Sure, it's missing some features (e.g. text shadows), but I can't think of any reason for browser implementations needing to be _that_ slow.

[1] https://github.com/phoboslab/Ejecta