This actually surprised me. Although I can rattle on and on about parallelism, and I know the main benefits of concurrency, and the theoretical benefits: I’ve actually come to realize that I actually couldn’t design one from scratch. And it’s pissing me off.

I came to this realization while actually working on moving the model loading off the main thread. Now, conceptually and based off experience, I know that this is just how it’s done.

I could even spawn multiple threads - Render thread, Loading thread etc. etc. And just slap on channels and call it a day. But that’s not even true concurrency, and comes back to the fact thta I don’t know how it works.

Plus, isn’t this the whole blocker behind Penny? I think i’ll take a couple of days to try and build it from scratch, and hopefully that teaches me something that I can use to create the state machine that Penny needs too (and finally scratch the itch)