Aviv Beeri

It'll probably be an unfinished disaster but I feel like tinkering.

I think the first goal is "minimal native, maximum Wren" for DOME. It won't be fast, but it'll be more portable. If it's cripplingly slow we can move the hotter pieces over to the native C portion.

So far, the code been a mixture of Handmade Penguin and the http://wren.io  embedding docs.

At some point, this is going to need a logo...

I'm basing DOMEs initial framework APIs on #pico8, because I like working with them, but it isn't a fantasy console. You could certainly extend it into one, I suppose.

I don't have like, a grand vision or lofty goals about why I'm making it.

My screen recording has also made it look much slower than it is.

Been working on some background tweaks to make development easier as well as loading images, which I hope to show off soon. Having a deep consideration about API design first.

(PS: I need better test images)

I'm building an example game (as in, following a tutorial for a very simple platform) as way of checking what utilities my framework is missing. Hopefully will have something to show soon #screenshotsaturday

Font stuff took longer than expected since I learned TTF sucks at low resolution (like DOME currently is).https://twitter.com/avivbeeri/status/1016091268869476352 …

Had some really good progress tonight. Saving something special for #indiedevhour tomorrow, so no screenshots today.

I've been building a copy of this fantastic tutorial designed for #pico8, as a way of checking off features my engine/framework is missing. It's a fantastic getting started tool, so give it a try! https://ztiromoritz.github.io/pico-8-shooter/ 

Considering some alternatives to pull out the SDL dependancy from DOME, because I want this to be both portable and easily distributable, which means a minimal reliance on external libraries, if possible.

Keeping SDL for now as it provides some good services, but the goal is not to use more than core SDL. I _think_ I've completed the initial research phase for implementing a basic audio engine for DOME as well, so implementation will commence soon.

Hm. Audio got more complex again. Diving into the realms of concurrent lock-free mechanisms. Quite the rollercoaster!

Even better, SDL actually has a good mechanism for what I was trying to achieve. Fantastic.

Gradually building up DOME to play Audio. No mixing yet, but nearly at "single-file-playing" which is a huge milestone. Generally happy with the architecture.

Audio work doesn't really fit with #screenshotsaturday, so just read back the thread a bit and see the gif of gameplay I managed to get DOME to produce.

Feeling like a boss: 4 days of trial, error, study, sweat and tears have resulted in a passable audio engine which can load WAV files (mono and stereo) and mix them to play simultaneously. Next is volume and pan controls!

Too late for #indiedevhour, but my audio engine is pretty much feature complete now, thanks to volume and panning features!

...and as a special extra, you can even load .OGG files to play now!

I got a bit multi-threaded with my tweeting, so here's a quick update: In order to implement asynchronous file loading, I've implemented a FIFO and thread pool (available on github seperately) and managed to create a new File api for polling when files are read to use.

The next step is actually to go back and integrate the file IO into the Image and Audio systems so those resources can be loaded async also.

Far too hot to code really... but going to try and do something.