Libretro Core about regoth-bs HOT 6 OPEN

I would postpone this until REGoth is in a state where it would make sense to look into RetroArch. Don't get me wrong, I use RetroArch daily and would love to see REGoth there, but the focus should be to get in a state where we can actually play.

In my opinion, that is ;)

from regoth-bs.

I dont think this is really in scope of this project.
What is the benefit? Portability to any platform retroarch runs on?
Is there any precedent for a libretro core of a engine reimplementation for a non-free game?

from regoth-bs.

Yes, indeed.
Tomb Raider: https://github.com/XProger/OpenLara
Doom: https://github.com/libretro/libretro-prboom

from regoth-bs.

Sure, I absolutely agree. All I wanted is register interest.
And just maybe some precautionary measures can already be taken during further development faciliating a possible upcoming Libretro core.

from regoth-bs.

Just be skimming through the libretrocore docs, it seems like we would need to target its own graphics and input backend. In that case, we would need BS::F to support such a thing, which is not the case right now.

from regoth-bs.

In addition to rendering and input, libretro also requires that all engine filesystem writes that are not temporary or volatile (say, named pipes) are redirected to (normally) a subdir of their configured 'system' or 'saves' dir. Which is why ~/.config/retroarch/ can grow to ridiculous sizes (though that is mostly savestates from dolphin tbf).

from regoth-bs.