only platforms on desktop we will need is X11, Wayland, Win32, and Cocoa i think?
only platforms on desktop we will need is X11, Wayland, Win32, and Cocoa i think?
PWe don’t even have maintenance effort to make a GLFW 3.0 let alone a full native 3.0
PMac doesn’t deserve our dev time tbh, but everything else yeah
Fsad Mac noises
RI'm still confused just to how the windowing implementation is supposed to work. The proposal does not help me in the slightest
Lmaybe i'll toy around with a DRM impl
Falthough apple computers arent supposed to make noise
Pfeel free to amend the proposal. It is open source after all.
PRight now all I’ve been told is “there are problems”
Pthe idea is you have an ISurface. If the backend can support it you’ll be able to cast it to a specialised surface. If the backend does support it, the TryEnabke methods return true
RThat helps explain it, I think, but I probably won't be back to work on it for a while, I think the PR and branch are still up if anyone wants to have a go. I have kinda started moving my personal project away from C#, so it means I am working elsewhere, I also have full time non programming work now, that has me up at 2 in the morning
RSo I have very little energy to do anything outside of work, and the energy I do have I end up putting into my own projects.
Pk
FIn Silk.Net tutorials (from repo) code is structured in classes like Shader, Texture etc. Is it how the real project should be organized or its just for demonstaration purposes for easy learning?
Lyou can structure it however you like
Labstractions over GL range a lot
ZThat's the best part tbh
Lits good to note you dont always have to abstract over every part of GL
Llike if you only ever need 1 shader you dont need a shader abstraction
ZI mean it's probably still good to have a basic abstraction in case you might need it later
ZBut depends
Lyeah it really depends
Zarchitecture can really get wild sometimes
Lmy engine has a shader abstraction that i really havent needed in months (i have cut my engine down to only need a single shader for everything) but i havent deleted them since they are helpful incase i do ever need it in the unlikely scenario
ZIs your engine GL-based?
Lit was originally built for GL, but has expanded much further
Zoh wow
ZThat sure is a lot
Lthis is after i culled veldrid too lol
LD3D9 is on another branch
Land GL is technically 2 separate backends, one for FFP and one for modern
ZWas there something wrong with it?
Lunneeded, and buggy
Leverything veldrid handled was handled better by specialized code
ZOh well I suppose I won't use it then (Not like I was going to but I considered it before)
Ltechpizza's fork of veldrid fixes a lot of the bugs with veldrid, but since i no longer had a use for veldrid, i just didnt bother switching to it
Lless code to maintain is always better too
ZHm I suppose I'll just stick to plain Silk.NET
ZI doubt I'll have to write THAT much code anyways
ZIt already does a lot for me
Li just like learning graphics APIs
ZDon't really have any major graphic programming experience
ZProbably will go with OpenGL
ZSeems like the right start
LOpenGL or D3D11 are good starting choices
Lam trying to abstract shit in vulkan rn and its painful and I have no clue where im going
ZIs D3D11 hard or Silk does some extra stuff as well other than bindings?
LD3D11 isnt hard, its about the same difficulty of GL, just a little bit more explicit and with less old cruft
ZWell, I heard you can do some impressive stuff with modern GL in terms of performance
Lyea performance from GL wont matter for 99% of cases
Lwhat makes D3D11 a decent option is the arguably better API than GL
