Lnot possible in silk right now
Cis there any insurmountable difficulty？
Lyes, it's the fact this is C#
CYesterday I tried to forward the webgl api to silk.net by myself, but failed
Lthe WASM dotnet runtime is too bad
CGL.GetApi(name => {
switch (name)
case "glGenTexture":
return GetFunctionPointerForDelegate(canvas.CreateTexture);
....
})
switch (name)
case "glGenTexture":
return GetFunctionPointerForDelegate(canvas.CreateTexture);
....
})
Lnot how it works
Cmicrosoft is junk 
Lit's a beta pushed to prod, if you keep that in mind it makes sense
CThe runtime of wasm comes from mono, right?
Li think so
CHow do you feel that many new functions of .net are mono. . . .
Lit makes sense
Lmono is more portable
CMicrosoft is trying to weaken mono's influence. . .
Lmono is integral to the dotnet ecosystem
CFor example, the existence of mono in .net5/6/7 is rarely mentioned, so that chatgpt does not know that mono is still being updated
Lit's an implementation detail, as it should be
CI wanted to embed mono into my c++ program before, and then found that the dll of mono was even renamed to coreclr...
CI solved this problem, fbo creation failed
Ljust gave it another shot, and yea its still borked
Llatest net8 compiler and everything
C.net 8's complier is aot ?
Lit has an AOT compiler
Lmultiple, actually
CNot only wasm, I hope that the aot compiler of .net can work as stably as golang
LNativeAOT is already very stable for desktop use
Land mobile has basically always been AOT'd iirc
LNativeAOT has even been ported to all 3 major console platforms
Lit's Mono WASM AOT thats my problem
Cwhen i did some embedding of .net runtime in to a c++ application, i just hosted the like, normal .net core runtime
Ccoreclr or whatever
Aohhh okay i see, makes sense
BI was just testing one of the web gpu experiments and noticed the nuget packages don't seem to be built with
libwgpu_native.dll, I built the dll myself and it works fine, is this intended? Running on Windows x64
Lalready fixed, just not released to nuget
Bahh ok
L@Perksey should we do a 2.17.2?
K(isn't SPIRV kinda close now and 2.18 would seem appropriate?)
Lidk, it's only been a week since 2.17.0 so 2.18.0 would be a bit early
Kye I guess, nothing wrong with it though imo
FI don't believe confidently incorrect poorly trained glorified markov chains to be useful in development, because a good bit of the information you get from them is completely incorrect, and is presented in such a way that it is difficult to tell the incorrect from the correct
and if you're using ChatGPT as a search engine or code autocomplete, you're wrong
and if you're using ChatGPT as a search engine or code autocomplete, you're wrong
PNah, this isn’t game breaking and emergency patches make us look bad
Lmakes sense
YHey! Is it possible to release the fix for Webgpu? Maybe in experimental feed
Pshould already be in the experimental feed
YAH My bad, yes you're right
Lrule of thumb is, if its in
mainare the direct3d & dxgi macros implemented anywhere? (specifically
currently making my own but I am just wondering
FAILED and SUCCEEDED)currently making my own but I am just wondering
Pyou can implicitly cast to a HResult
Pyou can do things like
if (HResult.IndicatesSuccess(...)) or something like that