[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Public WebGL] WebVulkan

I’m pretty sure Web developers have long agreed on the fact that web applications bring the magic of high portability in cost of performance degradations… And we are perfectly aware of these issues. Yet this what makes the Web “The Web” and that’s why we love it. You are going to write this piece of code and it’s going to work everywhere (or at least most probably will). Anyway, my point is building WebGL 3 on top of Vulkan shouldn’t be the future. Sure, 3D applications will definitely benefit from all the awesome features the almighty Vulkan will provide but let’s not forget the fact that Apple does not support Vulkan (and I can’t say they will, since they have Metal) and if my 3D content can’t run everywhere including apple devices, then the whole “webiness” of the Web just disappears. But we can’t also rely on OpenGL for our future, so either WebVulkan will have a layer (like ANGLE) that can switch between APIs on different platforms (DirectX12 or Vulkan on Windows, Metal on Apple) or we have to build a completely new 3D API for the web which is supported by all vendors (IMHO I don’t even think that’s a reasonable choice since it requires too much work). Some kind of a layer also means there will definitely be performance issues (but not too much) but I think that’s a kind of thing Web developers are ready to sacrifice to get the portability the Web provides