I`m already signed it in 2008 as far as I remember, see "Yeugen Golushkov (eugene_gff)" line in the AUTHORS file.Assaf Raman wrote:Eugene ... can you sign the contributor forum and send it back to me - so we can use it in the main trunk?
Reflection info is laid down in compiled shader file in very straightforward manner - structures on disk and in memory are almost always identical. So, developing another serialization format for D3DReflect information (with at least variable length strings for semantics, input/output/constant buffers descriptions and so on) is not necessary easier than partially support existing. Performance would be better if reflection info would not be duplicated, reliability can be ensured by comparing results of emulated D3DReflect calls with native ones in debug builds.Assaf Raman wrote: BTW: If we save the binary version of the shader, why not save the D3DReflect results as well? Shouldn't be hard at all and will be better performance and more reliable then the code you wrote, right?
I saw merging efforts and than saw some of bugfixes were rolled back. In any case, I need a solid foundation for my project (Live Interior 3D), therefore I can say that https://bitbucket.org/eugene_gff/ogre-winrt is tested as good as I can in both Win32 and WinRT modes, both CMake building process and runtime behavior of SampleBrowser. I can not say this about summer project, as it`s hard to follow: all winrt related commits were smashed into two or three megacommits than parts of code were commented, uncommented and commented again - it`s not stable enough for my purposes.Assaf Raman wrote: Eugene - What is the status with my summer student project, is all your code merged into his repo?