"Misha Koshelev" <[email protected]> wrote:
>> Instead of implementing D3DXCreateMeshFVF, wouldn't it be better to >> implement D3DXCreateMesh and call the latter from D3DXCreateMeshFVF by >> previously using D3DXDeclaratorFromFVF? >> >> D3DXCreateMeshFVF would be pretty straight then: >> 1. get the const LPD3DVERTEXELEMENT9 *pDeclaration from the supplied FVF >> 2. call D3DXCreateMesh >> >> BTW, D3DXGetFVFVertexSize could then also be forwarded to >> D3DXGetDeclVertexSize and the code simplified. >Thank you > >Thank you. Yes indeed this is the approach that has been recommended by Henri >Verbeet as well. > >Specifically, after the D3DXCreateSphere patch (in some form) is applied to >git: >http://bugs2.winehq.org/attachment.cgi?id=29947 > >You can see two patches for tests here: >http://bugs2.winehq.org/attachment.cgi?id=29948 >http://bugs2.winehq.org/attachment.cgi?id=29949 > >The only caveat is that we must first implement D3DXFVFFromDeclarator at the >very least, as the >CreateIndexBuffer call must be (to my understanding) supplied an FVF. > >Thank you >Misha > Fyi, I am currently taking a break from working on these patches as I feel like it may take some more time to get the D3DXCreateSphere tests into git given that it has been through so many tries. However, I hope that once this hurdle has passed I can once again continue to work on these implementations at full speed. Thank you Misha -- Sent from my Android phone with K-9 Mail. Please excuse my brevity.
