phirekid718

Thank you. I thought it was just me. It's been nearly a month and the only way I could figure to get this working was to go with the ToyMaker's implementation of mesh animating. Although it uses software skinning, It works. Has anyone ever taken so long on one task? Why or why not? Things like this make it seem like I don't have what It takes to make games. Again, thanks.

I ended up choosing to scrap xact and go with directsound as presented in the Toymaker's website. I may regret changing to this platform but for now it does exactly what I need it to do and the debug output displays as expected. I wish I didn't have to go through the process of changing the whole sound engine, but I can't see any other way, especially when the examples in the directx sdk such as "xactgame" have the same problem: loading up debug output at the start but not after the app closes. I guess I could have disabled the sound and continue building, but that would have been cheating for me. I feel that this problem is an indication of a much greater problem, but I don't know what. Any feedback would be appreciated.

I found that it has something to do with the XACT code that I am using from the directx sdk tutorials. I can't figure it out yet. It seems that the threads in the sound engine get in the way of the debug output from the other threads, if that makes sense.

I don't understand how someone gets away with publishing code that has memory leaks, leaving me to find them. I emailed the author, and I don't think I'm going to get a response.

I tried switching to the code found below, which is similar, and doesn't have any leaks. But the problem is that I am having trouble making it so that my character can hold something in his hand. With the textbook code, the application shows the character holding (a sword in this case) and moving it around just fine, but I can't find the memory leak after all this time. With the code in the article implemented the character can still make the motions, but the sword is not there.

If anyone has some insight please help. Maybe there's another resource on this subject? Thanks.

http://www.gamedev.net/page/resources/_/technical/game-programming/working-with-the-directx-x-file-format-and-animation-in-directx-90-r2079

Thanks for your help. I tried it but unfortunately that,s not the issue. If I do not load the mesh using the hierarchy (last line in the same init method) then there are no leaks.

This is the output when the mesh is loaded (From the directx debug runtime).

Direct3D9: :====> ENTER: DLLMAIN(6023d9a0): Process Detach 00001b20, tid=00001f54
Direct3D9: (INFO) :MemFini!
Direct3D9: (ERROR) :Memory still allocated!  Alloc count = 143

...

Direct3D9: (ERROR) :Total Memory Unfreed From Current Process = 2388080 bytes

It also shows resources that are not destroyed when running pix. So there is definitely a sneaky leak somewhere, again most likely in the bonealloc class.