i tried with pixel bender 3d and hardcoding with the same result. My code for the vertex:
varagalVertexSource:String="m44 vt0, va0, vc0\n"
"sub vt1, vt0, vc4\n" +
"mov op, vt1 \n" +
"mov v0, va1\n";
_context3D.setProgramConstantsFromMatrix( Context3DProgramType.VERTEX, 0, camTransform,
Context3DProgramType.VERTEX, 0, camTransform,true);
Context3DProgramType.VERTEX, 4, Vector.<Number>([0.5,0.0,0,0]));
All i want is a translation of the vertex (sub...) after the multiplication by a matrix (m44...). The result : the program translate the point and then multiply the result by the matrix.
just to precise something else. If i use a second matrix just to tranlsate the vertex it works:
m44 vt0, va0, vc0
m44 op, vt0, vc4
that's why i think there is something i don't know about the opcode sub.
Giving your vertex program a shot in my test program and I get the correct order of operations (matrix multiply first, then translation). Do you happen to have a test that can show the issue for me? You can post it here or email if to me directly at firstname.lastname@example.org.
thank you but i use only opcodes and AGALMiniAssembler now. i deleted other files.