On Thu, Aug 4, 2011 at 10:27 AM, Michael Mc Donnell michael@mcdonnell.dk wrote:
On Thu, Aug 4, 2011 at 8:48 AM, Michael Mc Donnell michael@mcdonnell.dk wrote:
On Wed, Aug 3, 2011 at 2:39 PM, Michael Mc Donnell michael@mcdonnell.dk wrote:
Anyway you're right it's probably a good idea to stay clear of them in order to avoid potential compiler issues. I've updated the patch to implement UDEC3 and UDEC3N using shifts and masks instead of bit-fields. It will still only work on little-endian machines though. Does it look ok?
I've rebased my D3DXWeldVertices patch. Git was having trouble applying it after Alexandre committed my ConvertPointRepsToAdjacency.
Sorry for the noise. I accidentally moved some of the code around during the rebase merge. I've moved things back in the original order.
Here's yet another update for D3DXWeldVertices. I'm working on writing a test for CloneMesh and re-using some of the parts from the D3DXWeldVertices test, which is why I keep finding these small bugs :-)
In the test I had forgotten to replace all fabs with fabsf. I've also changed the comparison test to find the maximum absolute difference for FLOAT1-3. I've finally added a comparison test for FLOAT4.