Haven't played around with the Cycles baker much but I'm willing to bet that issue means you're trying to bake to an image that is applied to one of your low or high poly objects.
First off, "OpenGL" or "Direct X" are not types of normals maps. Both OGL and DX can use either type, it just depends on how the shader is set up. It's an odd terminology that Substance uses for some reason. For instance, Toolbag 2 ran in both OpenGL (Mac) and DirectX (PC), but there was no need to use a special type of…
I don't really see any errors, what map type are you baking? Normals? Diffuse? Lightmaps? AO? Color Masks? What baker in what App? If you're referring to the weird reds in your padding, that's a Mental Ray issue in Max for example that cannot be solved, etc.
This can be due to the highpoly faces being backward, and can happen in some software if you mirror elements by using negative scaling. Make sure you have backface culling on in your 3D app so you can easily spot and fix backward faces. It looks like the resolution of the texture is very low in this area, so you have…
As a baker, I can gave you one tips, no baker ever would put a cake on a grid. 90% of chance of it breaking down, it's too soft, you can only put bread or viennoiserie on grid. no cake.
Every 3d package texture baker could bake from one UV to another or from one mesh to another. Could be done both ways actually. I don't use 3d max for a decade already but as far as I remember it could be perfectly done in render-to-texture dialog. Or any texture baker that support a texture on "hires" mesh. Substance…
Three possibilities: 1. Maybe there is not enough padding between the shells. 2. Maybe you did not use an averaged cage, so the corner is being missed in the bake. 3. Maybe the display is not synched to the baker. This is the most likely problem.
Splitting normals before baking is a good process. We have a bit about why this is important: http://wiki.polycount.com/wiki/Normal_Map_Modeling#Smoothing_Groups_.26_Hard_Edges Also some great examples here, along with info on why those extreme gradients are best to avoid…