Hi folks. I’ve been trying out baking various parts of meterials in the hopes of figuring out what’s safe to bake in all situations, and what can only be baked if certain conditions are met (like if shadows are being cast on an object or if the lighting is changing colour and so on).
From experimenting with the bake settings, i was fairly sure I’d hit upon something that was safe to bake under almost any circumstance: the diffuse colour of a material. Obviously, baking just the diffuse colour and plugging it into a Diffuse Shader only cuts out a much smaller portion of the rendering time (in the cases of just simple materials that being zero time), but as far as I could tell, the final image was always identical to an unbaked material which may require a little more rendering time for each frame. When I tried to apply this to an object I would like to eventually use in a scene, however, the baked render comes out darker than the unbaked version and I have no idea why. Can anyone explain to me where I’m going wrong? I’ll I’ve made an image that compares the two versions of the same object, but I can’t upload the Bled file it comes from 'cause the uploader just locks up any time I try to upload it.
Basically the grouped node creates the colourised pattern and then puts it through the same diffuse and glossy combination of nodes shown in the non-grouped version below the grouped node.
That’s weird. I wouldn’t have expected the file size to be a problem; the whole file is just the pillar and a slightly higher poly duplicate of it with a material. I’m minutes from going away for the week, but I’ll try it when I get back. Wouldn’t the file uploader just say the file’s too big, rather than completely locking up, though?
Ok, apologies for the insane delay in trying out your suggestion, but it didn’t seem to have an effect. I tried rendering with the bake saved as a 32 bit float EXR format image, and also with it not saved in any format just to avoid any potential data loss (regardless of how tiny it may be). I couldn’t see any difference from the PNG formatted baked image.
I realize the difference in the baked vs. unbaked pillars is small, and I’d be happy to continue as it is, but the real reason for me asking about why there might be a difference in the image at all is that I take it as an indication that there’s probably something I should know, but don’t. I worry that, whilst the difference is tiny in this very simple example, I might continue onward to build bigger scenes where the difference may become much more obvious and ugly, because I’m missing some bit of knowledge. With such a basic file, though, I can’t even begin to imagine what that might be.
I’m fairly sure I’ve managed to sort the file out now so that I can upload it and people can have a look at it for themselves. pillar-copy.blend (2.23 MB) It seems that th reason I couldn’t upload the file was because of a normal map that I had baked and packed into the file. As far as I know, it shouldn’t have anything to do with the problem because I just baked the normal map in preparation to apply it. I never got as far as actually using it, because I thought I’d better check that the diffuse bake was actually waorking as intended, before going further with applying the normal map. Any help would be greatly appreciated.
