I'm making some clothing for Oblivion, and am trying to get the specular to work right. From what I can gather, I need to get my spec map into the normal map's alpha channel. I'm using GIMP for this and can't seem to figure out how exactly to do this. Every bit of info for this that I've found is supremely unhelpful.
Replies
EDIT: Okay I've found 2 ways of getting my gloss map into the specular maps alpha channel, but not without it actually affecting the images alpha and mucking up the transparency of it all. How do I actually put an image into an alpha channel without it uh, actually affecting the images alpha?
If you use a format like PNG, the alpha can change the RGB channels. But a format like TIFF keeps it completely separate.
Thanks for the info though, I'll see if there still is a GIMP crytiff plugin.
EDIT: Bugger, I can't find a crytiff plugin and the .dds one I found does not work No fancy shaders for me! XD
Now you have to thumbnails in the layer view, the left one is the RGB view, the right one the alpha channel. Press left mouse button on each of this thumbnails to switch between them. The selected one get a white border (not easy to recognise) and all operations like painting, copy'n'paste ,color manipulation will be done on the selected one.
Clicking with the left mouse button on the right/alpha view while holding eiterh Ctrl or Alt, you can only view the alpha channel (green border) or disable the alpha channel (red border).
There are several ways to copy an existing image (greyscale) into the alpha channel, here's one:
1. open your normal map.
2. add transparency to layer.
3. load specmap as new layer. (should be of same size to simplify the process).
4. left click on spec map layer to select it, then Ctrl+C
5. left click on alpha channel of normal map to select it, then Ctrl+V
6. press the anchors icon
7. delete spec map layer
8. export
That's exactly what I did but when I do so the image becomes a semi transparent mess, although by the sounds of it that's normal then? If so then otherwise I can't export from GIMP in a suitable format since this didn't work in targa (and for some reason regular old tiff images don't work in the Cryengine 3)
What the CryTiff Plugin in Photoshop does (or at least how it can be best described) is save a Tiff File and then call the rc.exe (to be found in bin32/rc/) with the files location and the desired Texture Parameters. Check the rc.ini to find the different Names you can call. The RC then writes the necessary conversion data into the Tiff File-Header.
Note: The CryTiff Files are not in any way further compressed, damaged or corrupted by this operation. They are merely including an additional node now for the next step.
In the next step, upon calling the Texture, the Engine reads the CryTiff and checks if a DDS version already exists for that texture. If not (or the DDS doesn't fit to the CryTiff's Data) it will compile a DDS which will have the ideal settings for that specific texture format.
An easy example for this is the Displacement Map: A displacement Map is supposed to be just Greyscale. If you would just save a standard DDS it would not have an Alpha Channel and horrible compressed RGB channels. Now, if you look into the DDS specifications, there a different ways to compress DDS textures. One of them is to save only the Alpha Channel using full 8-Bits and killing RGB.
You could in theory save all your textures as DDS yourself and choose your own compressions, but a) that would take a ton of work on your side and b) you would have to figure out all the DDS settings your self and c) you have to make backup uncompressed copies of all your textures.
What CryTiff does is, it saves for you an uncompressed Tif File, using the RC's presets you always get the best compression method easily (and if you were to develop for consoles, it would pick the best compression per Platform automatically) and last but not least: It has nice features like "Bump2Normal", which can save you additional trouble.
When exporting your image, just add the necessary file extension, ie. .tga for targa, .png or .jpg. When exporting you often get an message, that the target fileformat is not suitable, this is because most file formats do not support layers etc. Just ignore it.
If you want to keep the layers etc., then save the file in .xcf or .xcfbz2 (compressed version)
Thanks for the replies
did you read my post at all?