Hard Light Productions Forums
Modding, Mission Design, and Coding => The Modding Workshop => Topic started by: Lynx on December 30, 2003, 09:50:35 am
-
I recently discovered this proggy and I'm trying to UV-map my model of a Banshee light fighter from the Union of the borderworlds with it.
There are 6 views of the model, but it seems that especially on the top and underside view some faces are missing or faces from the underside are shining through. And I want to know if that's normal or if there's way to fix it before I continue. I don't want to put my heart blood into ot just to discover that it doesn't work in the end. Here's a picture; as you can see, on the top and lowere view parts of the 'arms' are missing. Just want to know if it's normal.
(http://scifi-3d.de/wcsaga/Lynx's stuff/BansheeUV.jpg)
Thank you for answering my questions in advance, here's your reward, some screenies of the model :)
(http://www.scifi-3d.de/wcsaga/Lynx's stuff/Banshee1.jpg)
(http://www.scifi-3d.de/wcsaga/Lynx's stuff/Banshee3.jpg)
-
:/ tutorials could help a lot, expecially when you don't have a clue about how the proggy works.
the red polys are not different views of the same model, actually they are the uvcoordinates you are going to manipulate.
the first image is what it comes out from applying a cubic uvmap to the whole model, it just projects from all the ortoghonal 6 point of view (top/bottom/left/right/front/back) and choose what faces are best mapped from one point of view instead of the other (eg: from what point of view the face projecton has a wider surface), then it organize all the faces depending by the chosen point of view, and group them together, with the result you have in the pic (so face in a group is not in other groups).
-
What's gonna happen is that certain faces (the red outlines) will be layered over others. Best bet is to go through and seperate these faces. For instance, it looks like your cannons on those pods will have their UV map space layered over the space for the side of the nose. Find these and pull them away.
Next step is to find all mirror images, and I mean all! Flip them and arrange them over one another, so that you only need to texture one side and it corrosponds to the opposite. Not only does this speed up texturing but it also allows for better looking textures.
Third, spend a few minutes arranging all your new groups in a very tight perfect square. After you have done this increase their scale so that they completely fill the UV map space. AHAH! You are done! Well, after you save the map that is.
Have fun! (And after come texture my models, because I am just plain horrible at 2d art.) :)
~Beowulf
Edit: What exactly is this "Lithium Unwrap" of which you speak? ;)
-
save the UVmap, and make large coloured boxes over the different views, yellow for top, red for bottom, etc.
Then use that as an UV map to see if it works or not. That's the way I do it.
You'll notice missing faces etc right away.
-
This model has the potential to be extremely difficult for a beginner to uv map. The problem is that you have a lot of side facing polygons that will end up being inappropriately grouped together when you a generic box mapping. Such as the outer side polygons on the left engine and the inner side polygons on the right engine. The best approach is probably not to use a generic mapping at all. Use the preview window in wireframe mode and pick polygons using the mouse pointer that you want to group together, choose a mapping, and move them onto a new texture. For example, pick out the outer facing polygons on both engines and choose a side decal mapping. Continue until all the polygons are mapped.
The "gap" in the arms isn't really a gap. It just means that those polygons were (incorrectly) mapped along with a different group of polygons and are somewhere else on the uv map. If you go into the preview window and select the offending polygons on the actual model, they will be highlighted on the uv map in red. So you can see where Lith has put them. Unless your model is extremely simple Lith can often map polygons on opposite sides of symetrical models differently. You end up having to find and correct those mistakes. Another good reason to pick out the polygons you want to group together yourself.