MetKiller Joe
January 10th, 2009, 02:26 PM
I've been debugging a map for about a week now.
Errors kept popping up (random overlaps, degenerates, you name it, and they only moved when I solved one problem).
Finally, I got so frustrated that I went back to Max 7 from 2009 and tried using blitzkrieg. Now, I finally got what I wanted, a "VILS", which I remember as being a pointer to a error which doesn't allow the map to compile.
The problem is that tool is telling me that the error is at the origin, and there is nothing there, no geometry, no nothing.
So, anybody have any idea what is going on?
I am almost positive this is causing all other errors. There are no non-coplanars and other random things. I've done a sweep of this map at least a few dozen times to make sure of that.
Edit: I think I am getting warmer. I deleted the box surrounding the map (because the actual map and skybox are separate [think of a cylinder within a box]), and have not gotten the error. Though I'm not going to hold my breath because there are still some errors being reported (thankfully none at the origin).
CodeBrain
January 10th, 2009, 03:19 PM
This happened to my Test map I had. It wasnt this specific error however.
When I imported the .wrl file, it said where the piece was (It was a open edge error.) I said "Okay" and deleted the .wrl model. However, after I did that, I was unable to find it. I tried Mesh, Poly, to no avail.
The solution I unfortunately see in this (and mine) is to start over from square 1. Unless someone else knows how to fix it.
Remember, wait until others have posted before you use my solution, for it might not be the best one.
=sw=warlord
January 10th, 2009, 03:24 PM
Try using the "unhide all " tool in poly edit mode.
MetKiller Joe
January 10th, 2009, 09:27 PM
OK, at this point, my first assumption, that the file was simply corrupt, is incorrect. I separated the solid-surface geometry from the organics to try and pinpoint the problem. Sure enough, the solid-surface geometry was the problem, which means this error is geometry related.
Restarting from square one is not an option.
Some symptoms:
error "VILS", an error which will not let the level compile, is at the origin
this "VILS" will show up as a degenerate triangle in tool (though it might come up as something else, but this is what I got)
Possible/solutions causes:
"VILS" is the result of a simple degenerate triangle (easy)
"VILS" is the result of a grouping of degenerate triangles which are easier to rebuild than to isolate each degenerate triangle (because, remember, tool is not giving you the location of these things) (medium)
"VILS" is caused by a grouping of chunks, which MAY OR MAY NOT BE CONNECTED. For example, think of a house. If this origin anomoly appears, the house is at fault obviously, but if you take away the house, the error disappears; if you take away the smoke stack (and keep the house) it the error disappears. Meaning that the error is caused by both entities being connected in space (no one entity is resposnible). Take the less complicated chunk/entity and delete it, then rebuild. Hopefully this will solve the problem.
My method of isolating the problem chunks:
I've gotten rid of the errors chunk by chunk. The pattern I see emerging is that these errors are extremely elusive, but catch-able.
If you can isolate them to a chunk, you look for degenerate triangles. These degenerates are an order of magnitude more screwed up than the regular ones (so look where, on the editable mesh, the most triangles appear in a dense cluster), which may be why Tool is telling you to go to the origin.
The less of these degenerates there are (as you delete them), more seem to pop up, but I don't think this is the case. What I think is happening is that tool is dividing this "origin glitch" along these degenerates, so the fewer degenerates, the more unsolvable glitches you get. Now, hopefully, if I delete the degenerates, it won't just tell "No." (because if this division plays out to OVERHEAD/0 (degenerates). It might do something funky like give me an infinite number of glitches in the .wrl (causing tool to crash in the process I'm betting), it might tell me it cannot do an operation, or *hopefully* tool will not catch anything and the glitch will disappear.
I'm still not going to hope for anything until I get all of the errors and kinks out (open edge errors and a few degenerates that tool is now giving me on the actual geometry), but my deleting some faces and rebuilding some, I have managed to debug the origin error. I also urge any beginner who does get this error, to just start over. Myself and T1xAnton have been working on this map for a couple of months, so failure was not an option. It took many hours, but I'm sure anybody on a team could do this within a half a day or two with the prescribed method.
Powered by vBulletin® Version 4.2.5 Copyright © 2024 vBulletin Solutions Inc. All rights reserved.