Texture Issue

[ramirez 10]

So here i am again with a complicated problem. Im not sure why this is happeneing, but it is quite annoying and I do not know how to fix it. Basically the texture is getting warped, even though my uvmap is normal. I do not know what is causeing this, any suggestions on how to fix this? Ive already tried the following.

• Re-Mapping
  Moving the map around a little bit to see if anything changes
  repaired faces
  

http://i1222.photobucket.com/albums/dd494/RamirezLPD/arma2oa2012-03-0222-46-13-71.png

http://i1222.photobucket.com/albums/dd494/RamirezLPD/arma2oa2012-03-0222-56-11-90.png

http://i1222.photobucket.com/albums/dd494/RamirezLPD/annoying.png

http://i1222.photobucket.com/albums/dd494/RamirezLPD/Material156_co.png

Again, anything would help at this point. Thank you.

-Ramirez

Edited March 3, 2012 by Max Power

[max power 21]

Ramirez

Please watch your image sizes. I just noticed they are too big in your other topic, too.

§15) Do not hotlink images over 100kb (102400 bytes) in size

Do not link images over 100kb using the IMG tags to display an image in your post. If you wish to post an image larger then 100 kb feel free to post the URL instead of hotlinking.

And towards your question:

Are these faces quads or tris? Do they appear fine in the dx viewport?

[PuFu 4600]

i would say you got some sort of stretch right where the map is fucking up. As i cannot see (due to size of image) the close up where the UV is concerned, that is the only plausible explanation...

[max power 21]

I had a similar issue recently. It turned out to be a triangulation issue. If I just triangulated that quad the other way, the distortion disappeared... but it was only visible in buldozer and in the game, not in the viewport.

[PuFu 4600]

hm, that is a bit weird. AFAIK, direction of triangulation is important in the way light drops on the model, and hence can affect the normals map, not the diffuse one. Anyhow, i cannot tell how the geometry looks like at the culprit point from that image he posted.

[Sealife 22]

Maybe select the face in o2 where stretch occurs , click Uv and click button , select faces by selection in object see if there is some anomaly when vetices show in red and there position ?

[max power 21]

hm, that is a bit weird. AFAIK, direction of triangulation is important in the way light drops on the model, and hence can affect the normals map, not the diffuse one. Anyhow, i cannot tell how the geometry looks like at the culprit point from that image he posted.

Yeah, I agree.|

If I manually triangulated the mesh the wrong way, then I would see the distortion in the viewport too. I think it had something to do with the way Dx was triangulating the quads. I guess there was some distortion in there that was distributed evenly over both triangles if triangulated one way, but all on one and not the other if it was triangulated the other way.

Imagine a square, then pull one corner out away from the centre so it's a bit of a diamond shape. Now depending on the way you triangulate it, the distortion is on one triangle or on both.

[PuFu 4600]

Yeah, I agree.|

If I manually triangulated the mesh the wrong way, then I would see the distortion in the viewport too. I think it had something to do with the way Dx was triangulating the quads. I guess there was some distortion in there that was distributed evenly over both triangles if triangulated one way, but all on one and not the other if it was triangulated the other way.

Imagine a square, then pull one corner out away from the centre so it's a bit of a diamond shape. Now depending on the way you triangulate it, the distortion is on one triangle or on both.

I know very well what you mean. But then again, i see no point happening to a map applied to a surface, unless that surface has been deformed in the UV space...

[ramirez 10]

So any solution? I havn't really heard any solution, i can send the MLOD to one of you so you can take a look at it if you would like. That way you can see exactly what is going on, PuFu or Max, just send me a PM on how you would like me to send it.

[max power 21]

I know very well what you mean. But then again, i see no point happening to a map applied to a surface, unless that surface has been deformed in the UV space...

The only thing I can think of in my case (and what makes me think it's the same issue in his case) is the UV space is automatically deformed because the map he's using isn't square.

---------- Post added at 12:02 ---------- Previous post was at 12:01 ----------

So any solution? I havn't really heard any solution, i can send the MLOD to one of you so you can take a look at it if you would like. That way you can see exactly what is going on, PuFu or Max, just send me a PM on how you would like me to send it.

We would like to see a close up of the deformed area on your UV map.

Edited March 3, 2012 by Max Power

[PuFu 4600]

So any solution? I havn't really heard any solution, i can send the MLOD to one of you so you can take a look at it if you would like. That way you can see exactly what is going on, PuFu or Max, just send me a PM on how you would like me to send it.

now, can't give you a solution unless i/we know what the problem is in the first place...no?

I for one think that it has something to do with a problem on your uv unwrap. On the other hand, max believes it has something to do with face triangulation. Either way the easiest thing would be to provide a close up on the UV where the stretch occurs.

as a side note, while you seem to be using a 2^n by 2x2^n texture format, you are wasting a LOT of the UV space, so i would advice for a new (and better) unwrap.

[ramirez 10]

I fixed the problem, i don't know how i fixed it, but i did lol.