PDA

View Full Version : Uv mapping in 8.2



peteb
02-03-2005, 04:19 PM
Just trying out the Uv-mapping in 8.2 and using the new mde where you odn't have to un-weld stuff. This seems good but I find it a bit messy that it leaves points behind from where it was joined. Am I just doing something wrong or is this just the way it is?

Pete B

jeanphi
02-05-2005, 03:59 AM
Select those points and use delete map.
The new UV in LW 8.2 are not full fonctionnal for now. They need more improvments. Others UV tools like UV copy/past, UV flip need to be update too.
They still need the unweld process to work correctly.

peteb
02-05-2005, 04:45 PM
Oh that's good I thought it might just be me being really crap. One other thing I've just discovered is that I can't convert uv to subpatch after they've been assigned. Unless of course I go back to make new UVs and then have to re-map them. It would be good to have a button where I can go from linear to subpatch to any other without having to remake the uvs. Or maybe there is a way and I don't know it.
Pete B

evenflcw
02-05-2005, 06:27 PM
You didn't have a look at Protons demo videos yet? That's exactly what he shows. Have a look at the Vmap Browser, they updated so you can change interpolation and sketchcolor (specific for UVs) anytime you like!

About the stray points. I think jeanphi is right. They need to tweak the workflow and redo most of everything dealing with UVs for this new stuff to work intuitivly. My guess is that those stray points belong to the neighbouring polygons of the polygons you selected to assign UVs for. How did those get in there? They got there thanks to what maybe should now be legacy functionality. If you think about it, previously, with non-per-polygon UVs, one 3d point could only have one corresponding UV point. That's why we had the whole unweld issue, no? As of 8.2 however one 3D point can have as many corresponding UV points as there are polygons connected to that point. So when you brought your selected polygons points into UV you did not only bring IT's corresponding UV points, but also that of it's neighbouring polygons. Logical? Yes! Intuitive? Not anymore! In other words, we got the reverse of the unweld problem. Instead of too few points, we now get too many. :) NT needs to redo UV assignments to assign UVs per-poly in some cases.

Please if someone else agrees OR disagrees with this theory, please say so. I'm not entirely sure, but would like to be before I report bugs and send in requests.

I'd say NT are at the verge of a really sweet UV system, it just needs alittle more work and some added tools.

EDIT: Fixed spelling mainly.

Dodgy
02-05-2005, 06:35 PM
I agree with this. :) You're not alone.

They do need to sort out their workflow now with this new system, in theory you don't even need to see the UV's as being discontinuous, this should all be transparent to the user.

evenflcw
02-05-2005, 07:04 PM
Ah good. Nice and quick answer too. I'll rest easy tonight :)

I reread the 8.2 addendum and it did say basically that it is still work-in-progress, so they are probably well aware of things not working that smoothly.

I could agree about it not being that important to see discontinuous points, all the time! Instead I think it would be more usefull if they only went red if I selected one of their buddies, so I can quickly spot who goes out with who (spare me all the nasty details). It also seems wrong to me that the first UV "instance" of a point isn't labeled as discontionuous (red) when the other instances are.

peteb
07-04-2005, 02:07 PM
He he, I just did a search on UV's as I'm having a problem and saw this one and thought it sounded interesting only to discover I posted it.

Anyway back on Subject, it seems to me that UV's are still a real pain in Lightwave. I've had to use MAya for my new job and I have to say in nearly everyway it sucks arse, well for modelling, I'm sure the animation tools are great? But the Uv editor although not the best I still think it's a **** of a lot better then Lightwaves. All I'm trying to do is apply a Planer map to a sub-d mesh using the sub-d uv option and the result I get is a complete mess. Is this something that most people get or just me?

Pete B