View Full Version : Trying to add third group with CTU
omegastarr82
4th Jul 2011, 06:55 AM
I'm trying to make an outfit with three groups using CTU, Milkshape, MorphMaker and MorphMatcher. Groups 1 and 3 work perfectly, group 2 gives me the morph fail spikey stuff, see pic. The steps I took are as follows, Changed base meshes to my liking, cloned af halter with ctu, updated meshes, used morph matcher to make morphs from base meshes, then used morph maker. Now with morph matcher I didn't use the renumb feature since I didn't add anything to the meshes, I only took away and moved around. Should I renumber them? If so at what starting number? I've never understood the point of the renumbering. Did I miss something? Could it be that I'm using different gender meshes? The two that work are male meshes, the female one doesn't. The groups are as follows, group 1 is the top, 2 is the skirt, 3 are the legs.
I tried this yesterday too and got the same results, I thought I may have did something wrong so I started over today from scratch and got the same thing.
Bloom, you're the go to guy for stuff like this, what am I missing here? I swear I wouldn't get anything done without your help.
Cocomama
4th Jul 2011, 09:39 AM
Sorry, I'm not Bloom.
How does your comment looks like? All parts/groups from the same LOD must have the same comment, so its not possible to have 2 male groups and 1 female group in one LOD.
omegastarr82
4th Jul 2011, 04:02 PM
I just checked the comments and they seem to be in order. What i meant by male/female groups is that the top and legs are from a male mesh and the skirt is from a female mesh, I don't know if that has anything to do with anything but I thought I'd mention it. (I didn't know you're on MTS also, good stuff)
I got it working. I renumbered everything and it kinda worked, the morphs got all mixed up somehow, PEBCAK. I eventually started all over again and merged it into one group. Works now.
Cocomama
4th Jul 2011, 06:22 PM
I think you should renumber.
Start at 5000 for a fullbody like this.
Import Lod 1 with all its morphs, take note of the number that will be given to you after renumbering with Milkshape plugin 'Sims3 VertID AutoNum V0.16 -by Wesley Howe' . (eg 5600) Export all.
Load all of Lod 1-1 with morphs and this time use the starting number you noted (5600) take note of the next number that will be given to you (e.g. 5650) Export all.
Load all of Lod 1-2 with morphs and the starting number you noted (5650) take note of the next number that will be given to you (e.g.5700) Export all.
Load all of Lod 2 with morphs and start with the last number you noted (5700) take note of the next number that will be given to you (e.g.6300) Export all.
etcetera untill you renumbered all.
Some say you can start counting here with 5000 again, but I never did that.
If renumbering does not help, you'd better attach the file here for Bloomsbase to look at it.
BloomsBase
4th Jul 2011, 06:23 PM
The distortions were most likly caused by the leg and skirt having identical vertID's, you should renumber.
Edit,
Coco beat me to it and she is right :)
omegastarr82
4th Jul 2011, 06:32 PM
The distortions were most likly caused by the leg and skirt having identical vertID's, you should renumber.
Edit,
Coco beat me to it and she is right :)
Thanks guys, I did that but was still having problems, different problems, the renumbering worked. I ended up regrouping the three meshes into one group and that worked flawlessly, I still renumbered everything. Which I only tried on a fluke because I thought you couldn't do that? Or is that only with TSRW?
CmarNYC
5th Jul 2011, 06:15 PM
I don't know of any reason you can't combine groups - maybe Bloom does. Updating (and renaming) MorphMatcher to include a mesh combine function is on my list so I'll be looking into it more.
Cocomama, it's possible to have multiple meshes in a lod with different comments. The top and bottom below were each done by Frankensteining two meshes and setting the shader/EmbeddedType to different values. I think the pants also have different tagvals in the two meshes. So far so good but I'm still testing - maybe there's problems under certain circumstances?
omegastarr82
5th Jul 2011, 06:37 PM
Well I know with TSWR it makes the morphs go all spiky. This is my first full CAS meshing project using only CTU, so I'm still experimenting. Is that transparent clothing? Can we do this now? That's pretty awesome.
CmarNYC
5th Jul 2011, 07:30 PM
I've never used TSRW, but I know it has its own way of doing things and maybe it just doesn't handle combining meshes correctly.
Heh, I may be the first to do transparent clothing. At least I recently stumbled on a method that works and don't know of anyone else who's done it.
Cocomama
6th Jul 2011, 09:24 AM
About the comment thing, what I wanted to say is, that it is not possible to have to different skeletons 2 male and 1 female part in 1 LOD.
Appearently it is possible to have different texture settings.
Bloomsbase has been working on transparent clothing for a very long time already Cmar, he will be very pleased to see it is possible!!!
It's awesome, it looks like you've made a true alpha mesh!! :gjob:
Tutorial please :heyhey:
CmarNYC
6th Jul 2011, 12:50 PM
That makes sense; if the skeletons are different the animations wouldn't work.
I'll get a quickie tutorial done for the CAS section. :D
BloomsBase
7th Jul 2011, 02:45 AM
I don't know of any reason you can't combine groups - maybe Bloom does.
Combining morphs is just not possible, unless the parts you are using are already divided on the uvmap by EA.
Like the arms/sleeves/hands on some GEOMS
I never really understood why althoug Wes tried to explain here:
err, cant find it....
Alot of people already had problems when trying to regroup diffrent morphs, the reason i advised to use outfits with 2 meshgroups in it.
If your lucky it works but often gives you messed up outfits.
A
Bloomsbase has been working on transparent clothing for a very long time already Cmar, he will be very pleased to see it is possible!!!
Tutorial please :heyhey:
Getting the parts transparant isn't the problem, i just can not figure out how to recolor them :(
CmarNYC
7th Jul 2011, 04:55 PM
By combining morphs, do you mean regrouping two morph meshes / groups? I don't understand why that wouldn't work either, or why them being divided on the UV would affect it. If you find that discussion with Wes please let me know.
BloomsBase
8th Jul 2011, 07:51 PM
yes, its been a pain for many.....i simply refused to believe it wasn't possible at the time(we never had any problems with it in sims 2, so why doesn't it work in sims 3.)
Its because alot of EA's base meshes and morphs are somehow build diffrent.(you must have seen the diffrence in vertice count on base and morphs?)
Wes his exporter even refuses to export those combined morphs.
Maybe your Morphmatcher does fix some settings i dont know?
You can do a few simple tests if you like by combining tops and bottoms into a fullbody outfit, 90% of the times you get a messed up morph in CAS
If you can regroup them in Milkshape(and export) in the first place.....
CmarNYC
8th Jul 2011, 11:12 PM
Strange. I think I remember some issues like this with the face, teeth, eyes, etc. meshes - seem to recall I had to make my ear morphs from scratch because the game ones wouldn't import. As soon as I can get back to the MorphMatcher project I'll get into this. All the morphs contain is the usual header data, the differences in position and normals, and the same face list as the base mesh so it's possible to fiddle the data and make any morph match any base mesh as long as the vertex IDs match up, at least in theory. This is what MorphMatcher does now. It should be equally possible to take the data from two base meshes and their morphs and write out one combined base and the combined matching morphs, again as long as the vertices are numbered correctly.
BloomsBase
10th Jul 2011, 12:37 AM
According to Wes this isn't possible altho alot(including me) were stuborn and tried anyway.
You must have seen this thread: http://www.modthesims.info/showthread.php?t=408803
Shows a few tops and bottoms that are giving problems.
vBulletin v3.0.14, Copyright ©2000-2013, Jelsoft Enterprises Ltd.