New Clothing Body Style Converter Beta v0.89f (10-26-2014)

[ger4]

OK. This is getting silly

 

Clothing Converter Release Candidate 0.76

 

Download:

 

Portable Package 76 Upgrade (upgrades Portable Package v.70, v.72, v.73, v.74, v.75 to v.76):  Convert Clothing RC_76_Upgrade.7z

Light Download 76 (scripts only. Recommended if you intend to use 64 Bit version of Python):  Convert Clothing RC_76_Light.7z

 

 

ChangeLog

 

v.75 to v.76

1. Target nifs with complete skeletons were having their skeletons improperly flattened and transformed. This bug has been resolved.

WTF is happening here 

[gerra6]

WTF is happening here

[xarathos]

Lattice successfully generated using .73. Just launched a test conversion using .75 on the UNPBase->7BBombshell lattice. It'll be interesting to see what happens with these - I'm trying it out on the "Hmm What To Wear" set of meshes. I had to give them up ages ago because the skirts didn't fit right.

 

First batch has completed seemingly without incident, so I'm gonna do the second and then try loading them in game to see what happens.

 

EDIT: Previewing in Nifskope, I'm noticing some odd distortions in a few meshes that feature things like rings and whatnot - I think similar issues have cropped up in other conversions throughout the thread so I'm gonna skim back through later and see if I can find the solutions that were offered to them the first time xD Still no errors or anything like that, though. Of course as I said, yet to try them in game. Conversion is still running.

 

EDIT(Again): This sort of distortion.

 

 

 

 

 

 

 

[gerra6]

EDIT: Previewing in Nifskope, I'm noticing some odd distortions in a few meshes that feature things like rings and whatnot - I think similar issues have cropped up in other conversions throughout the thread so I'm gonna skim back through later and see if I can find the solutions that were offered to them the first time xD Still no errors or anything like that, though. Of course as I said, yet to try them in game. Conversion is still running.

Would you be willing to send me the From and To files you used for that lattice, as well as the bikini nif with the ring problems. I'd like to see if I can figure out what's causing the distortion.

[xarathos]

Would you be willing to send me the From and To files you used for that lattice, as well as the bikini nif with the ring problems. I'd like to see if I can figure out what's causing the distortion.

 

Done. Hope it proves useful.

[zippy57]

Not sure what information you need for error reports, but after running it two or three times I managed to get a screenshot of the error message before the window closed itself.

UV Verts Searched 2250 of 4331
UV Verts Searched 2501 of 4331
Traceback (most recent call last):
  File "PythonPortable\App\Scripts\kgtools\make_lattice.py", line 205, in <module>
    Main()
  File "PythonPortable\App\Scripts\kgtools\make-lattice.py", line 171, in Main
    good_targets += uvSearch(cs, to_mesh, from_mesh, act_p_vert = search_res, ctx_p_vert = lattice_res, face_search = True)
  File "PythonPortable\App\lib\site-packages\kg\search_util.py", line 1816, in getVertexInfluence
    v1, v2, v3 = face_.verts
UnboundLocalError: local variable 'face_' referenced before assignment

This was creating a lattice from the Type 3 body from here (specifically HD Base.nif) to the BNB body here. Turning off UV Search allows the process to complete successfully. I'm fairly certain Type 3 textures can be used with BNB but I don't know if that means the actual models are mapped the same way and I honestly have no clue how to check that.

[gerra6]

Not sure what information you need for error reports, but after running it two or three times I managed to get a screenshot of the error message before the window closed itself.

 

This was creating a lattice from the Type 3 body from here (specifically HD Base.nif) to the BNB body here. Turning off UV Search allows the process to complete successfully. I'm fairly certain Type 3 textures can be used with BNB but I don't know if that means the actual models are mapped the same way and I honestly have no clue how to check that.

Thanks for the bug report. It feels good to crush bugs.

 

Which version are you running? I believe that particular bug should be fixed in the 0.76 update located here: http://www.loverslab.com/topic/27557-new-clothing-body-style-converter-release-candidate-v072-3-30-2014/?p=735819

 

I'm just vomiting out versions today, so I figure I'll wait until I'm *sure* that I'm done iterating for today before updating the front page.

[xarathos]

The good news re: that last batch of conversions? The vampire skirt from the "Hmm What To Wear - Dawnguard" parts converted perfectly. I tried like six times to convert that manually and gave up because I couldn't get the bone weights to carry across properly AAAGH

 

But this version handled it like a charm. No clipping that I can see, no weird stretching issues, incredible. A++

[gerra6]

That is damn good to hear.

[zippy57]

Whoops. I should take my own advice. That was using v72; using v76 fixed the error.

[orgs1n]

Hey, I used v76 to convert Aradia's Lace Dress. Works perfectly... except for the legs.

 

 

Judging from your first post, Padding would seem the solution to this. But how is that setting accessed? The dialog looks like this:

 

 

When clicking Customize Mesh, I can see the button depress, but no dialog or extra settings appear.

[gerra6]

It's a bit weird.

 

If customize mesh is enabled, then once the tool loads a nif, it will open a menu that allows you to customize how individual geometry blocks (meshes) in the nif are processed.

 

Basically, just enable it, click OK, and wait until the menu appears. Once the nif is loaded and preliminary processing is complete, the tool will open the menu.

 

Note: the customize menu will open for every nif processed by that run of the tool.

[orgs1n]

Padding does improve on things... These are at 0.1 and 0.18 (all).

 

 

.. But even at 0.18, some gaps remain and the stockings visibly start to 'float'. Are there any other settings I could try?

 

(Edit: for the record, Growth bias 1.1 didn't seem to make a difference.)

[ger4]

Padding does improve on things... These are at 0.1 and 0.18 (all).

 

padding.0.1.pngpadding.0.18.png

 

.. But even at 0.18, some gaps remain and the stockings visibly start to 'float'. Are there any other settings I could try?

 

(Edit: for the record, Growth bias 1.1 didn't seem to make a difference.)

Probably really needs some kind of clipping detection.

[ger4]

Quick question about a future library of UV maps. Would it allow a HGEC UV to convert to say, a UNP skyrim map and vice versa? Or will it just be for UV maps from the same game (i.e. HGEC --> Robert's Female)?

[gerra6]

Padding does improve on things... These are at 0.1 and 0.18 (all).

 

.. But even at 0.18, some gaps remain and the stockings visibly start to 'float'. Are there any other settings I could try?

 

(Edit: for the record, Growth bias 1.1 didn't seem to make a difference.)

Not really.

 

I'm developing a framework designed to automatically detect intermesh clipping and automatically repair it. But the collision detection in the latest version of the clipping removal tool is a spectacular failure (I think there is a very subtle bug somewhere in the collision code that I haven't yet tracked down).

[gerra6]

Quick question about a future library of UV maps. Would it allow a HGEC UV to convert to say, a UNP skyrim map and vice versa? Or will it just be for UV maps from the same game (i.e. HGEC --> Robert's Female)?

Technically, Clothing Converter is part of a larger library with a single goal in mind. I want to make it possible to automatically interconvert Oblivion, Fallout, and Skyrim nifs with each other, and to interconvert any clothing made for one body style with any other.

 

We're about halfway there

 

Pose Converter can automatically adjust the bind position and crude shape of a nif between Oblivion, Fallout and Skyrim base shapes. (I need to improve the finger positions a bit, but I haven't yet because finishing it requires a whole lot of manual posing in Blender)

 

Mesh Rigger can automatically convert the skinning and rigging of a nif between Oblivion, Fallout, and Skyrim (and any other arbitrary rigs)

 

Clothing Converter can automatically convert clothing and armor between any two bodies that share the same UV map. (still a WIP, but getting damn close)

 

The next piece is to build a UV library that would, on the back-end, map any previously analyzed UV map onto a common generic set of coordinates. If my idea works, it will be possible to use that library of analyzed maps to do two things. The first stage would be simply to allow Clothing converter to use the UV search function on meshes with different, but previously analyzed, UV maps.

 

The second stage (more complex than it sounds), will be to write a tool that automatically converts the UV map from one standard to another. The tricky part is that the Nif format only stores a single UV coordinate per vertex. So, in order to move a UV seam, I need to do two things. I need to remove the old seam by removing the vertex doubles along the old UV seam. Second, I need to cut the mesh along the new seam and generate entirely new vertex double pairs there. It all should be doable with enough work.

 

After that, my plan is to build a material library that can be used to convert shader, texture, and material blocks between Oblivion, Fallout, and Skyrim types.

 

Finally, if I ever find a Python .DDS library that I like, I'll need to write an automatic tool to convert textures between the Oblivion, Fallout, and Skyrim formats as needed.

[Monsto Brukes]

Oh...I also have a question for anyone using the latest version of Clothing Converter.

 

Clothing Converter has a bit of a design flaw...not a big one, but one that has been bothering me a bit for the last few days.

 

I designed the 'Search Distance' as a hard edge to the clothing conversion.  Up to that distance, the searching vertex will receive the full brunt of whatever transform was coded into the lattice surface at the nearest point to the searching vertex.  .00001 units past the 'Search Distance' the vertex receives no transformation at all.

 

This...makes me grumpy.  So...I'd really like to add some form of edge smoothing.  The idea is that, as vertices near the upper limit of the search distance, the percentage of the transformation that they receives gradually drops off until it reaches zero at the search distance limit.

 

Any thoughts?  Good idea, bad idea?  What sort of implementation would be both useful and user friendly?

 

One idea I had is to add a second distance slider, one whose value is added to current value of the Search Distance Slider.  So, anything less than Search Distance receives the full transformation coded into the lattice.  Anything between Search Distance and Smoothing Distance receives a reduces version of the transform, dropping off as the distance increases.

 

Do some gradient math on the edge with a "Transform Edge Smoothing" check box, then add a value slider that goes from 0-100% gradient. The only problem with this is getting a sense of how it's going to look without sitting thru the conversion.

 

It's kind of a Pandora's box, tho. once you introduce any kind of gradiation, SOMEONE will INEVITABLY start hollering about grad type choice... Linear won't be good enough by itself. They'll want log, reverse log, sine, etc etc. A bunch of shit that nobody will ever use.

 

Altho I think some kinda sine is probably the best choice as it avoids abruptness.

[gerra6]

Sine is probably the way to go, maybe a BSpline implementation of sine with zero dropoff out to a certain distance.

[Nightwynd]

This is AMAZING!

 

I've tested it with the following.

 

DMRA to HGEC AA cup

DMRA to HGEC K cup

HGEC B cup to DMRA Guts

 

My only problem was the last conversion, as the breasts clipped through the top.

 

 

[gerra6]

It's a fun tool.

 

Clipping is probably going to be an issue until I finish development of Clipping Mender.

 

In the meantime, enabling 'Customize Meshes' and setting padding to .1 can help. Setting the magnify movement and/or growth bias can help fine tune the movement for a given mesh. Unfortunately, for some reason 'Growth Bias' and magnify movement doesn't appear to be doing their job, but I haven't spotted the bug in the code yet.

 

I'm also working on some back-end improvements that should hopefully resolve a few of the remaining sources of artifacts.

[myuhinny]

One thing you can do on minor clipping is to tweak the vertices of the piece in nifskope. 

[gerra6]

Clothing Converter Release Candidate 0.78

 

This update brings several improvements to the quality and performance of Clothing Converter.

 

Note: Please generate new Lattices for use with 0.78. Older lattices will continue to suffer from pre-existing artifacts.

 

Download:

 

Portable Package 78 Upgrade (upgrades Portable Package v.70, v.72, v.73, v.74, v.75, v.76 to v.78):  Convert Clothing RC_78_Upgrade.7z

Light Download 78 (scripts only. Recommended if you intend to use 64 Bit version of Python):  Convert Clothing RC_78_Light.7z

 

 

ChangeLog

 

v.76 to v.78

1. Several potential sources of artifacts have been identified and resolved

2. Quality of lattice generation has been improved.

3. Speed of lattice generation for meshes with compatible UV maps has been dramatically improved

4. New option added to Make Lattice tool: Skin Only. When enabled, this forces the Make Lattice tool to match only meshes with skin material types. This resolves transformation bugs caused by stray clothing or Fallout Meat Caps.

 

Outstanding Issues.

 

1. Some sources of transformation artifacts remain. I believe I've narrowed down the causes and am working to resolve them.

[xarathos]

Ooooo fun a new version to stress test. xD

 

Fresh lattice generated; running conversion now. Will report back with results. ^^

 

EDIT: Have I got good news for you, sir!

 

 

 

 

 

 

 

 

Huzzah!

[gerra6]

***Uh Oh***