Wall Worm Forums

Quote from: wallworm on May 12, 2017, 01:20:59 PMI suggest avoid installing this update until after more software developers (including myself) have officially tested and given the green light on it.

Too late for that unfortunately. I haven't encountered any problems yet on 2016 though.

Oh that's what you meant by "loaded it". Ok cool. Well I await the multiple flex dominator fix with much anticipation then.  :P

Ah, such a simple solution. I've never used the Vertex Paint modifier before this so I still have much to learn.

I await the dominator fix with much anticipation.  :)

The stereo painting works 100% as it should. Just wish I knew how to turn the black/white coloring and flat shading off and keep the balance modifier in the stack.  :P

So after some looking around, I don't think the eyelid flex "problem" is a problem at all. I think that's just how they are supposed to look. I've seen some models exported with SMD and VTA that do some more complex calculations to have the same eyelid movement, and their eyelids start at a more opened state. However, thats VTA and it's very complicated to do so.

Seems most DMX models I see have the eyelids default to half closed when the eyelid bool is set to 1. Just as long as they work and the multi_CloseLid flex works as it's supposed to, I can live with that.  ;D

Alright, so I'm back with some general observations about DMX flex exports.

Domination rules don't work. But I've identified one of the root causes. In the dominators section of a WW exported DMX file, the word "suppressed" is spelled as "supressed". Changing it to suppressed fixes the problem.

However there is another problem. Dominating flexes only seems to be able to dominate one flex at a time. Add a second flex, and the domination rule breaks and it won't dominate either flex.

I'm also still looking into the eyelid flex problem.

Cool. That works for now.

I've got everything working fine, with the exception of eyelids. Following Pte Jack's guide, I've got eyelids moving with the eyeball flexes fine, but the eyelid flex (in this case called "CloseLid") defaults to half-closed at zero. I don't know if that's a WW issue or not, but I don't think so. Either way, a temporary fix for that is to just include separate eyelid open flexes, open them first so the eyelids will be at a normal state, and then the eyeballs when raised and lowered will still open the eyelids.

I haven't checked back on here in a few days, so no. But wow that was quick work! I'll test it out as soon as I can.

EDIT: It works! Man I've been waiting a long time for a feature like this, I should've requested it sooner! :P You're awesome Shawn.

EDIT 2: Although I seem to be having a small problem. For flex target meshes that already have flex controller info, trying to add the new DMX controller info on top of them doesn't work, and I don't know how to remove the existing info. The "Remove Selected" button seems like it would do that, but it doesn't work.

Quote from: wallworm on December 30, 2016, 04:52:45 PM
How many tris is this model? I would be interested to see if the Blender DMX is significantly smaller than the WW export with similar meshes.

In terms of the stereo function, what exactly do you do to set it the manual way? Seems to me it would just be better to have a checkbox in WW for a morph target that sets it as stereo and then let the exporter just handle it. I've not messed with stereo but if you explain to me exactly how you set it up then I can probably just automate it.

Well the way it works with SMD/VTA is you have to set the flex you want as stereo manually in the qc file. It's similar for DMX, but you have to manually set it in the DMX file itself. Seeing as WW already creates the necessary flexcontrollers in the qc files for you when exporting to VTA, it seems like it would be possible to have said stereo checkbox, and have WW create stereo flexes in the qc by making the necessary adjustments (changing flex to flexpair etc.)

From what I understand in this guide by Pte Jack, for DMX it's as simple as setting the stereo boolean per flex from 0 > 1. However, when applying this to the DMX file outputted by WW, it separates the flex into stereo, but the stereo balance is set in such a way that the right slider controls the entire flex animation, with the left slider not doing anything. I'm not sure if it's possible to implement the stereo balance utility into WW as it is in Blender Source Tools,as shown here:

Quote from: wallworm on December 30, 2016, 04:12:37 PM
There was an update to the DMX exporter several months ago (V.2.835) that should have drastically reduced the file-size. Are you still experiencing such large file sizes?

The size for the head with flexes is 24 mb, but there is still over 1 million lines of code. It would just be drastically too inefficient to try to sort through all of it to enable stereo flexes, when I could just make separate stereo flexes in the morpher modifier. AFAIK, the process is much simpler with Blender's Source Tools, but I can't bring myself to learn Blender's weird interface.

Resetting the XForm solved the problem. Thanks Shawn!

As for the difference between SMD and DMX, DMX allows for wrinklemaps. However, from what I understand, the only way to make stereo flexes with DMX files is to open the DMX file itself and modify the flexes that way (as I've seen on tutorials for doing it Blender), but the DMX files that come out of 3ds Max are something like 6 million lines long.

I've been having this problem for quite a while now, but I've never bothered to post about it till now because I've just been using SMD instead in its place. For the record, I never used to have this problem with DMX, but somewhere along the line it changed.

I'm running the latest version of WW Pro on Max 2014. 2016 produces the same result, so it's not the older version that is the problem.

This is how the model should look (body exported with SMD):



And this is how the body looks when exported with DMX:



As you can see, the rim lighting completely bugs out on DMX, and the phong screws up too.

It gets even worse in SFM when you combine ambient occlusion in the mix: