UMA Wizard

Sadly, I'm not a coder yet. It may eventually come to that, but it will surely be a while before that day.

As for the DNA translators, I'm not sure how that works, but maybe that is like the "constraints" I mentioned ...? When I brought this idea up before on an UMA thread (maybe the pre-release one), I think I was told that messing with the DNA would be a way to do it, but a complicated way.

What I was thinking of, which might be easier, is to let the developer design a figure in UMA using relaxed constraints - so that they could make the head 1/2 the size of the body, if desired - and then save that UMA and use it in the future as a template for future UMAs. So you could make an orc or a cartoon character or whatever, and then specify that any random or user-made changes can only be up to +/- 10% on any adjustment bone using that template. Something like that. So you can have individual variation based on the template, but not wild variation that looks out of type.

Also, I agree that there should be a way of making an UMA be 6' or 2m tall ... even if it means lining them up against a measuring stick on the screen and dragging an UMA height slider till the height matches.

Speaking of constraints ... developers will also need a way of constraining slots and overlays, so that elves, for example, would be restricted to this or that type of pointy ear, while humans would be restricted to non-pointy ear types. In my game I wouldn't have those sorts of constraints, but I'd want a way to lock/unlock features individually, so that players would start with a restricted set of appearance options and gain more as they play.

As for higher detailed UMAs ... I'm glad you agree. At some point there has to be a way of making higher detailed characters, or in two years - what with the usual tech advances - UMA will be obsolete.

As for the expense ... UMA won't be free, as there will be costs in man-hours or cash at the asset store. But developers (and players) will get more flexibility for their money.

 

I'd also like to see a few physical adjustments combined to make a "Slender - Overweight" slider, where you can make the whole body adjustment in just one pass. Also "Slender - Muscular."

There could still be the other sliders for fine tuning, but it would be good to have some sensible rough adjustment type sliders.

Also, while there is a height slider, IIRC, there is no width slider.

 

hi hopeful,

the slender overweight, muscular etc. its what i call presets (i can change the name if this one doesn't make sense). they will be stored in xml files, and you will be able to store your shapes according to these templates. on the shape/preset menu (please refer to the big picture i posted), you will select between one or two presets. so it will work liket this:

1. you go into shapes/presets. select one of the available presets.
2. go into each of the body sections and make manual adjustments.

or

1. go into shapes/prsets. select two of teh available presets.
2. a slider will appear, it will be the amount of blend of each one. you select from this slider.
3. go into each body part and do additional editting

within this shape menu., there is a button called "store as preset". if you do, next time you'll see this as an option. in that way each user will be able to build their own library. regarding the +/-10% i agree, that is what i intend to do with the classes. you save a base class, and then go to the class wizard and specify a range for the randomness in the crowd. so if you want an army of green hulks, you won't get skinny ones, nor caucasian ones. so yes, we are indeed on the same page. cool!

regarding the DNA, it isn't that hard, at all. if i may, you can see the DNA script as a transform from few to many. there are many adjust bones on the UMAs, and they get mapped into a smaller set of "shapes". so for example if you examine HumanFemaleDNAConverterBehaviour.cs, you can see that the parameter "LowerBackBelly" (a bone in the rig, if you open the base uma on a 3d editor), its scaled based on umaDna.belly and umaDna.waist. and that's all. i assume fernando figured out this formula. in that case is just scaling. other bones get rotated or translated. this is the DNA. and it can be greatly improved if looked in more detail. not saying currently is wrong! you can even create your own race. say a human too. add even more bones and give it a higher degree of adjustment. and the only single script you need to do is this DNAConverterBehaviour, and that's it (well you also need to skin the mesh, and rig it, but that is done on the 3d sw). a simple automated process is what i think joen wants to do with his power rigger. it will be a very cool and handy tool!!!

in my case, once i have the wizard stable and i get to focus again on my game, either i make a new base mesh for male and female with a different topology (i like DAZs very much, so something similar (im not copying!)), or maybe keep using the current UMA one, but i will very likely add additional bones and implement more customization features (something closer to makehuman, and even makehuman is missing stuff on certain parts IMHO). then i'll figure out the DNA converter, and use that for my game. i don't want to use the current texturing system, so i will only use UMAs mesh/shape part. for teh textures i already figure out a different workflow, but ill leave that for myself. and ill use the wizard scripts.

in regards to how tall and the measuring stuff, that is also another formula that can be written based on a few parameters from either the umaDna parameters, or maybe with an even far simpler method that is reading the position of the top most vertex of the skinned mesh. that is the nice part of all being standard, you can actually know which particular vertex we are talking about. ill actually add that to the next update.

i agree with you, for me the higher detailed everything is extremely important. but this is more an artist stuff than actually a coding one, and the dev team is basically focused on coding and debugging (the coder guys), and on getting nice shaders working (the artist guys). this IS a must and has higher priority, so the issue is not with the dev team. we need to get more artists to generate high quality content for UMA.

finally regarding the wizard, im currently on the filtering issue. meaning this slot matches with these overlays. the body parts are these and those. etc. but since we will now have quadruped UMAs too (animals, monsters, etc.... cars??), i need to make the wizard general enough, or be prepared to recode what im currently doing, or only supporting humanoids for the time being,... etc.

anyway, i'll continue working.. will keep you guys posted on progress and post pics once progress becomes more "visual"

cheers!

 

fear not, i just figured out how to address the potenciallyInfinite-multiSpecies-type of UMAs filtering issue

 

I hear you. For some time now I've been wondering ... horses, wolves, monsters ... cars? Buildings? Worlds? Can UMA do everything?! lol

Someday I'll have to try to take a peek at the DNA section. It's been about 9 months, I guess, since I last looked at UMA in any depth.

 

lol, well the concept on its self is simple. uma uses skinned meshes. any skin mesh can be a potential uma. what uma adds is that uma uses additional bones in the rig. some bones handle the normal skinned mesh, and the others are used for scaling (relocation uses the first set).

http://en.wikipedia.org/wiki/Morph_target_animation (uma uses skeletal animations, but for adjusting the uma, not only for animations)

add to that a mesh assembler system and layered texture atlas, and have all that done dynamically and that is uma (i know there is far more,.. hope the devs don't get mad at me if i oversynthetized). so any mesh that is a skinned mesh can be made an uma. you set some additional bones, and write a DNA converter if you want stuff to work with "sliders". and voila! new race

 

hi guys, just to let you know im still here. we are implementing some stuff into the slots and overlays to help the editors know what belongs with what, etc. this is being done in the main uma project, and it will be automatically propagated into the new wizard update. it has been a bit slow (i have had extra work this week at my job), but it will change the editing workflow to something im sure you guys will like more - me too, im a uma/umaw user

please allow a bit more time while it gets finished. i assure you guys the wait will be worth it.

cheers!

 

can we sync uma models over network?

 

hi daeuk,

of course you can. im actually preparing an additional tool for that, but that one will come after this long update for the wizard. guys, btw, im working on the new workflow and that added to unexpected work load from my main job has made everything take a bit more, but its on its way.

meanwhile, please note that at least the UMA as such is made from a recipe that is coded as text. and this recipe text is a simple JSON. json is just data stored in a long text string (that you can open in notepad or any text editor). since it is text, you can just send it via whatever method you prefer. you can email it, or you can have scripts to automatically send these recipes. the receiving part will just build a recipe with that text and have the umagenerator draw it (assuming it has the same slots and overlays too). hope it makes sense.

some tools will soon be available to help wtih that.

cheers!

 

hi guys, i just wanted to tell you all that im here and work on the wizard continues. i just have had both issues at my main job and personal matters, sometimes they just come at the same time, and it has taken an absurd amount of time and energy. i expect to be free by wednesday and resume work. i have continue working, but have not been able to do so with the required intensity. apologies.

 

Things have been quiet and slow lately all over UMA. Hopefully before the end of May things will pick up again.

 

hi hopeful and guys,

well, i'm in the middle of those moments in life when problems come all at the same time from several directions... anyway, hopefully will have some free time finally by the end of next week or beginning of june's second. now i'm totally absorbed between family issues, and a mobile project that is giving more trouble than expected.

anyway, i received an email from a user stating there was an issue with the new 4.5 version. i'll try to at least address that during this weekend. the problem is that when i temporarily suspended work on the wizard, i was mid way with a big change and all the scripts are in "low power mode". i also modified them for a change i was going to commit to the master, which i actually finished but didn't commit because i was having issues generating the dll files... so i'm not sure how much time i need to resync the project. i'll try to figure out a solution to address this and the 4.5 problem this weekend, and then when i finally get more time in a week+, i'll carry on with the delayed roadmap.

the main focus is achieving a more stable release, in a more modular form, that will allow you guys to make better use of the tool even if i have to temporarily be far so to speak for some weeks. and of course, all the gui options.

in summary:

1. i'll address any stability issues with 4.5 this weekend. by monday i'll let you know if a new intermediate solution was submitted to the asset store.
2. roadmap will resume once i'm able to solve some issues at main job, they have taken more time than initially expected. i now assume this will happen by june's second week.

cheers,

 

How is the stability coming? I just bought this and tried to follow your youtube tutorial here:

and got a lot of IndexOutOfRangeException: Array index is out of range. when trying to add meshes.

Doesn't seem to be practically useable at this point, but it looks like a great tool, so I'm hopeful. Also please consider using the default font instead of black? Should be the same result on Free but on Pro using the dark skin I can barely read some of the buttons and labels.

 

Is this still actively being developed? I purchased recently, and we are currently trying to figure out what extensions we'll leverage related to UMA. Development seems to have dropped off considerably on both the main UMA track, and this extension. Hope everything is going well Erwin!

For now we are using it in 4.3 for evaluation, but our new projects are largely in 4.5 now. Just FYI, we are also using Powertools for the prefab savedown. I'm really looking forward to changes you mentioned most recently, though that was almost 2 months ago and you'd mentioned having something out that next week (which was like 7 weeks ago). I know how side project timelines can slip under the weight of personal life and primary projects.

-Andrew

 

bummer.

 

Hi

After buying your tool I've run into a few problems. Primarily After clicking off an UMA and then back onto one all of the meshes disappear except for the hair on female characters and legs on the troll character currently making the system unusable. Also the skin colour swash at the beginning doesn't do anything, if I don't select an animator; the system fails on creation, and if I try to add my own meshes I receive xml issues and have to reimport the plugin entirely to allow it to function again. Sometimes Unity will also just freeze out.

What advice do you have?

Thank you

 

Hi, can this be used for UMA2? thanks.

Just saw that the asset has not been updated since Mar 25, 2014. Has it been abandoned?

 

esaavedra76 Is this asset still active and will it work with newer versions of UMA, I believe they are at something like 2.12? The content page on the asset store says this tool can make recipes. Is that recipes from existing UMA wardrobe collections, or is that new recipes. I am looking for a tool that can take existing mechanim prefabs that have attached skinned meshes of the clothes and turn them into an UMA, does this tool do that?