UMA - Unity Multipurpose Avatar on the Asset Store!

Is it much work to update from the old to the new uma (dcs)? What are the steps?

 

Link on product page Visit Publisher's Website shows to nowhere... Demo scenes are not working... Is UMA alive?

 

The updated UMA is backward compatible, so to use the new one, you only have to import the new package. Might be best to delete the old one first.

To use DCS will take some more work, as you will have to create recipes for your wardrobe pieces. It's not too hard (right click, create recipe, add slots, add overlays), but if you have a ton of them, it can get a bit tedious. We've tried to streamline the process as best we can (adding drag and drop, etc). But it's well worth it.

 

UMA is alive. The wiki that was hosted by Unity died, and we haven't had a lot of luck getting back up and having it remain up. So we moved it here: http://umawiki.secretanorak.com/

We're working on an update (lot of things broke in Unity 5.5), and it should be available soon. There's a lot of new features and content coming in the update (I highlighted most of them a few messages ago). But you can get the WIP here:

https://github.com/huika/UMA/tree/unity53

Here's the message with the updated content:

https://forum.unity3d.com/threads/u...n-the-asset-store.219175/page-90#post-2890255

 

We are very much alive, new update contains loads of new exciting features that will be well worth the wait : )

 

Thumbs up for the backward compatibility. We've pretty much equipment pieces, so I probably will test the new workflow before switching. Is it possible to use the new asset bundle feature when not using DCS?

 

Really can not wait any longer anymore ^^
Dont like MCS or DAZ, and MakeHuman is not ready to use with Unity.

I hope this will be a full release, not "early access" "beta" or not complete release.

 

Should be - you'll need to switch to using the "dynamic" version of the library components, and use the loader (there's a prefab for you all set up). It can load from resources and asset bundles in addition to out of the library (race library, slot library, overlay library). Then you'll have to manage the asset bundles yourself (where is it, is it downloaded yet, etc).

 

Awesome, thanks for the quick replies Jaimi, can't wait trying out the new features.

 

We're working hard on it, you can check out the current state on git here if you like:

https://github.com/huika/UMA/tree/unity53

Here's a quick screenshot showing the new overlays, meshes, shaders, etc. Yes, it's not perfect, but I believe that it looks much nicer than before.

Note that this is just the first sample scene that includes randomly generated characters. It's not a scene with commercial quality outfits, etc.

 

Looks definitely better than the current version, nice. But what I personally don't like is that everything looks plasticky, no skin shader and no cloth shader applied. I guess you guys are aware of that and fix that, I just wanted to tell it anyway ^^

Edit: "first sample scene that includes randomly generated characters" - I guess thats why this looks like that? But wouldn't it make sense giving the characters, even if they are randomly generated, a skin shader and the cloths a "cloth shader"? Well those swimming trunks can look more plasticky so its realistic in that case haha.

 

Yeah, the level of shine on the skin is defined in the color set attached to the randomizer, I just gave everyone a bit of metallic shine. It works better on some color combinations than others. I'll be playing around with the LUX skin shader as well, but using that precludes using overlays on the skin (like the swim trunks), as then it would have to share the shader.

 

You can use your own shaders, of course.

But maybe there should be a demo that uses multiple shaders (like skin and hair from the Blacksmith demo) to show how it can be done.

 

The current one is actually using multiple shaders through three different materials - the hair is using the LUX hair shader, the skin is using the standard shader (opaque), and the eyelashes are using the Standard shader (fade). I do have the LUX skin shader in the package, and I will likely add a scene using it specifically, but it's more work, as no overlays can be used on the body, so quite a bit more setup.

 

I see. Yeah, under the old diffuse shader, there was really no problem applying clothes to the skin as overlays. But now, if we're using a skin shader, probably only things like tattoos, body hair, and freckles could be added.

 

I downloaded the latest version to check out the DCS, it seems like there is a shader missing for the New Hair materials, located here: UMA\Content\UMA\New Body\Hair\HairMat Std and Aniso. Any idea where I can get those shaders?

 

The shader should be in the Lux 2.01 Personal folder at the top level - there was a few minutes last night where I had goofed up the shaders, and then checked in a fix - maybe you got it in between?

 

Ah, oops. I didn't copy that folder over; I thought it was an optional folder for people already using Lux. It's working now! Thanks!

 

In the release version, I'll have that in the shaders folder so it won't be confusing.

 

I ve upgraded from Motion builder 2015 to 2017 and Uma rig is displayed there as spheres and I didnt find way to display it properly other than recreating the whole hierarchy, it doesnt have all options like standard skeleton. There was change between FBX 2015 and 2017, so maybe this is the reason.

 

Hi guys,

Just wondering if there is anything built into UMA for verifying avatar recipes (for server-authoritative architecture, so I can immediately check to see if a client has submitted an invalid recipe and then drop their request if they have, instead of having to actually process a load call and wait to see if it fails, maybe catch exceptions for it or whatever). I haven't started looking around in the API yet so apologies if there's something really obvious for this.

 

Running into a slight issue with the newest build of UMA from Github. Some of the filenames, such as

\UMA-unity53\UMAProject\Assets\UMA\Content\UMA\PBR\Slots\Female\FemaleSwappedHead\UMA_Human_Female_Head_ElvenEars\UMA_Human_Female_Head_ElvenEars_Slot.asset.meta

are getting too long to unpack the project in anything but a fairly short directory name. Is there any way these could be trimmed up a bit?

 

I'll see what I can do. I'm rearranging the folder structure a bit.

 

Not really. You could unpack it, that would give you a minimal check, though it wouldn't validate slot names, overlays, etc.

 

Thanks Jaimi,

That sounds like a great starting point actually, I'm basically just wanting to double check that they haven't sent totally invalid data, since I don't want to have serversided character creation so all I trust clients with is their avatar recipe. So I'll simply unpack it and then check it against the slot and overlay libraries and maybe make sure that all the scales are within the correct ranges.

 

I'm interested in UMA2 but I have one crucial question about it.

I have no interest in breaking up my meshes into chunks, which would be quite a lot of them.
Is there some sort of support for morphing a base mesh instead of having chunks?

UCC (Universal Character Customization) did a phenomenal job on character customization in my opinion.
Unfortunately, just as I wanted to buy it, the developer officially canned it.
Would there be a level of customization similar to UCC?

Also, from my perspective as an artist, I see a lot of potential in literally re-designing or re-thinking the way UMA2 does things... it just seems to be outdated and using old technologies instead of embracing new ones or re-thinking how to allow for more diverse modification.

This asset could be so much more and is a reason I haven't used it.

Humans aren't the only living thing in existence and I haven't seen support for anything other than humans and allowing different methods of customization.

 

So, whats those old and new technologies you are talking about?

 

UMA is just a framework. It's value is to deform meshes and combine their meshes and textures on the fly.
The humans with it are samples of what you can do, you can create your own meshes to use it, and you can split them up in chunks, or not, all up to you. You could just as easily use it for horses or dogs, for example.

UMA uses bones instead of blend-shapes. There are advantages and disadvantages to this. I wouldn't consider either method outdated or old. There are all sorts of problems trying to make a solution with envelopes (etc) and apply those to random clothing. UMA's model of requiring the clothes to be rigged bypasses all of that, and the chunked humans allow you to hide portions of the model so there is no overdraw or z-fighting. You can always add more bones for customization purposes (for example, you could add a pony tail that had it's own bones, and they would merge in).

Anything you think would make it better, we're more than happy to hear about of course.

 

An install and setup guide would be very helpful. Especially with the one from github. I'm getting multiple fundamental issues and errors dealing meshrender, lighting, and slotnames. And this is before I add on the DK UMA.

Add to the confusion is not knowing if I am missing some critical setup steps with a new project.

 

Should be to just copy the UMA and StandardAssets folders into your Assets folder, load up an example scene and hit play.
What version are you using, git or asset store?

 

Installed latest Unity 5.5.0f3 and Downloaded Unity53 branch from github. Both just with it and over UMA 5.0.5.

I just now pasted just the assets you mentioned into the assets folder on a brand new project. And now it works... I think I was copying over too much with the Project and User settings - and maybe the Linux stuff.

Still some yellow issues, but should be fine there. Now for the DK.

Thanks!

 

Yeah, you should only copy the things inside the assets folder : )
Glad you git it to work.

 

I love playing with the customizer. Will's skins, faces, hair, and clothing make it fun.

 

FWIW - I'm not sure what DK all does, but there is a character wardrobe system now built in to UMA - you can see it in some of the sample scenes - such as "1-DynamicCharacterSystem - Simple Setup"

 

I am going to Twitch me using the built in character wardrobe system to customize UMA's. I will post the video when done, if I succeed. First time using Twitch and it could be a disaster. lol

 

That's *right now* for you thread watchers:

https://www.twitch.tv/watregames

Edit: It's over with. It was great!

 

Yikes! Didn't know you posted it here. Thanks, Jaimi. I am glad you all were there. I was so nervous. So far, the constructive criticism is that I need to calm down. LOL

 

It was great, looking forward to next time . )

 

Hi Everyone,

It's been a while since I made any tutorials, but I've got plans to make a new series for the upcoming UMA 2.1. The next update will make UMA more accessible than ever and open up a whole new realm of possibilities so I thought I'd share what I've spent the last couple of days making.

If anyone has any Crazy Ideas for me to try out, please let me know (UMA buildings are in the pipeline)

 

As much as I appreciate the effort you put into your videos, not all of us have multi-monitor alienware systems that would let us have a video full screen (big enough to see what the heck you're doing) on one monitor while Unity (resource hog that it is) is on another monitor to try to follow along and learn from them. Additionally, with constant updates to UMA, they become outdated quickly. I'd really rather there was written documentation with cropped screenshots for reference and step-by-step walk-throughs of how to use the product and how to accomplish common tasks. They're a lot easier to maintain, and they're searchable, where the videos aren't.

Try to remember that we didn't build this, we don't have the first idea how it works, we can't help document something we don't know about. Assume we're all coming in cold and blind. It's a complex system...it's not a trivial thing one just gets in 5 minutes.

Thanks again for all you've done in support of UMA!

 

There are also a lot of ppl that prefer video tutorials instead of written text, hard to cater to all needs.
Writing documentation is also a VERY tedious and boring process compared to doing a video.
I would love to have both, but since UMA is a spare time project for everyone involved, i doubt that will ever happen.

 

(I still think there's so much new under the hood that it should be called UMA 3.0.)

 

Well, looks like everything might not make it into the next release, like cloth.

 

Then cloth can be 3.1.

 

: D

 

Please make a more in depth video on how to make new races and add the DNA components. I think this would be invaluable.

 

Yeah, no one gets paid to do this so while I also prefer written docs, I see the benefits of videos as well. They are certainly easier and take much less time to make.

 

I understand how this works, and I just decided to use UMA myself.

My game needs an entirely different rig setup for furry creatures, does UMA require a particular bone standard?
For bones however, I want to do lip-sync for speech and I am not sure how UMA works around that.
I don't mind the character customization using things like muzzle size, lip size, eye types, etc. but how can you do that if UMA requires bones to deform a model?

Are there any limitations to mesh deformation so the player doesn't explode a model or make them too big?

What about dynamic bones or soft bodies?
Is it possible to allow for particular areas of the model on the vert level only have soft body features?
The models are generated at run-time, so how would you be able to control which parts of the model have soft bodies?

About documentation.
For one, people don't read anymore and creating a documentation is a waste of time.
Create tutorials on humanoids, animals, monsters, anthropomorphic and maybe even insect creatures.
Not every game out there is going to use humans and the fact that it's almost impossible to find good information cripples more people using UMA to begin with.

I doubt the developers will even bother with what I am saying, but if they do... I will be amazed.
Look at popular systems like Adventure Creator, Final IK, Invector, etc. they all have easy to understand terminology and developers have a great time understanding them and using it.

UMA seems to try to be its own 'thing' but fails miserably at doing so.

For one, it doesn't appear the developers are concerned with newer technology and advances with Unity like some very popular assets are.
On GitHub, it appears the project is basically dead and commits are non-existent.
If you take a look around the popular assets, they update to allow for better improvements and features Unity has.

Most assets either integrate with other popular assets to support what worked or works.
For example, Adventure Creator and PlayMaker support so many assets it's hard to find ones that don't support them... very extensible and approachable and a great design.

What about optional procedural vert deformation like how UCC does it?
People loved it but few seem to want to do it.
Scaling bones seems sort of limited in my opinion, especially if you want to have a specific change... how would you do that?

Slots, Overlays, etc.?
UMA needs a renaming scheme that actually makes sense and this doesn't.
I was actually tempted to rename things on my own project because I thought it was so dull and lame, no offense.
People are also saying this, but I am told you guys are interested to know what I say yet I see nothing about that.

Why not go on Patreon and ask for support if you can't be bothered to sell it on the Asset Store?
You obviously have a fan-base and people are asking for more... so it's a perfect opportunity.

Honestly, like DAZ of Morph3D did, I think this project needs a major over-haul and re-consideration.
Most people downloading UMA on the Asset Store complain that it doesn't reflect any changes and almost confirms UMA is dead.

What would be done about all of this?
Unity is progressing and UMA appears to be in its own corner gathering dust.

 

Creating documentation is most certainly NOT a waste of time.

The UMA GitHub is very active; you need to look at the Unity53 branch where all of the current development is taking place. It's not complete, so it hasn't replaced the official release yet.

Comparing UMA to the "popular systems"that you mentioned misses one critically important factor -- those are paid assets, and UMA is not. That makes a tremendous difference in how much time the developer(s) can devote to it.

I'd be amazed if the developers bother with what you say, too. Might have something to do with your attitude.

 

@Janet-Merai
Just saying that we use "outdated tech" is a waste of time if anything, you need to be more specific than that.
UMA works just fine with lip sync and you can use UMA to modify anything you can imagine, did you not see the videos just posted a few posts up with a UMA spaceship and a UMA horse?