Recruiting for Makehumans Future!

Locked forum where the devs and staff post news about the makehuman software

Re: Recruiting for Makehumans Future!

Postby punkduck » Sat Feb 25, 2023 7:38 pm

tomcat wrote:Legacy addiction very often gets in the way of development. This applies to programs to the fullest extent. In addition to retaining old users, it is good to think about attracting new users. And, will all the old users give up on the new features? And how many MakeHuman users fled to MetaHuman?

I have a MetaHuman implementation in one of the unreal versions at home. Most people use it for unreal engine so far. The characters were professional and huge (memory)... you will not get one mesh, you get more than one (head + body). And then try to understand how the character is formed. I got it barely running on my box and still ask myself why I need a Gigabyte per character on windows. For Linux does not even work correctly on my Box. Metahuman is a character generator for Unreal. They have possibilities in budget and ressources we never will have.

tomcat wrote:
We are not a DAZ clone

Not a clone, but competitors are an important nuance.

No I do mot see them as competitors. We are open source. Either people buy DAZ or they steal characters like with smutba.se and mostly add sexual parts .. :lol: . And when I look at the art in deviantart most of the characters look kinda identical. Cute babyfaced young girls. It is a human-out-of-the-box system like when kids buy a barbie-doll. Ah yes ... and try to make clothes for DAZ :P ... flexibility is not really wanted, they want to sell things, that's what they do for living. Until a long time they even tried to avoid exporting there stuff to Blender.

tomcat wrote:It's actually quite handy when you can customize the panels "for yourself". And "running" through the whole screen to select one garment, then switch tags, select another garment, repeat… — is not very convenient.

In a certain range it will be okay. But I will avoid spending most time in fixing code because a system is over-flexible instead that we program new features. We made this fault already by allowing stupid filenames and native language support in places where it broke e.g. JSON structures

tomcat wrote:Namely, that "not all people do that". Often tags are neglected, so they can't replace categories.

that's the reason why the tags available will populate a database and this database entries can be changed by the user. If the user is too lazy to do that it will not be better, we will not guess if the asset is underwear or not, so best one should do the own classification after download. For most user this would be okay. If someone loads everything available, then it's his own problem.

tomcat wrote:It's not a good term and bad breath. ...


The term is one of 4 different terms used in English and not my invention.

https://en.wikipedia.org/wiki/Diaphragmatic_breathing

We Germans use a term like "belly breathing" which is also more or less wrong.

tomcat wrote:
is usually upperarm + shoulder section.

The only difficulty here is that they have to move relative to each other in an interconnected way. And the bones are controlled separately.

Only difficulty? Well and they do not all work like ball-joints at all. Some of them slide (scapula can move up and down and is able to do protraction, retraction). This bone-chain furthermore does not support all axis for all bones and it is not linear. So doing that with a weighted bone-chain is also not correct.

https://www.youtube.com/watch?v=B5O1h73ltKc

No, honestly, that is far more difficult than just extending the breast for breathing.

tomcat wrote:Why one page? All the settings are already categorized by tabs. All that is needed is a value entry box beside the name.


when custom slider will be part of this, we might get these long lists. Because people add them .... We need to do a smarter method imho
On the other side one normally changes one item per time ... so it could be done different

Ideas we have enough. Just another example: why not copying a head from one character to a kind of standard body for a game? If a game needs characters of same size to make interaction with objects easier the same body size and arm/leg-length would be needed. But the head should look different ...

But the main question is ... will we have enough programmers to do it? At least this brainstorming showed me that I am not that wrong with my ideas.
I even struggle with the basics ... not with the details atm. But we will do a plan how we can go on.
User avatar
punkduck
 
Posts: 1216
Joined: Mon Oct 17, 2016 7:24 pm
Location: Nuremberg, Germany

Re: Recruiting for Makehumans Future!

Postby Elvaerwyn » Sun Feb 26, 2023 11:35 am

jorgeo wrote:I'm experienced in Python programming and SQL databases. MakeHuman has been good to me, so I'd like to give back as I can


Jorgeo I sent you a pm, please respond to me when you get time, as we have a lot to ask and communicate! Ty so much for your interest and involvement in the project!

Elv
She who knows in repose with prose...
User avatar
Elvaerwyn
 
Posts: 373
Joined: Tue Aug 22, 2017 11:28 pm
Location: Canada

Re: Recruiting for Makehumans Future!

Postby Elvaerwyn » Sun Feb 26, 2023 11:37 am

guenrij wrote:hi @Elvaerwyn I am Guenrij Silva and I want to collaborate with the Make Human Project as a 3d artist > https://guenrij.wixsite.com/3dart https://www.instagram.com/silbender/ Cheers!


Greetings Guenrij!

I will pm you shortly with some questions and comments! Ty so much for your time and genuine interest in this project! We love artists and value their time and focus!

Elv
She who knows in repose with prose...
User avatar
Elvaerwyn
 
Posts: 373
Joined: Tue Aug 22, 2017 11:28 pm
Location: Canada

Re: Recruiting for Makehumans Future!

Postby Elvaerwyn » Sun Feb 26, 2023 11:42 am

Greetings

Salty, Tom and Ricardo- Ty so much for your time and offers for testing etc going forward! It is wonderful to see so much genuine interest in the rejuvenation of makehuman! I look forward to seeing how much we can do and where this will go in the months of future of the project.

We value all of you and in fact all of the makehuman community as this is what keeps our software alive and well!

As you are needed we will ask of your time and involvement, however always feel free to work on any projects of your own and present them and in fact any ideas you may have also as these are helping to form an outline for our development going forward!

Very happy with all of you, ty again!

Elv :D
She who knows in repose with prose...
User avatar
Elvaerwyn
 
Posts: 373
Joined: Tue Aug 22, 2017 11:28 pm
Location: Canada

Re: Recruiting for Makehumans Future!

Postby tomcat » Sun Feb 26, 2023 3:03 pm

A few sentences and reasoning.

First, I will unequivocally agree with
punkduck wrote:and mostly add sexual parts ..

Sexual content is, indeed, woefully lacking! I fully agree that it desperately needs to be expanded.

"Before we argue, let's talk about concepts." (c) The ancient Greeks.

No I do mot see them as competitors.

What are "competitors"? When we open DeviantArt and ArtStation, we see a lot of work with human characters. How many of them are made in MakeHuman, and how many in DAZ? Although, MakeHuman is positioned for artists. Although DAZ is paid, and with the creation of clothes problems, but the vast majority are using it. Why?

Among many other things and that's why too:
If someone loads everything available, then it's his own problem.

When a user downloads and gets one of these:

Image

then it's not exactly comilfo.

Here, I corrected it, but why does everyone need to do that?

Image

I propose that somehow more responsible place content. Especially since it is already enough and can choose — some kind of filter on the quality would be good to enter.

I know a few games that used MH for character creation. One studio used the body, but made their own head — they had to do the aging of the character.

The term is one of 4 different terms used in English and not my invention.

I didn't say that you invented the term. It's just that when a term is used by an expert, he understands what is behind it. And a person who is far from medicine can interpret it directly, which is wrong. That's too far to the side, though. What I mean is that it's possible, to begin with, to do simple normal "chest breathing".

Only difficulty?

What exactly is the problem? Which bone is responsible for the scapula? How can you make a figure like that?

Image

In my opinion this video shows the mechanics better.

At the very beginning, you can see that first the arm rises by a small degree, and then, along with it, the clavicle begins to rise. This can be programmed by a script in the simulator based on the game engine. And how to do it when all the bones are controlled individually… only by sculpting, morphing… probably. And apply manually.

I want to draw attention to an unoccupied niche — biophysical modeling of the human body. Same as Z-Anatomy, only in dynamics. Take a skeleton from there as a basis. Flesh is modeled — muscles and fat, and covered with skin. Thick/thin/muscled would depend on the amount of soft tissue involved. But that's a pretty radical suggestion. But then you can solve a lot of problems, for example with the same shoulder girdle.

Perhaps this could be considered for the long term.

For my part, I can place a recruiting ad on the Godot forum. Since MH can be used to create characters in games, it would be quite appropriate. And there are enough programmers out there.

Wouldn't it be easier to take a game engine as a basis and make a new MH on it? This doesn't mean that models can only be used in this engine — Godot allows exporting to glTF format — in fact a new standard for 3D.

They have possibilities in budget and ressources we never will have.

Never Say Never 8-)
Foreigners' reactions to Russian "Bird's Milk" candies
— Are your birds being milked?
— In Russia everyone is milked. Here even the zucchini is used to make caviar.
User avatar
tomcat
 
Posts: 336
Joined: Sun Sep 27, 2015 7:53 pm
Location: Moscow (Orcish Stan), The Aggressive Evil Empire

Re: Recruiting for Makehumans Future!

Postby punkduck » Sun Feb 26, 2023 6:43 pm

tomcat, I know there are concepts with volumes modeled on bones to simulate a few parts. If you have the knowledge, feel free to make a try. For me most of it will not fit to a system which cannot easily be rewritten, because we do not have a real 3d system, we have planes which are moved by bones. That is more or less 2d on curves.

What you need is a finite element system with a complete different approach. For me that's another software. I cannot write such a solution with my knowledge, maybe you can.

Moderation of assets is not yet wanted. A quality gate might help. With the danger that then nobody delivers assets at all. People over here do not like dictators :?

When you show me pictures with old ugly characters in deviantart which are made with DAZ then I might agree with you. Not zombies, real characters. When I open my page I get a bunch of nude young girls as "see also" ... honestly not all people in deviantart are real artists in my point of view. I don't even consider myself as one, especially when the only effect was to do a "cute" more or less nude female character. When someone is able to unpack a box with a barbie inside he is far away from being an artist. Of course also real artists (in my eyes) use DAZ and it is also the base for some games. Part of DAZ concept IS to sell attractive characters. Same goes for iClone. Metahuman is still for free, they indeed have more realistic "normal" characters there.

gltf might work in the end. The ASCII solution would be easier, but the binary one is a problem, python is not c++ or c ... it can write binary code but it is really no fun. You know from a PM long ago I already thought about that.

Sexuality. Well if you really need a functional MakeHuman character with working private parts: I have that at home. Male and female. Remodeled private parts of course, a few shape keys for her and a few additional bones + shapekeys for him. And it works in Unreal including sound and it looks very natural using blendspaces with random values. The whole kamasutra btw. Unfortunately I will not give that to the audience. You can now believe it, or not. It's up to your fantasy. :shock: :lol:

I ask the other way round: we need people for documentation also. Would you do it? Or as an asset moderator? Maybe we convince the people ... for me that would not be a problem.

greetings pornd.. eh punkduck :mrgreen:
User avatar
punkduck
 
Posts: 1216
Joined: Mon Oct 17, 2016 7:24 pm
Location: Nuremberg, Germany

Re: Recruiting for Makehumans Future!

Postby davewoodrum » Sun Feb 26, 2023 7:53 pm

I have some interest in doing 3d aspects, accessories, and monstrous parts design (my programming abilities are limited to variants of BASIC and mostly text manipulation). I'm worried though that I may or may not be too useful, based purely on how I design, what I design, etc... hence why I'm mulling it over.
I come from a tabletop game design background (I run Fishwife Games and have done work in the past for Monkey House Games, Spectrum Games, Fantasy Games Unlimited, Smif Ink, Dark Quest, Dark Furies, etc.). My 3d modeling has been mostly for table top game design interests in the element of 3d printing. As such, I've not had to worry about colorizing/skin overlays in the past so that will be a wee bit of a learning curve for me. Another learning curve, if needed, will be rigging models/accessory parts. I did some contract work for Desktophero 2.0 (now Titancraft) in the past designing weapons, accessories, and robot heads, and most of these items did not require rigging. I've been meaning to learn rigging models though as it would quite frankly help me mass produce more printable mini designs in different poses, etc. So this might be the push I need.

I really don't know how useful I'll be, but willing to give it a shot. (I've also arranged electronic music from both royalty free/commercial ready and self made samples... don't know if there's any need there but I can help if demo video background sounds are needed)

Lastly, beyond the weapons design and so forth, my stuff is... well, kinda weird. Not weird in a controversial or risque way, more of weird in a creepy critter 60s sci-fi gumball machine monster sort of way. A sample of some stuff I've made for printing in the past (note this is set up a bit different (such as fill in areas between fingers, etc... for printing purposes): Image.

\
User avatar
davewoodrum
 
Posts: 2
Joined: Sun Feb 26, 2023 7:33 pm

Re: Recruiting for Makehumans Future!

Postby davewoodrum » Mon Feb 27, 2023 2:41 am

Here's a shot of some of the weapons I designed for DesktopHero 2.0. Again, no colored "skins" over any of this as was not needed for a 3d printable perspective. Something for me to learn. These were made a couple of years or so ago. I've got about 4 or 5 years of 3d design under my belt now, so these represent my "midway" point work. I was more mechanical early on (a whole lot of Tinkercad work and little else)... I drift more into organics now (using Tinkercad at times for the base model shape and then modifying in Meshmixer and Blender... also use Sculptris quite a bit for heads and other body forms).

Image
User avatar
davewoodrum
 
Posts: 2
Joined: Sun Feb 26, 2023 7:33 pm

Re: Recruiting for Makehumans Future!

Postby tomcat » Mon Feb 27, 2023 7:41 pm

punkduck wrote:because we do not have a real 3d system, we have planes which are moved by bones.

Well, it turns out that it is possible to explain in a clear and understandable way… if you want to… At least now I know what to expect. Actually, it's very useful information.

I ask the other way round: we need people for documentation also. Would you do it? Or as an asset moderator?

That's an extremely strange question. You know my situation — in fact, we are expecting a civil war any day now. ImageImage War always comes back to the one who started it. So at this moment, making promises and taking responsibility… is, to put it mildly, too reckless.

And I'm not going to promise anything, and I can only take on non-critical tasks.

I have my own project, which is quite closely related to MH. So I am quite interested in its development. I see the first step as creating a character generator. On a game engine. And it's not … Unreal.

For the knees I tried a double joint .. so one small bone where the kneecap is.

As far as I understand it, it comes from the fact that the bones of the real skeleton have a thickness that is not accounted for in the game rig.

When you show me pictures with old ugly characters in deviantart which are made with DAZ then I might agree with you.

Easy. Type in "old man DAZ" into a search. Voila:
https://www.deviantart.com/saidge42/art ... -711824277
https://www.deviantart.com/rae134/art/S ... -837088015
And "ugliness" is a very subjective concept.

Sexuality.

Sexuality (in games) is not my primary concern. I'll probably do adult DLC eventually, but it definitely won't be a priority. But requests to expand such features are coming in regularly.

And it works in Unreal including sound and it looks very natural using blendspaces with random values. The whole kamasutra btw.

Kamasutra in Unreal doesn't interest me at all. I use a different engine. Maybe I'll make an "expanded and extended" edition of Kamasutra on it, but that won't be soon. Image

Unfortunately I will not give that to the audience.

It is in the highest degree selfish! Image

I know there are concepts with volumes modeled on bones to simulate a few parts. If you have the knowledge, feel free to make a try.

Physics and volume — maybe I'll try those… but also in the very distant future… If I survive the troubled times and live to see calmer times. When a more 3D-adapted version of the engine is released. Also… you're not going to rewrite the program by yourself, but invite helpers? So am I — if I can find interested programmers, we'll do it. And if not, then no.

People over here do not like dictators

…but recent practice shows that most people adore dictators… unfortunately. If a dictator is not adored enough, then he is not a dictator and a tyrant enough. Image

But the main question is ... will we have enough programmers to do it?

So, should I ask the programmers in the engine forum? Image
Last edited by tomcat on Tue Feb 28, 2023 12:38 am, edited 1 time in total.
Foreigners' reactions to Russian "Bird's Milk" candies
— Are your birds being milked?
— In Russia everyone is milked. Here even the zucchini is used to make caviar.
User avatar
tomcat
 
Posts: 336
Joined: Sun Sep 27, 2015 7:53 pm
Location: Moscow (Orcish Stan), The Aggressive Evil Empire

Re: Recruiting for Makehumans Future!

Postby spamrakuen » Mon Feb 27, 2023 10:45 pm

First of all, sorry for my poor english.

My three cents;

1 - Focus on MFBP. Or, in other words, convert MH in a Blender addon.
-- Why?
--- Many people knows how to use it, and love it. This should help growing MH community (so, more assets for MH).
--- Wasting time trying to give MH features already available in Blender is... well... a waste of time (IMHO).
2 - Many assets are, frankly, low quality. This makes MH look like software that is not good for creating quality content.
-- On sites like blendswap you have "staff picks" that help show the potential quality available.
3 - Anime Style base mesh would be nice.

As an user of MH I've found several issues trying to use it for my Unity game projects. To workaround those problems I've:
- Made my own addon for Make Human (for exporting multiple .mhx2 files).
- Made my own addons for Blender for importing those .mhx2 files, create blend shapes, load custom weight data from json files, adjust rig, and so.
- Made my own addon for saving/storing clothes in my own format (which preserves stuff like face normals, material ids, and so)
- Made my own clothes because the ones that come with makehuman get too deformed in my anime style body. Also, finding goog assets (they exists!) is a pain (lot of time wasted finding some good clothes for a character).
- Made my own poses because the ones that come with makehuman are impossible to use on my rigged blender mesh.

Some of my issues may be already been "fixed" / workarounded with MFBP.

So, I believe that many potential users just ignore this software, not because is aging, just because its hard to use. I'we wasted a lot of time creating my own workflow mh -> my games, and I'm quite sure many people just find MH too hard to be used for their pourposes.

Also, I've found quite complex trying to contribute MH community with my own assets. I've shared a few assets with MH community (I'm a coder, not an artist, do not ask me for high quiality assets), but it is also a pain in the ass.

A newbie may go here;

http://www.makehumancommunity.org/wiki/ ... :ClothesV2

A quite complex workflow of another tool (makeclothes). This discourages many users.

Also, some workflows are missing, like one for creating a pose or an expression.

I'm a coder, I can create my own tools and dig into undocumented code or unkown file formats, but if you want good artits contributing with nice assets, make their life as easy as possible.

I would like to help, but I've two little children and almost zero free time.
spamrakuen
 
Posts: 18
Joined: Sat Feb 13, 2021 11:25 pm

PreviousNext

Return to News from the crew

Who is online

Users browsing this forum: No registered users and 1 guest