Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[YAHG]

therahedwig , thanks for the colors palette, probably will use some of its colors.

Here is draft of one of the possible styles for the visualizer, little simplified and edgy.

Spoiler (click to show/hide)

I think he looks badass/adorable he is like a cute lil midget biker <3. The nose is throwing me off being such a different color, but its important to remember that I never know what I am talking about and to discount any opinions I have accordingly .

[CharonM72]

What a fantastic idea this is! This would really help the player emotionally connect with their dwarves. I'll be keeping an eye on this too.

[Dirst]

I really like this idea, though to start I would stick with the passport photo to decrease the number of traits to you need to track.  Once you can generate a picture, don't worry some other 3rd-party utilities will find a way of inserting it into the interface.

You might want to look at Stonesense's large sprites.  The program already knows what Dwarves are wearing, how their hair is styled, and pulls skin/hair colors from somewhere and applies them to the sprites.  You'd probably want to pick up iris color as well.  The other modifiers (close-set eyes, lip thickness, etc.) are numeric scores.  These scores can do at least a little of the work in positioning/stretching bodypart images, though I wouldn't recommend relying on it to produce great results across the entire range.  The troublesome one would be scars, since I don't know of any existing tool that uses scars to provide an example.

One idea to keep in mind is being able to down-convert your images into Stonesense 128x128 sprites.  This becomes more relevant when you ignore my advice above and go for rendering the whole body.

[therahedwig]

One idea to keep in mind is being able to down-convert your images into Stonesense 128x128 sprites.  This becomes more relevant when you ignore my advice above and go for rendering the whole body.

Stonsense can't actually target a specific dwarf with a graphic. It rather reads the traits of the dwarf and assembles a sprite from that. It doesn't really use facial features either.

That said, I like the look of the sample-dwarf.

[AndrewChewie]

Thanks for the advice and feedback, appreciate it
I think portraits will be generated with a little correction to the base ingame descriptions to look good.

Here is some test haircuts for the dwarfs:
   

   

UPD added different hair color

[Lysabild]

Personally I liked the first sketch more, but either way this is an amazing project and I wish you the best of luck and motivation for seeing it through to the end!

[Megaman_zx]

Awesome!!! i suggested someone do this once, but have neither the time nor skill, amazing work, and best of luck!!!!!

btw, do you think you will do this for the other vanilla races? humans, elves, goblins... kobolds? 

Definitely gonna follow your development regardless

[King Mir]

This is cool, and I look forward to being able to use it.

[Meph]

Looks really good.

The units in Stonesense can already show the beard and hair styles and color, so best to ask Japa if he knows how you can extract that data. Since the code already exists in some form I hope its not too much work to convert it to your project.

[AndrewChewie]

Thanks for support =)
Will be updating this thread asap.

Personally I liked the first sketch more, but either way this is an amazing project and I wish you the best of luck and motivation for seeing it through to the end!

You mean first lady dwarf sketch that is more realistic?
It'll be hard to produce all face parts in this detail.

Awesome!!! i suggested someone do this once, but have neither the time nor skill, amazing work, and best of luck!!!!!

btw, do you think you will do this for the other vanilla races? humans, elves, goblins... kobolds? 

Definitely gonna follow your development regardless

Thanks, if everithing will go well then it's not a problem to produce art for other races as well

[Petr Ga]

Great work!

I tried to do this some years ago via python and DFHack and was partially successfull, but had lack of attributes via DFHack API and art skills.

So, do you know how to program it? Becaused right now it looks more like tries of art style (which looks GREAT)

[AndrewChewie]

Great work!

I tried to do this some years ago via python and DFHack and was partially successfull, but had lack of attributes via DFHack API and art skills.

So, do you know how to program it? Becaused right now it looks more like tries of art style (which looks GREAT)

Yes, currently I'm researching art style for visualizer and then I'll move on to the coding or there will be someone who can take this part of work.
I can make simple portrait generator in Unity3d, but I lack skills to make it fully functional utility.

[Dirst]

One idea to keep in mind is being able to down-convert your images into Stonesense 128x128 sprites.  This becomes more relevant when you ignore my advice above and go for rendering the whole body.

Stonsense can't actually target a specific dwarf with a graphic. It rather reads the traits of the dwarf and assembles a sprite from that. It doesn't really use facial features either.

That said, I like the look of the sample-dwarf.

The current version can't, but Stonesense obviously knows which dwarf is which.  If Japa is aggreable, it might be possible to have Chewie's tool generate "naked" sprites for each dwarf then let Stonesense do it's magic with the clothes and equipment.  The hand-off needs to fail gracefully so that creatures that Chewie can't render still get default sprites.

Again, this is all way down the line.  For now just a passport photo would be immensely helpful with immersion.

[bicueites]

Do you do 3D art?
I've already implemented the dwarf fortress data extract (lua script) and the python-ogre build scene, unfortunately my models are bland and dull.

see
http://www.bay12forums.com/smf/index.php?topic=138229.0

[AndrewChewie]

Do you do 3D art?
I've already implemented the dwarf fortress data extract (lua script) and the python-ogre build scene, unfortunately my models are bland and dull.

see
http://www.bay12forums.com/smf/index.php?topic=138229.0

Yes I have skills in producing 3d content as well, I've posted my test dwarf model sowhere here a long time ago.
here is some:
http://imgur.com/a/s0o75

Why you desided to use Ogre?