Dragnix.Net

It has come to my attention that no one has ever tried to make scales that behaves like "real" scales on a CG model. All the models of dragons I have seen that have scales is either static (can't move, like a statue) or uses a cheap texture based approach (surface with scales "painted" on). The result is no way near as stunning as "real" scales would look like. But then again, few has the attention to details that I have.

Making scales stick to the surface (skin) of a model is difficult. But this is the simple part, making them behave like scale is where you hit a brick wall. Some would claim it is stupidity to even try. I tried.


2 years later


A working rig.

With 22 scales.


The scales closely follows the surface when it flexes.


Even at extreme poses the scales stays neatly on track



This is by no means a perfect rig. The scales are anchored to surface and can only rotate in the axis parallel to the root side of the scale. This means they tend to tilt up when twisted in certain directions.


There is still a long way until any usable animations can be made with this rig:
http://www.youtube.com/watch?v=guqlPff7fks


Edit: Woops, wrong forum, was supposed to post this in the artworks section.

That's very good!

I do hope you intend to someday complete all the scales, and if so, I do hope you'll show us the finished product. But yes, a CGI dragon with movable scales is a grand idea, and does look very good, and true to the idea of what one expects when they think of how dragons move and act.

@Sholl: The problem with completing the rest is that there is no body, only a head. Might change in the future.

Current scale plan:

From what I understand, the hard, armored scales needn't cover every last part of the dragon's body. Smaller and more specific body parts, like the fingers and the toes and the like can be uncovered, I think.

The model of the head, with the current scale plan, looks very good to me.

Covering every part of the dragon with scales is impossible. Just try putting something stiff between your elbows. The arms, body, neck, etc... will not be completely covered. Most of the head will, except for the areas underneath. Which is still unclear.



The top side is also a bit unclear. There is a dominant line on the center (yellow in the image) with slightly bigger scales. But i can't decide whether keep it longer back. The area between the horns is not done yet.

I just came to the conclusion that creating the about 110 scales is going to take a very, very long time. Not to mention the scripting required to mirror the scale rig.

I just came to the conclusion that creating the about 110 scales is going to take a very, very long time. Not to mention the scripting required to mirror the scale rig.

Oh really?

What program are you using for this?
It's very good, can't wait to see your progress...

7 down 93 to go

Some 300 lines of python scripting and the creation and duplication of a 4 part scale is getting convenient. And the incremental version number on the save file (I create a new file for almost each save) just reached v1.00!

@Marik: Blender 2.49

Yeah, I thought it might be, it looked familiar.

Looking good so far...

About 30% done

The rest doesn't move in test animations and is thus postponed.

@Marik: Was it the pink selection color, or the 3D cursor that gave it away?

Vamp, could you please move this thread to the Artworks forum, This feels a lot more like a "I can make a dragon" thread than a discussion.

Rig test, need i say more?

Sivert3 wrote:Vamp, could you please move this thread to the Artworks forum, This feels a lot more like a "I can make a dragon" thread than a discussion.

Done

Love it!

I'm going to switch to Blender 2.5, I estimate that will take about 2 weeks (a lot has changed, especially the python API). So don't expect any visual progress during that time.

I was supposed to be logged in...

Sivert3 wrote:I estimate that will take about 2 weeks

Um, more like 7 months of inactivity and 1 week of work.

I've ported a minimal subset to 2.5 and I'm now going full scale (pun intended), But first, I need to make the dragon (and rig it).

Making some progress. A draft body has been completed. The idea for the feet is that they are held towards the body during flight. Wing folding is still a headache.

Nice work there Sivert

The shoulder needs more work, the wings have to be redone from the ground up, the feet does not even have joints. There is no wing fabric, heck it more of a mock up than an actual model. It will take a lot of time to refine the shape and topology, and rig it. And I still have relatively little experience with rigging. When that is done I'll do a proper scale rig and change that dull gray color.
I've always fond the size of the wings to most depictions of dragon to be way off. For small dragons; there no way such a creature can fly with such tiny wings, especially not with a human mount, as for the large dragons; It will not fly on its own regardless of the size of the wings.

The wings on this dragon will not be able to carry a mount. The only problem is folding them together. It just does not work, the joints moves way out of position (not visible on the image).

My skills in modeling is better than I thought, making the second pass over the model a lot faster than anticipated. Almost done with the body. Then it's over to rigging and finishing the feet and wings before a final topology review. Then it's over to the scales.

Looking really good as per usual, even with the long delay!
Might I ask if you could point me at any tuts for rigging models in Blender? I REALLY can't get the hang of it...

@Marik: I never found any good tutorial on rigging. Most of them shows how to rig one specific thing such as arms and feet. Rigging is really just using the right setup of bones to make animating as simple as possible. Keep that in mind and experiment with different setups. Animate the setups and find out what works and what does not work

The rig is going to take time to complete. A lot of experimentation is needed so expect some walk cycles and wing flaps in the near future. The preliminary deformation rig is set up and works! Now I have to figure out how to make a good rig.

You are so silent. No replies in a week. Did you leave this thread? Does it look wrong? Or is it just not your taste? You just leave me here wondering. Do you think I will just continue like I don't care what you say?. Heck, why should I even post updates here, when nobody can come up with something better than "Nice work, please continue I want to see it finished". I want to hear your opinions and ideas, even if they are silly and negative. They will contribute more than a 1000 years of silence.

Making a rig without any previous experience is just frustrating. There is practically no documentation on the animation system. Things just don't work as expected and a lot of things has to bee redone from scratch. On top of that blender does not like me and decided to crash whenever I edit the armature with the outliner visible. I also got caught up creating a gigantic castle in Minecraft Creative.
Create an account on minecraft.net (free) and go to the classic multiplayer server "24/7 FreeBuild on Big Flat Map" (also free) and type "/g sivertsen3" in the chat if you are interested

I reported the crash to the blender bug tracker an got a reply a day later:
"fixed in r40247". I'm going to compile the SVN version of blender, this will hopefully make things better.

No matter how this turns out in the end (At the moment it is amazing) I will be watching, keep posting please

I guess there is nothing more to say than "nice work, please continue".

Walk cycle test (on youtube)