Hello everyone! Before you say anything about how inappropriate it is (which I’ve been constantly being hammered with as I’ve been working on this character) I honestly don’t think it’s something that people should be too fussed about. That being said, the manga, “暴食妃の剣 Boushoku-Hi No Ken”
the manga gave me so many cool scenes that I wanted to learn from, and translate it into blender.
Now, I have never done anything remotely like this before (other than old disney styled character, and turns out that anime and disney style are quite apart) and trying to get the compositing and lighting and all the texture right to achieve that authentic anime- 3D look that other of the artists are doing these days with the youtube sculpting video. (also trying to differentiate myself)
I am trying to recreate that speed and power, as well as making a beutifully lit scene (while staying true to the orignal art) and honestly, I am stuck.
Good thing that you post the updates, the result is much better now.
I have found the same idea when translating a picture from 2D to 3D I will definitely put twice the effort in shaping (shape keys - lattice - mesh deform). It means to forgetting everything about 3D and think only in 2D.
At first I was very anxious because using such methods is totally against to how 3D content works.
But after learning a lot about the “Guilty Gear” workflows used in the game, each animation frame was shaped and supervised by a 2D artist if it looked right. Definitely is the right way to do in these cases.
I actually do wonder if it’s necessary for a 2D artist to outline the shapes instead of conventional inverted hull method. No matter how good that method is, it will never beat the human hand of 2D artist…
This is also another huge factor that plays a role on how 2D-ness a picture has.
The ideal case is to do 100% everything with hand drawn lines, because it looks much more cool. But due to time constraints using inverted hull is essential.
I have watched a dozen of manga drawing videos to get ideas (eg: nihongogamer) and only one suggestion sticks with me the most. That lines not need to be flat, but they must have variations, sometimes very thin, sometimes bold. This is what makes the image look good, otherwise it looks boring.
Still I haven’t found how to translate this to Blender. Such as using modifiers or deforming the actual inverted hull itself.
The only way that it would look feasible is to get the 70% done by Blender (simple inverted hull) but the rest of the 30% must be done manually with hand to get more expression and details where needed. Far better than doing 100% manually or 100% automatically (which might result to boring result).
I also think that a very-significant part of “the reference drawings” is that they do consist of “black-and-white lines,” and this creates the context in which the entire drawing is viewed: not just the character, but “every single thing it’s crash-landing in.” In the reference artwork, the visual trope of “tightly laid-in vertical lines” is used nearly everywhere. The “landing pad” treatment is very-definitely "blasting up" while the character is "blasting down."
You’ve done a really great job with the character.
I still need to do some deformation but I figured I needed to get the lighting right first!
I am now wondering whether I should stick to this kind of aesthetic instead of staying too close to the original art. Though I really need to get the blasting part better, like @sundialsvc4 had said!
