Development-Exploratory Practice Blog
- Crane ZHANG
- Mar 2, 2023
- 6 min read
Updated: Apr 19, 2023
Week 1
This week's tasks
Click to jump to the relevant page
⭐ means I like it a lot
√ represents I think it's relatively okay, can refine and look again
mood board
Week2
25/2/23 ~ 28/2/23
Draft Development
I first chose the role I want to develop, and then a comprehensive consideration of the background, design, and my personal preferences to choose three, drawing further development.
While I was developing the three sketches, I re-watched the Watchtower character background and researched the character design. In fact, I actually disliked the game's story background a little bit before, but I suddenly fell in love with the game after I learned more about it and became enthusiastic about making this style of character.
And after the second development, I learned a lot and connected my sketches to the Watchtower story, which made me feel that my characters were more alive, such as Red panda from HorizonLunar Colony, a tree man character archetype that fits Reinhardt's backstory, and Ramattra's rebellion that also made me think of The third role that I ironically had.
I was torn between choosing one of these three as the final product. I actually personally prefer to make small creatures, and at the beginning I always intended to make the first character. But after I asked people around me for their opinions and tried to refine them, I found it difficult to convey my feelings about Walker to others.
My second favorite character is Ford, because in my heart he is also a living being, and his story has a sense of drama, so after getting Robert's feeling, I also decided which character I want to refine.
27/2/23 ~ 1/3/23
Concept art
After reconfirming that, my mind actually started to conjure up a series of moves for Ford, who I imagined to be a super internet celebrity and to have a sports car-like shell, with structural references to Orisa, and I also did more research in Overwatch.
When I refined it further, I found that it was much easier for me to produce a creative sketch than to refine it, and I needed to put more attention as well as thought into it.
One of the main points is that I need to refine again at the same time to keep the feeling from a distance to keep consistent with my sketch, while taking into account the movable range of mechanical structures (I think I will continue to experiment with 3d afterwards), while I can think about the small ideas of various parts, such as the location of the link cloak, after which may be, for example, open large when projecting three-dimensional projection, appearing the whole character is very dazzling, and the design of the sword unfolded from the hand.
Week 3
2/3/23
When I got home from class today, I finished the sketches of the legs and shoes.
3/3/23
I did a lot of work today. In the morning, I made a scene from a collaborative project, drew the material and adjusted the lighting to build the scene.
Today I colored Ford and drew a draft of the back, but I'm not very clear about the structure of the small arms, I plan to do a simple 3d draft after.
And after the first two weeks of drawing, I became really good at drawing with Photoshop. I drew this character's lines for three days before, but I believe it will take much less time back there.
And I also have a better understanding of perspective, although there are still some problems, but compared to Sketch, I have to Concept's leg perspective has become much more comfortable.
I also did research on Watchtower's Concept Art when drawing the concept, such as color rendering, line style, most obviously the background color, and even some small mechanical structures, which helped me a lot.
4/3/23
I used zbrush to make a simple shape
After refinement, I found that it was a bit out of shape, so I took yesterday's as a reference to change the shape, and I should copy a few more layers and put them underneath as a reference, I think this is one of my weaknesses, I'm still a bit unsure about the perspective.
5/3/23
Today I focused mainly on finishing the back sketch and coloring, and tried my best to lean towards Blizzard's Concept Art, and then also tried some different color combinations.
Color
I tried the classic color scheme of some different sports cars, and to be honest, I prefer black, but it doesn't go well with the character background.
Next I'm going to model in 3d, I'm not going to draw detailed three views because for me I like to explore more in 3d, and I'm very clear about how I want the character to feel, and I'm not 100% accurate about perspective, instead of making inaccurate three views, I prefer to keep exploring in 3d.
6/3/23
I watched Blizzard designers, modelers workflow, and my very favorite artist zhelong xu, they are not drawing specific three views, but they are very skilled for sculpting, in my experience, is to capture the soul of the character, that is, the characteristics, and show it in the production, which is the most important.
And the importance of the three views is clear to me, that is, when the team works together, the concept design should clearly communicate their ideas to the modeler, but I think then the modeler is just doing repetitive, uninteresting, and not much creative work anymore.
Block out
Because zbrush is very well optimized for tall models (blender often flickers when sculpting tall models), I was going to try making them in zbrush, but I found I was more proficient with blender...
8/3/23
Week 4
9/3/23
Because I thought the lighting would give me a fuller view of the character's shape, I divided it into two collections so I could quickly switch between them, and I added keyframes to the lighting and camera so I could quickly switch between the front and back on the left side.
11/3/23
Modeled the hands, refined the face, legs, added the cloak
Overwatch Pioneer latest linkage for my performance Overwatch cloak on help a lot!
12/3/23
I wanted to use sculpting to draw out the shape first and then topology afterwards, but I found that many parts could be modeled directly, but I stuck with the unified process and next I plan to keep the topology correct throughout the model.
15/3/23
Different colors and materials were divided, cloak sculptures were made, the range of movable joints was thought about, and many things above the shape were adjusted, such as the thorax, the large structure of the pelvis, and the small arms were optimized.
Week 5
16/3/23
Trying Retopology today, and having made several iterations of it in
But there are still some problems, which really bother me a lot
Luckily, I have Orisa's model as a reference, which will at least ensure that my final wiring is not too far off
At the same time, I restored Orisa's model, his head has a total of 800 faces, and then if you remove the triangulation, it should be more than 400 faces, I now have more than 200 faces, after adding the structure inside should be in a right category
Adding an extra looped edge on the head worked much better.
I tried to do the best I could:
But I found that his shape because I did not adsorption caused and my sculpture does not match at all, so I chose to give up the perfect curve, as far as possible to fit the original look, the other I want to continue to adjust more in the details of the sculpture after
Spring Break - Week 6
(use 2 week)
I used Mark sharp very frequently during this topology, which is crucial for hard surfaces, and to think about the edges that are laid out during the topology.
However, when I was working on it, I found that it was more of a design process because I didn't have the final high-model version before I made the topography, and because of this I had to cross-reference it with the concept art and the sculpted drafts several times.
and use the sculpted Move brush to adjust the shape when adjusting
Week 7
Ended Unwrapping UV and split the main blue material into two texture sets with the rest of the material.
Then bake in Substance painter
19/4/23
Modified: the wiring was adjusted, some uv's were missing, and there was a problem with the sharp angle setting causing black spots.
To be honest, the smoothness of the planes is still problematic and there are unfixable jaggies where the meshes overlap, but they still need to be put together for procedural curves and AO
After debugging I found that a baking size of 2048 has no black dots, if it is 4096 it has...
Comentarios