VUE without competition
Can you please give us a walkthrough how to implement this into Maya? would be super helpful. Thanks a lot.
Vladislav Kudrya shared a very interesting workflow, which allowed him to build complex 3d space station environment with soft and hard elements.
My name is Vladislav, I’m 22 years old. I grew up in Russia, in a small town near Novosibirsk. Yeah, that’s the place with bears on the streets and people playing balalaika from time to time. Since my childhood I’ve always wanted to paint and create new worlds. I started with art school and continued with university, where I got my Industrial Design Degree. Now I finally got some time to work on my portfolio.
It took me long enough to choose profession: I’ve studied web design, interfaces, then switched to product design and motion graphics, ended up in volume production. All that was really interesting, but I wanted more. I wanted to create characters, some visual shapes, which stuck in my head, but my skills were not advanced enough.
I don’t have much experience when it comes to the industry. I haven’t had a chance to dive into 3d during the first two years at my university since the eduction there was mostly about classical art and painting. I’ve been doing it as something more than just a hobby since the third year at the university, so it’s safe to say that I have about 2 years of experience in serious 3d. I’ve contributed my art to several indie projects and ads. That was cool as I didn’t spend too much time on these projects, managing to find time for the university.
That’s an unusual story. My friend once asked me to design some bags in Marvelous Designer. This task wasn’t too hard and time-consuming since I knew how to work with the software. He sent me some references and I created 25 different designs, which he could use in his project. Then I sent these to my friends for their criticism. They actually liked my work and asked to create more. That’s how I ended up with 100 elements for kitbash and a new extraordinary project.
Using Maya and Marvelous Designer
All the soft elements were created in Marvelous Designer. That’s wasn’t hard – I created primitives from Maya and imported them as avatars in Marvelous Designer. Then I created main shapes and duplicated geometry to get the final result.
Below you’ll find a video from kitbash, which shows my production process. Repeat the process and you’l get the exact same result.
All the other hard surface details were created in Maya. Final composition, lighting and rendering were made in 3ds max because, in my opinion, it is the easiest software to learn. But when it comes to modeling I choose Maya as I’ve changed so many hotkeys there, speeding up the process in a way no other software would allow me.
Still, if you are not that strict about Topology, you could use Cad Modeling, software like Fusion 360 or Mol 3D that bring good results really quickly.
In the video below I’ve spent an hour on Useless Cylinder. This task would have taken much more time if I’ve chosen classic modeling approach.
Modeling Hurd Surface Elements
First of all, you should learn your package, its basics and tools. Then you should work on your own hotkeys and scripts. I started my first day with subdivision modeling, making basic shapes, but then realized I spent too much on cleaning up and optimizing my mesh. I purchased and used ADN Modeler Tools and CreasePlus. Those scripts can do miracles! You create a form, cut it with booleans, use Shape Shifter and get the result you need! I didn’t have enough time to learn the software in detail, but I still got the amazing result.
I highly recommend studying other projects’ meshes, on Pinterest for example. That would help you a lot if you are at the very beginning. Also, don’t forget about watching experts’ tutorials. All the techniques and tricks are out there, you just have to find them and adapt for your pipeline. I highly recommend watching tutorials by Tor Frick, Vitaly Bulgarov and Simon Fuchs. These are professionals with a deep knowledge of different software packages, so you might find something useful for you own projects.
Assembling the Scene
I created all the elements as blocks so that they would fit my proportions. That solved most of the problems with combining elements.
I always do block-out of my elements – it helps me understand what is required in the specific situation. Then I do a screenshot overpaint in Photoshop to define the level of detail needed and to plan the scene. Then I start adding the details. At the very beginning you also should define the final composition that would make you scene more attractive. You can really benefit from that step, so don’t neglect it!
For example, when creating soft elements on the wall, I used only two variations of the panels, but once I reflected them horizontally and vertically, I got 8!
I’ve chosen Corona Render, because I knew it quite well and I’ve had a chance to work with this software. It’s a simple one-click renderer that brings a quick result.
For example, the render above was done in Arnold. When I was assembling the stage it was important to understand what the spectator would look at first and, also, to set the general mood.
It took about 10 hours to render the scene in Corona. I don’t have a super-powerful computer or farm to render scenes in 30 minutes. There is nothing wrong with not having a super powerful computer to work on your ideas, it’s just a tool. So try another way.
In fact, I’m just learning how to set up the right lighting. I advise watching the way lighting is used in films and photos. 3D is not that different from the real life, so try to understand how it works, and then repeat what you’ve learned. Try to do as much as possible in 3D, so that when it comes to post-processing you have less work left. Not sure about your lighting? Make another render pass with different lighting, experiment. There is no formula that would work all the time.
Each light source should be set up separately and sequentially. Good lighting is half the work, while the poor one would spoil your scene. I thought about frame composition at the stage of block-out, then intentionally used a red light to guide the viewer’s eye.
If you want to learn more about Lighting setup, I recommend reading Matthew Scott Cinematography Blog as he handles this task much better than me. And don’t forget about learning from everything that surrounds you!
When it came to final composing, I tried using render elements to finish the work.
Indirect, Dept and Emission passes are really important there. Then I just had to break the composition for different shots.
Would it be possible to achieve this kind of quality in a real-time render?
Yes! The scene could be even better, but it would take 3 times longer as you would have to make the Low poly model, unwrap, bake all maps and create final texture set for each of your elements. Then you would adjust shaders in some editor and bake lighting. You can do it, you just have to spend more time on it.