The main project I embarked on last year during the lockdown was this 3D food truck scene that's based on a concept art by Olga Permiakova. While I have attempted handpainted models before, this is the first time I am trying out on a scene of such a scale. Initially I wanted to use this opportunity to try out modeling and texturing in Blender, but I decided to go back to 3DS Max as there were too much new information to absorb. I will make it an effort to use Blender next time!
Some additional renders of the scene
It's been awhile but I have been working on the Schoolism assignment. As it turns out, I kinda misinterpreted the previous assignment and went off course. Regardless, I will be taking whatever I learned from the lesson and using that knowledge to apply to whatever I intend to do moving forward. Here's my second "assignment" where I include Substance Painter into the workflow.
I have been meaning to improve my ZBrush skills in preparation of possibly teaching the subject next year. If you follow my art, you will know that I dabble my feet into 3D once in a while. During my days in DigiPen, we were taught 3DS Max and ZBrush. I really want to improve my digital sculpting skills and eventually move on to high level texturing via Substance Painter.
Last month, I enrolled myself into the Justin Goby Fields' Schoolism Introduction to ZBrush course. However, due to the ongoing Inktober event, I didn't really have time to invest all my time into the assignments. In this post I will share with you what I've done for the first assignment.
The assignment is to sculpt 5 busts. I tried to be a little ambitious and added colors and lighting to better present the final image. I also purposefully selected subjects that were different from one another. Some of them turned out pretty bad while others were fine. Stay tuned for more diary entries of my assignment progress.
"Fire Boi" started as a recap exercise for myself before I started training the kids for WorldSkills. However due to other work matters, external projects, other personal art projects and of course my personal life, I was only able to work on this intermittently. Well, I'm glad to finally call this done for now (don't intend to rig this unfortunately).
Moving forward, I will be focusing more on polishing up on my ZBrush skills and learning more about Substance Painter in preparation of the next WorldSkills. Thus there will be less "big-scaled" 3D projects like this one and more small practice assignments.
The concept art of this model is from Tooth Wu. Do check out the rest of his amazing artwork.
I have left my job at The Good Mood Creators at the start of 2017 and have been working on bolstering my 3D portfolio. I'm focusing on generating more realistic assets.
Here's a 3D character game model that I've been working on this past week. The character is based on a concept art by Aleksey Bayura. It was sculpted in ZBrush, retopologized in Topogun, textured in ZBrush and Photoshop and rendered with Marmoset Toolbag 3.
Modeling a small village/town has always been a project that I wanted to pursue ever since learning more about environmental modeling from my full time job. I recently found some time to attempt it and I began gathering ideas. I knew I wanted to skip the concepting phase so I decided that I will model something based off another video game.
Initially, I was contemplating on doing the entire first level of Metal Slug X. After spending one whole day just sculpting rocks in Zbrush, I knew I would not have the patience to complete the whole level. I decided to give myself a one week duration, and turned towards the sandy beaches of Costa Del Sol, from Final Fantasy VII.
The original environment looked a little dated, so I decided to do a quick paint over to re-establish the colors and to clarify some of the ambiguity in design. While doing the paint over, I quickly realize that the buildings were in fact not scaled properly. The van in the background is as tall as the opening of the door. It made sense back in the FFVII world as characters were represented by a 'chibi' version due to technical limitations of the Playstation. I decided to stay true to the roots and not edit the building proportions.
The gray box models were massed out to give a clearer picture of the proportions in the 3D space. I was debating between whether to include ZBrush in the workflow, as it wasn't particularly necessary. I decided to put off the decision to a later date and concentrated on the textures first.
Unlike the Class 30 model, the workflow for this project began with the texturing phase first. I allocated what I felt was sufficient space for each element in the concept based on the screen real estate they occupy.
I underestimated the time required for texturing, and the whole project was extended by two days. Eventually, I settled on not using ZBrush, as that would further extend the production period. The normal maps were generated using nDo.
I tried doing the final render using XoliulShader, but the transparency from the opacity resulted in a lot of noise. I went back to Marmoset toolbag to finish things up.
Overall I'm quite happy with how it turned out and definitely learned a thing or two from this project.
A few weeks ago I completed Modern Warfare 2 (Yes, 2. I'm so far behind on the latest games) and I was pretty impressed with the graphical quality of a 4-year old game. I was inspired to model a military vehicle and was contemplating between the good ol' 5-tonner (soldier transporting vehicle used by the Singapore Armed Forces) or the Class 30. In the end, I selected the Class 30 as it was a more unique vehicle and it was related to my vocation, field engineers.
The Class 30 is a mobile trackway vehicle that lays down vehicle tracks on terrain that are unsuitable for vehicle movement (like soft, muddy ground). As field engineers, our role is to 'unroll' the gigantic trackway that is coiled up at the back of the vehicle. It is one of the most dreaded task to befall a section during a military exercise as it is labour intensive and the entire operation of unrolling and keeping the track can take between 45 minutes to an hour.
I opted not to use Zbrush for this model, partially because I'm feeling a little lazy. However, most of the hard edged details could be easily achieved using nDo. This is also the first time I tried using Marmoset to render, and I find that it's a pretty powerful tool.