Chose this as one of my developments as I had already invested two weeks of work into it (vs. one week for the other projects) so it seemed to make sense to take forward the project with the most work. I had some goals that I wanted to achieve from my initial starting point:
- Creation of a low poly, game ready model of the Dunk.
- I don’t focus much on prop design, so I felt that I should do something there with the Ink Grenade concept.
With some help from tutorials and classmates I took the highpoly model, did some re-toplogy and baked the details down from the zbrush model. I then had a nice clean model that I could take into substance painter and add some textures too – as this was something I was keen to do from the early concept stage.
Some visual research into tropical fish markings led me to a hybrid design that made sense within the context (red coral in the environment for hunting camouflage).
Here’s a screenshot from Substance of the low poly model:
Taking a note from artists such as Peter Konig I did one last paintover and added some environment lighting effects to visualize how it might be presented in an in-game scenario.
This project has been an invaluable way for me to learn the asset creation workflow. Even though my main focus is on the design and conceptual stages I knew that I wanted to go through this process to understand the technical limitations that my designs may come across later on in the process when handed on to a 3D modeler to create.
The Ink Grenade
Further elaborating on an early concept that I came up with for the diver to use as a means of escape (evasive cover underwater in the same way cephalopods do), I decided to develop the idea into something more functional looking with a little more thought put into the design.
My first design iteration (after a thumbnailing process) had a corkscrew around the outside that would give the grenade movement through the water, further dispersing ink that it shot out.
Unfortunately when I came to make a model of this (using Modo) it was incredibly… phallic, and I ended up making it more of a ‘squat’ design to compensate against that.
I actually did not like that at all and went back to the drawing board, creating a much more compact design, mirroring that of a jellyfish more closely.
As I’ve always been very interested by the ‘Art of Star Wars’ books with all their schematics and exploded views. I feel that they present a very clearly thought through universe (although fictional), so I was keen to bring elements of that into my work.
Final image of a swarm of ink grenades. These were all rendered in Modo’s inbuilt system using the basic materials in the Modo library. It is a PBR based system and it was a joy to learn about specularity, roughness, the refractive index and all that good stuff that as a 2D artist I was aware existed, but never paid a huge amount of attention to.
For me the process is all about creating and refining designs, so I’m happy that my use of 3D highlighted an issue that I was otherwise oblivious to had I stuck with 2D for this project.
Always start with a nice warm-up painting! Wanted to be nice and loose with this as it was only meant to be a warm up, but I wanted to have a go at loosely suggesting texture/materials with a good ‘brush economy’
Have been reading up on ‘Brush Economy’ – where old masters of painting, and looking at the work of the late Jon Singer Sargent. Although his use was mainly in portraiture, the theory still applies to whatever the subject matter is. Matt Korr from Ctrlpaint.com describes this perfectly:
“…when an artist evokes a rich, detailed, image in the mind of the viewer through implied detail. Where photographs shine is in their specificity, though paintings offer a different opportunity: simplification”
So I tried to apply the principle, making brush strokes in photoshop as ‘efficient’ as possible, whist still trying to define different materials (the lighting affects the ‘globules’ on top of the head of the mushroom in a different way as they are translucent).
After exploring design shapes via my sketches and the reference board that I put together, I realised that this guy looks a little ‘clunky’ and unable to quickly manoeuver away from the fearsome Dunkleosteus. Time to deign something to even the odds!
Taking the ‘base generic’ concept I discussed in an earlier post, I start making adjustments to create visual interest. Made helmet more octopus shaped…
Added decal details and rust to suit to make it more convincing. The octopus decals are a nice touch I think that fits in well with the steampunk aesthetic.
And he’s got an Ink Grenade to use as a distraction to escape Big Red. I was going to add more weapons and bits in the way of defence, but time was of the element, and something else would have suffered for it, so I felt it best to balance this evenly as my main foucs for the project was creature/character design, not props.
Final design! Overall fairly happy with this, it’s a solid base to hand over to a 3D modeller, and I’ve made it in such a way that all the elements are individually separable. It’s easy to see the definition between the undersuit/belt/gloves/tank/chestpiece/helmet. I may even model this myself when it comes to the later project.
So the whole project is published on Artstation here, so I guess that means it’s official.
Had a lot of fun on this one, was pretty good to force myself to use ZBrush again as it’s been so long and I picked up some good techniques and tips from Lee Devonald.
So I was approaching this from an as ‘logical’ approach as I could, I had done some flat images for the turnaround and brought them up in ZBrush to use the image plane to carve them out. It was taking forever and the result seemed very… stiff?
Then, as per Lee’s suggestion, I ditched that approach and just went about sculpting it organically. I had already done all the visual research and structural sketches (which really paid off) so I knew exactly what the main shapes were and how I wanted to block it out.
Using a combination of pretty much just x3 brushes in ZBrush – clay tubes, move and trim dynamic – I was pretty oblivious about how to use them efficiently, so I’m glad that I got some insights into the modelling process.
The finished model, rendered in ZBrush in the standard ‘Basic Material’. I’ve a lot to learn about the rendering process (specifically PBR, which I’m looking forward to getting my teeth sunk into), so for this project I just wanted to finish up the model as best I could.
As a concept artist, my process is about the actual design, as opposed to creating an actually usable asset. My mission is to communicate the nuances and shapes, and so at this stage I’m happy to leave this at it’s current level.
This project has opened my eyes to working more in 3D within a concept art pipeline – something I’m a lot more comfortable with now. I’ve been looking at work from a lot of concept artists who incorporate 3D – Luis Carrasco is one of those artists and has a tutorial about his pipeline using the tool.
Selection of the work I’ve done thus far for this project. Will be presenting on this later and will be good to get feedback on designs.
Initial rough character sketches:
Studies of the Dunkleosteus – it’s important for me to understand the structural breakdown of a creature at an early stage, that way I have a solid base to build off and make design changes. Here I’m doing some quick notes about the jaw motion, skeleton and the tail/fin structure and how I could change that to affect the overall silhouette.
I’m using a similar technique here to concept artist Scott Robertson, (although it’s widely used in the industrial design industry), where you draw from a very structural point of view, indicating form with line. I’ve got organic forms here, whereas Robertson mainly focuses on hard surface designs, such as mechs and cars, but the principle is the same. He’s also much better at this than me!
Rough sketch of diver. My first iteration of design is to usually go with something ‘generic’ and expected, and then work on tweaks from that. As a concept artist it’s important for me to be aware of the fact that these designs have to look good, but they can’t be ‘different for different’s sake’, so I’ve done some visual research on actual Victorian-era diving suits to get an idea of the shapes, and what I know is ‘functional’. It’s then a matter of creating interesting shapes from that.
Quick anatomy study, there was no real need for this, but practice never hurts.
Face designs. It helps for me to limit the scope in this, I don’t want to have to go through the design process for every single element of the character because of time constraints. I keep things quick here by painting up a base and then doing some fast overpaints of it to have a quick look at scarring/facial hair etc.
Silhouette breakdown of the Dunkleosteus based on other prehistoric fish, as well as existing fish fin shapes.
Dunkleosteus skeleton analysis
Dunkleosteus final shape design + front view. I want to create a model of this, and as a beginner in 3D (have a few models under my belt, but nothing to write home about), I think it’s important for me to have these views to refer to when creating it.
Dunkleosteus texture explorations – part of my design thinking was that the fish would be red to blend in with red coral as camouflage (hence the backdrop that I made). This would be a logical decision as a large predator that’s bright red would otherwise have difficulty sneaking up on prey.
First mission is to decide what Red/Blue are, and why Red would be chasing. After doing a bit of research into horrors of the deep, I decided that I wanted the protagonist, of this project to be the mighty Dunkleosteus – a 6 metre long pre-historic fish from the Devonian period.
(For those of you wondering, that is pronounced as ‘Dun-klee-os-tee-us’)
I’m usually careful to avoid ‘secondary references’, for fear of being tainted by a very good design idea that I would copy, or something along those lines, so have been sticking to fossil records and paleontological artist renditions of ‘The Dunk’
With all of these things, research is always going to help when creating a convincing design. In this case this was an actual creature, although as it is extinct, and the only remains that have been discovered are the bones in the head/jaw (as the rest was cartilage and lost in the fossilisation process). This is good for me as it gives me a certain amount of creativity and room for design.
I like to watch documentaries and read up about creatures as it increases my knowledge of biology. There are all sorts of adaptations and problems that nature solves in very interesting ways, and applying this thinking to my design process enables me to create grounded and believable creatures that are also visually interesting.
Many artists take this approach, especially renowned creature designer Bobby Chiu who has a very solid understanding of a huge variety of animals, and uses his design knowledge to take elements from the real world and convert into an interesting design.
Obviously as I’m a graphic designer, I start the project with a good old fashioned mood board and logo… going for a nautical/steampunk theme here, undecided who is going to be red/blue at this stage.
My mood board on pintrest is right here.
On a side note, I love pintrest as it’s such a rich catalogue of creative ideas and sources of inspiration. There are some really fantastic reference images on there and I use it all the time to quickly generate boards for projects that I’m working on.