I was prepping for Fan Expo Vancouver, so not quite as productive as other weeks on my personal work, but here’s what I did this week!
I didn’t realize I forgot to post last weekend, so here is a large dump of two week’s worth of warm-up sketches, renders and things I did to learn new stuff in software. This week, as part of #inktober, I modeled and rendered my scenes that I drew in pen, so I’ve included those drawings as well :)
Another week of rapid 3D sketching, mostly using Gravity Sketch, Oculus Medium and Octane, with a bit of Photoshop to tie things together. I try to keep my working time on these to about an hour each, so that I can learn things without getting hung up on little details.
When I don't have a lot of time to make personal work, I try to make environment sketches, either drawn or in 3D. They help keep me sharp, they let me practice and try new compositions and ways to handle textures and surfacing. Here are a couple I've done in the last little while.
What do you do to practice and keep yourself sharp?
I decided to spend a couple of days unpacking the character design logic from Gears of War. My own personal characters have tended towards a minimalism that I got from Alex Toth being a huge influence on me, and I'm trying to broaden my style to make it a bit more current. The GoW style is *too* micro-detail for my personal tastes, but I thought pushing past my comfort zone might be a good way to learn.
I started with a relatively simple design from the GoW3 art book, and did a master copy.
Next, I was interested in how they portrayed women as well as men, so I found a picture of a 3D model of one of the characters, and drew that as well.
As my third and final (for now) master copy, I found one of the characters that was "bursting" with micro-detail, and copied him.
Only after those 3 studies did I attempt to apply the design language to my own character. I didn't try to slavishly copy exactly, but rather to learn the ideas of the master copies and apply them my way.
I had a lot of fun with this process, and I recommend it for anyone who is trying to stretch their visual libraries and sense of design. Learn first by observing and copying, and then try to take that to your own idea.
Thanks for reading!
When I was in Iceland this month, I picked up a book of Icelandic Sagas, the stories from the 8th-11th century that chronicle the population of the island and the deeds of the families involved. Being me, my first thought was, "Wow, this could make a cool basis for a sci-fi story!"
The layout of the Viking world makes it tailor-made for such a thing. First off, most locations are simply farms. There are few cities or towns. Second, travel between them is mostly by boat. Converting that world to one of small asteroid-based settlements connect by spaceship is a very small jump. Each ruler of an area would be the largest asteroid in a small cluster of vassal rocks.
Obviously, my world names and such may change, but right now, I'm thinking in terms of this:
- The North Belt - The largest concentration of settlements. At the start of the story, it is in the process of being unified under one ruler.
- The Ice Belt - Newly colonized and somewhat remote. Attractive to those escaping politics in the North Belt, or those simply looking to make a fresh start.
- The Green Belt - The outer limits of settlements, although explorers have found a planet beyond it.
- Umbria - A planet further in-system than the belts. Rough and tumble by the standards of the in-system planets, still considered soft and easy pickings by the belt-dwellers. Often raided for materials and slaves, although some belt-dwellers have found riches serving as warriors for the agricultural centers on the planet.
- There are more systems closer to the sun that are highly populated and civilized, but they won't be detailed in this first pass.
The Old Norse word for port or harbor is "Vik", and many believe that is the source of the word "Viking" - Basically, to go to ports for goods (often raiding, but sometimes trade.)
For my story, given asteroids as the settlements, I'm going to use the term "Rock" and "Rocking" in much the same way. Rockers are those who leave their homes to search for treasure and slaves and glory.
Visually, I decided to start this one with characters, which is a little different from my usual methodology. I've been taking traditional Norse costumes and silhouettes and trying to update them to a tech look without losing their design language. Here are a couple of the ones I've done so far:
Here's the first of the women I've tried - I think they may change more from the historical, as I have no interest in creating a "Man's World" story, and actual medieval Norse women's garb isn't that suited to deeds of adventure.
Obviously, these are just sketches to get my brain thinking, and not finished works of art, or even concept art.
I am enjoying the process, and I'm going to start thinking about specific characters and locations from the original saga, and how to translate them into a tech story. Thanks for following along!
For the rest of August, I'm going to try to do a 45 minute-an hour material study painting. Here's today's:
I'm hoping this will help me take my 3D+painting technique to the next level.
Ok, we got the house modeled and textured. I'm not really going to show any pictures from the texturing process, as they are pretty boring and I don't think would explain a lot. I used a custom adobe/stucco texture on the walls, with the windows and roof made up of an aluminium texture with some added dark bits in the cracks. I put the whole thing through a "dirt" smart texture in 3D Coat that scruffed it up and added a lot of grime into the corners and crevasses.
As I mentioned earlier, I made the choice to model the house in sections, which I then loaded and created Unreal Engine materials for. This is also a pretty straightforward task, 3D coat creates UV jpgs for base color, roughness, normal maps and all the rest. It's basically plug and play.
Now the fun part! I used the same desert map that I used for my last 2 vehicle designs, but found a different POV and built the house. I did have to increase the size of every component by 1.7, which I figured out by loading the first section next to a model of a human. Doors are great for this, they frame the people and give you a very quick read if the sizes are working in comparison. Once the model was constructed, I adjusted the lighting, moved the camera around, adjusted the lighting some more...in general, just stuff you do to find your composition. The advantage to 3D is that when your design is complete, you don't have to do more thumbnails for scene composition, you just move your camera around and tweak things until you've tried all your ideas and picked the one you want. Unreal lets you create camera viewpoints with "Control+<number>", so you can cycle through different camera angles and compare them.
Here's the Unreal Engine render.
The other great thing about UE is that because it is a real-time render engine, you get great results the whole time you are finding your shot. I used to hate setting things up in Vue, running a 3 hour render, and then discovering that something was off in the high quality render that didn't show up in the preview. This is *much* more efficient.
The rest reminds me more of matte painting than anything else. I took the render into Photoshop, replaced the sky, added some mountains and some foreground interest and did a bit of color tweaking. Since I was happy with my 3D model, I didn't do a lot of paint-over to change parts, but this would also be the time you could "fix" low-res models with more detail if you needed to.
Here's the final image:
Thanks for following along with me through my design process from thumbnails to concept art painting. I hope you enjoyed it, and if you have any questions, feel free to drop me an email :)
After the thumbnail process was complete, I drew my first choice for the house design. You'll note that I used pen and paper - These interim steps are for me, not for clients, and tool choice is almost irrelevant. Do it however you feel like, the important part is to increase your understanding of what the thumbnail represented.
To make sure my understanding was there, I chose to make notes. This can be a really good idea if you want to share with an AD or teammate before you go further.
After the drawing was done, I jumped into 3D-Coat. Because of the voxel nature of 3D Coat, it is great for loose 3D sketching, but pretty heavy on system resources. For this reason, I went back and forth on if I should model the entire structure as one thing, or build it in components. In order to keep sizes consistent, I opted to create each part as a separate section, but then down-rezzed it and made a *very* low-poly reference model of the entire thing. This ended up being a very good thing for me, as during the the texturing phase I accidentally overwrote one component, and having the entire house let me recreate it very easily.
There's the fundamental design for the house! The next stage is to quickly texture the sections in 3D-Coat, and then bring them into Unreal Engine to render with an environment.
If this were a studio project, I would definitely share what I have at this stage, as it would be very common for the AD to have notes of things to change, and this sketch phase can be very iterative. Since I'm my own AD, I have the advantage of saying, "Yes! You knocked it out of the park!" on the first pass :)
Next step, getting ready for the final painting!
In 2005, a book called "The Skillful Huntsman" was published by Design Studio Press, and concept art was forever changed. For maybe the first time, students were able to see process steps to making entertainment design, and of course, those students began to emulate the process laid out in that book.
One of the parts of that book that got perhaps the most attention was the concept of thumbnail silhouettes. The idea is actually, really, really good - to think of the shapes and read of what you are trying to design instead of getting bogged down in the details that are so easy to fall prey to in line drawing. Of course, when students started aping this technique, many of them didn't have the understanding of form that is REQUIRED to make this a useful methodology.
The result, frankly, created wave after wave of incredibly shitty designs in our industry. Instead of focusing on readability and form, people got caught up in how "interesting" a shape they could generate - usually by adding a shit-ton of spikes. When that didn't happen, the designs were unreadable and not taken further. Portfolio after portfolio were filled with pages of little black blobs that could never inform anyone further down a pipeline with information needed to do the work.
In the last couple of years, there seems to be less of that going on. The word got out that it wasn't a "cool" technique to get jobs. Now, I'd say the pendulum is swinging the other way, towards tighter and tighter linework. That will probably cause it's own set of problems, but is beyond the scope of today.
This morning, I started to rough out some designs for a building in my "Strange Futures" project. It's going to be the home of an isolated old guy living out in the desert, where he has to ward off attacks by mutant dinosaurs. I want to capture the feeling I got driving through the Badlands of South Dakota and seeing old shacks in the middle of nowhere, but also have a bit of "cool" to it, and show the mish-mash of technology that is the cornerstone of Strange Futures.
I'm not ashamed to admit, I did a page of silhouettes. My tendency is to keep things too close to "reality", and I want something with more panache than just an old tin shack or mobile home. Silhouettes almost force you to go beyond the basic cubes of those realistic homes.
Here's my first page:
Obviously, as I moved down the page, I started combining earlier ideas into new mash-ups, which is completely fine. I decided I really liked #14, #17, #18 and #19, so I took those 4, made a new sheet, and started drawing into those shapes to give an indication of what the blobs might actually be.
From these tighter silhouette sketches, I'm leaning towards the top 2. In the next part of this development process I'll take those 2 and completely redraw them, trying to add details and clarity without losing form. I'll also probably start thinking of what they look like from other angles, in a very loose way. After that, I'll jump into some simple 3D modeling to kick out something I can rotate around and get a feel for. I'll share those steps with you guys later on in the week :)
Rather than an opinion piece this week, I thought I'd share some work I've done on a personal project over the last 2 weeks. I wanted to work on my take of a "Superhero vs Aliens" movie story, set in Vancouver B.C. Style-wise, it was based on an "X-Men meets the Avengers" vibe, with normal people who have superpowers instead of costumed heroes.
My four "superheroes" are all college students at Simon Fraser University in Vancouver. I don't go into how they got their powers, but Nicolas has telekinesis and can fly, Lauren is your basic "Super Soldier" archetype, stronger, faster and more sturdy than Olympic athletes. Her boyfriend, Calvin, has telepathy, and Jenny can create destructive energy blasts from her hands. I will admit, it made my day to create a team with no beefcake white dudes, and where the white woman has "normal" human proportions, and a bit of cellulite.
As Lauren and Calvin relax and watch the summer fireworks...
An alien breach team gates in. They have been monitoring via deep-cover agents, and have an agenda before the full invasion begins.
Part of that agenda is removing potential threats, like our four heroes. Nicolas narrowly escapes an explosion in his apartment.
The alien spaceship has come down to Earth and is waiting in northern B.C.
When the assault team fails, fighters are dispatched.
They make the trip down to Vancouver as the mothership moves more slowly behind them.
Lauren and Nicolas engage the fighters while Calvin and Jenny sneak back to the mothership through the closing warp gate used by the assault team.
Jenny takes out some guards and she and Calvin advance to try to find some way to stop the ship.
Unfortunately, they are captured just after discovering one of the housing units for the mothership's power cells.
With no time to spare, Nicolas leaves Lauren to fight on the ground, and goes to rescue his friends and help them stop the invasion.
He sneaks his way aboard the vessel and is reunited with his friends.
What happens next? Not sure! I've got enough paintings for this project for right now, so I'm actually getting ready to start a new project. I really find this "storytelling approach" to development really helpful for keeping things interesting and consistent. You don't have to have a full script mapped out when you start, often as you are painting, new things will come up. There was originally going to be one more character, Jenny's 11 year old young brother, but as I started working, he didn't fit where the story was going, and he was dropped.
Timing-wise, this was about 2 weeks' worth of work. I tried to do at least one image a day, plus some paper thumbnailing of ideas. Technique-wise, this was a bit of a mix. A couple were straight "Photo-plate and painting" images, some were Daz Studio and Photoshop, and some were full 3D environments kitbashed in Unreal Engine and then brought into Photoshop.
They were not painted in story-sequence order, I jumped around all over the place. If I were going to keep working on it more, obviously I'd need a finale, and I could really use an image or two of Calvin being heroic....telepathy isn't an easy power to make look dynamic!
Thanks for reading! Two weeks today, I'm flying down to Spectrum Live! I'll have a new sketchbook with many of these paintings, plus some other projects. If you're going to be there, stop by and say "Hey" :)
So, Monday was the deadline for the Ancient Civ keyframe and environment design challenges, and I submitted stuff for both of them. I thought I might talk about my ideas, how they led to world building, and then my final paintings.
So when I started thinking about this, I knew I wanted to keep it based in a "movie reality" version of the real world - no magic, no ancient aliens, but still allowing for things that would never happen in the real world. Probably close to a "Laura Croft" universe. I knew I wanted to show it as a living society, not as some Indiana Jones discovery story. Finally, I *really* didn't want to tell yet another Orientalism appropriation story.
With those as my drivers, I started brain storming. I threw down a bunch of thumbnail sketches (they're all up on my Instagram if you're curious), and I kept coming back to an iron age Pict/Celtish inspired village. What I ended up with was an isolated, idyllic village inside an extinct volcano. I figured that the volcanic activity heated up the area, and created an artificial green-house effect, so that the temperature and terrain was sub-tropical jungle inside a volcano in someplace Northern, like Iceland. The story I worked up was that they were refugees from a war several hundred years ago that found this isolated place and made a peaceful, happy home there, secret from the rest of the world. They subsist primarily on farming, birds and a bit of fishing, as the ground is mostly swamp and they don't have a lot of room for grazing.
Now, the story - It is not fully fleshed out, as I was looking for key moments and not a full script, but the core of it is this: The protagonist and antagonists are the twin children of the village chief, a boy and a girl. The daughter is technically first born, and will lead the tribe. The son does something that endangers the village and his father exiles him. Outside the volcano, he meets a warband of viking slavers, and saves himself by telling them about the easy pickings in the village. He leads them back to his former home, where the reavers torch the village and kill the chief. The daughter confronts her brother in the lava temple, and unleashes the flow to destroy the warband, at the cost of their home. She leads the survivors out of the volcano to try and find a new life.
With that, here are my keyframes and designs:
I'm reasonably happy with the work I've done, although looking back at it, I think I might change up my keyframe choices a bit to sell the project better. These ones are necessary, but they might not be the best 4 to showcase the project.
I really had a fun time with this project, and learned a ton. There really is nothing like working on a contained idea to force you to grow. I have signed up to do the matte painting challenge, where you have to take an environment from the first phase and make 2, photo-real matte paintings based on it. Should be a good time!
If you worked on this challenge as well, I'd love to see what you did for it :)
Thanks for reading!
Hey guys, I don't have a lot to talk about today, but I thought I'd share some of the copies I've been doing recently.
So, as of 2 weeks ago, I had done two Sargent studies. I did a few more of those, ending with this one:
After that, I did two J. Allen St. John studies - If you don't know him, he's kinda the "proto-Frazetta" of the 20th century.
I wanted to try a photo-study instead of a painting, to work out my own ways of simplifying shapes and values.
To mix things up, this morning I went into DAZ studio, rendered out a figure, and then did a study of that.
I liked that one, but I didn't have time to "bake in" good lighting, so I spent a bit more time today pushing things past where I'd normally take them, and then putting away the reference and adding lighting to make it "pop".
To me, this is the point of doing these copies. No one cares if you can replicate the work of an ancient master - you're doing it to push yourself to try new things, and find new ways of problem solving. After you've done that, you should try to apply those tools to your own work. I can see some of the things I learned from the Sargents and the St. John studies in today's painting, which is as it should be. I also feel like I'm learning a ton and getting better very quickly. Even just doing 1-2 hours of studies a day has had a huge impact on my ability to straight up paint and render. All of these were done on one layer in photoshop with no textures or photos.
Anyway, I'm sorry I don't have a more text-heavy update, I hope you've enjoyed seeing some of my progress :) If you start doing daily studies like this, please, let me know your results :)
Hey guys, sorry I missed Thursday's update, but hey, better late than never, right?
This last week, I've been trying to paint outside of my comfort zone. I've been doing "straight" painting, mostly one layer, no photos, no 3D, based on a drawn sketch. It's been fun, and frustrating. I feel like I got a lot of out of it, which is awesome - But one of the things I got out of it was confirmation of some fairly large weaknesses, which feels less awesome.
I've ALWAYS hated learning to do other people's stuff. I hated learning other bands' songs when I was a musician, and I've never liked mastercopies as a visual artist. After seeing my shortcomings stare me in the face on Thursday though, I am biting the bullet. I could figure out how to solve all my problems on my own, but WHY? There have been painters for thousands of years, and studying what they've done keeps me from having to re-invent the wheel.
I'm starting off with Sargent, 'cause I love his stuff, and because he's great at characters, which is definitely an area I need to improve. I started yesterday with a one-hour study of Madame X:
One hour isn't enough to get the subtle brilliance of Sargent - but I can do it as a warm-up at work most days, and I'm hoping at the stage I'm at, we're talking 80/20 rule.
This morning, I did another hour study, this time of a Sargent self portrait.
If things go as I am planning, tomorrow I'm going to try for a longer session, and one that has multiple figures and a background.
I am also happy to report that I'm having more fun doing these than I thought I would. Instead of being tedious, I'm seeing it as a puzzle, and a chance to really push my observation.
Because it's what I want to work on, all of these, while digital, are done on one layer. I am sampling colours sometimes from the original, because I know I can match colour, it just takes longer. What I'm focusing on is strokes, and *which* colours are laid down next to each other. Remember, when you are doing a master-copy, there is no such thing as "cheating." You can do it 100% by eye with no tracing or sampling if you want to, and that will be helpful - but you can also chose to focus on the things you feel you need the most work on, and recognize that you don't have enough hours in the day to do everything the hard way.
Anyway - once again, sorry I'm 2 days late, and thanks for reading!
Yesterday at lunch, I drew these 3 thumbnail sketches. Each one took about 2-3 minutes.
This morning for a warm up, I figured I'd paint them into quick colour-comps. I started by fixing the borders of the drawings and converting to pure black and white.
After that, I spent about 25 minutes in total painting.
Supposed to be late evening in the forest, just before the sky turns colour for sunset. These were all painted on one layer, with one brush (the standard chalk brush in Photoshop with transparency and brush size turned on.) The point of these is not to have finished paintings, but to be able to see a little better what colour and lighting would do to the locations.
Busy day at work, and I don't have a lot of cycles for a blog post, sorry :( Rather than skip though, I thought I would share two paintings. The first I did recently, the second I did 4 years ago today, at least according to Facebook.
Improvement happens. It doesn't happen as fast as you want it to. It doesn't come easily. It does happen.
Just keep swimming.
Ok, today is election day in the states, and I'm too stressed for a full painting analysis blog post...but I figured I'd share the process work I did for this film festival poster.
My only brief was to "Make a cool poster." I went to the festival this year, and there was a sweet French cartoon that reminded me of Moebius' work, so I thought I'd take it in that direction. I started with a page of simple thumbnails:
Definitely borrowing some of Moebius' compositions...but if you're going to do that, do it from the best!
Nothing leaped out at me as "good" or "bad" from those 8, so I decided to take all of them and make colour comps. I just took a picture with my phone and mailed it to myself. In Photoshop, I set the layer as 'multiply' and then put colour underneath.
I showed these to the client, and she liked all the ones on the top row. For the final poster, I drew it in Photoshop on an 8.5x11 image at 600DPI. I changed the colours around a little bit, but I followed my ideas pretty closely. Here's the final poster:
If you are an animation student, I encourage you to enter, the show was very popular this year, and there are cash prizes!
Ok, that's all from me today, sorry for the short entry, but real life intrudes. If you are American and registered, GO VOTE. If you are anything else, find a registered American and tell them to GO VOTE.
Thanks again! Talk to you on Thursday :)