Doing some photoshop tutorials from Aaron Blaise youtube channel. I'll need to use photoshop to create some ornate pattern designs for my 3d project soon but I'm kind of slow and uncomfortable in photoshop, so just doing a variety of tutorials until working in it feels more natural. From there I can look up techniques as I need, but I like to have that basic level of comfortability first.
This isn't great artwork, but I had a lot of fun doing it. So I win, okay? I really like the simplicity of 2d, and how flexible and forgiving working in PS is, especially for somebody who can't draw a straight line.
Another Aaron Blaise photoshop tutorial. Had a ton of fun with this and definitely am more comfortable working in photoshop now, especially creating custom brushes and managing layers.
The art's not great but I've never painted a thing in my life so I'm proud of it anyway. Next time I do an exercise like this on my own it will be better as I'll be able to work my way rather than follow along someone else.
ok, one last photoshop painting exercise to help solidify what i've learned. This one I did on my own, and used similar workflow to what I do in substance painter. Basically i exclusively use fill layers with mask and just paint on the mask. this way i can tweak colors till my hearts content -- very little commitment and lots of flexibility. I also did most of the work with some homemade brushes, which I actually find to be way easier than trying to figure out somebody elses brushes. They are easy to make, so I don't see a reason to purchase pre-made brushes (this is what I am saying after I bought a bunch.)
So, here's my dog-ter, Poppy Jane. Possibly the most prettiest dog that ever was a dog. This makes my third painting I've ever done, and while I like the way the dog looks I didn't really have a clue what to do with the background. So it doesn't tie together very well. A more painterly style similar to the dog would probably be better. Maybe I'll play with it more in the future, but I got to get back to 3d now.
Having some fun this month making a game with a friend. Good to take a break from portfolio and do something practical.
We are more or less cloning an old turn based strategy game called Fantasy Wars. To get the project finished in a month we are turning it into a very simple board game. My work is to finish 31 characters -- just static little figurines -- and also make the game board.
My plan is to make 2-3 figurines a day, and the rest of the time will go into doing as much as I can with game board to make it look nice. Also play testing to find bugs, of course.
First week has been planning and iterating both on design of the figurines, game board, and also figuring out the most efficient workflow. There was some time I began to doubt I could really finish these characters to the standard I want in time, but after getting a few kinks out now I can output at the pace I planned for. Basically I have assembled a large kitbashing scene in maya, largely composed of stuff I created for previous projects -- get it all UV'd and then slap shit together, do a quick sculpt pass in zbrush, and then I've got some smart materials made in Substance Painter.
Making tons of little figurines like this is sort of like sketching... it aint about polish and looking pretty, it's just a chance to try out lots of stuff, learn a lot real fast, nd have some fun too.
Instead of your typical goblins and orc's, I've turned the enemies into a Wendol-like race, with enslaved bear-werewolves as the beast race. I know it's not the job of a 3d artist to design, typically, but I find it a lot of fun and doing quick static sculpts in pose is a great way to experiment with designs. Try to find ways to communicate the units role, and look cool too.
Before I make each character I first go over how it will behave game mechanics wise, then I write a quick lore blurb, then I find some quick reference online to fit my lore and get to work. Each one takes just a few hours. I try to tell most about the character with pose. Peasant on the right is a weak kneed turd. Archer is mission focused, dangerous dude. Militia is ready to stand his ground, but he ain't leading no charge. Bearman is a bearman, so watch out. Wild tribal dude is crazy, but he's just got a stupid club. Be careful anyway.
Getting out 2-3 a day is very challenging. Most time goes into organizing UV's and then setting up the pose -- it is time consuming with characters that have a lot of separate pieces. I use a combination of maya quick rig, then manually moving pieces which should not deform. I still haven't found a single process which works for all characters, so production is slower than it could be. I am not sure there will be a single solution which works for all characters. But with each character completed, that gives me finished assets which I can pull from to quickly build new ones.
Big thoughts on design is keeping unit behavior and class readable from high camera distance. I think having unique colors associated with unit class helps, though for enemies I am keeping a dark theme. One thing I liked a lot about the Wendol in 13th Warrior movie is how they always remained mysterious -- even up until the end of the movie you barely got a clear look at them. They were always hidden in shadow. This helped get you in the same mindset the vikings must have been in -- they believed the Wendol to be demons/monsters, not mere men. So I want to capture that sense as best I can with our Wendol inspired enemies.
Not that these quick sketch characters really deserve beauty renders, but in the right light -- or lack of light -- i think they look pretty cool
Enemy archer I based on warriors from New Guinea. These guys are really fucking impressive looking. I want to make a high quality, realistic model of one of them some day.
Last of the bottom tier characters. Still haven't settled on cohesive material scheme yet. I kind of am partial to a monochrome approach because that shows off the details of the models best, however from a gameplay perspective it does not deliver as good of readability, and there is a lot of information we can convey with colors. So I dunno, I just try something different with each one and eventually will find something I like, then go back and update everything.
Just a basic game board. We probably won't have time but I made a pull-out drawer which could house the pieces. Then at beginning of game could have it open, pieces come out and go into place. Simple animation but as neither of us are animators, it would likely take some time to do
So once I got everything kind of assembled in Unity, I realized we didn't have a strong cohesive theme. Rather than just roll forward anyway, I decided to rework the theme. The environment wiil be fully abstract -- just a literal 3d game board with all pieces being constructed as if they were real life figurines -- and the enemies will come from early christian mythology. Headless men, bat-wing demons, dog-head men and multi-headed abominations... cool shit.
Two reasons for this -- one a better, more cohesive theme thats easily identifiable, and two, the monsters will mostly be single water tight meshes which makes my work a lot easier. So even though I got to redo about five characters, it should be much easier overall.
For enviro props such as trees and mountains, I've tried to mimic styles I've foudn in medieval paintings. I am not so sure the mountains read as such, but I don't really want to insert the HImalayas in here. Perhaps a cap of snow could make them more apparent?
New enemies. These are taken from medieval christian mythology. Really kind of goofy and sometimes hilarious in design, but I kind of find that charming in a way.
Key benefit is that these are really fast and easy to make. I can just copy directly from old paintings. Designing as you go is just too time consuming.
So far we've got a headless man, a demon stabber, and a basic imp. Some the higher tier enemies coming are just gonna be fucking hilarious to make.
We are at halfway point for the production phase. Tomorrow should have fully playable minimum viable product. I'm a little behind schedule with characters because of the theme change but that's ok because I started a week early. Always good to do as much as you can when you can so that you've got flexibility down the road.
There was some technical troubles with scale and pivots. It's really important in this case as the movement calculations and some fragmentation shaders we are using depend on it. Lesson learned is, get that shit figured out early before you have many pieces. Lost half a day having to rescale and carefully go through each piece.
Besides that, I fucking love these new characters. Lots of fun. I am trying to keep them looking like they are hand painted figurines still, even though they are getting custom paint jobs now. To meet that end, I just paint them by hand. Haha. Occasionally some texture or photo with a warp and blur filter makes a nice result though. I've made some nice brushes in Painter and am getting practice blending colors, and do kill some time just playing around with colors to see what looks cool or not.
I try to get two done in a day. This normally means about 10 hours of work. I've got the workflow pretty well dialed in. It is actually quicker not to kitbash so much as that makes UV organizing a big chore. For the most part I start by modifying a base mesh -- not usually connecting additional parts but just intersecting the geometry -- and then as the topology is very simple unwrapping after the sculpt only takes 20-30 minutes.
One thing I don't think I'll get a good result from is the lighting in Unity. I just don't know a lot about it and also I lack a vision for what final result should look like. I'd like it to be crisp and clear looking -- I hate overly dark shadows and high contrast in games -- but I'm not able to find a good solution just poking around, and I've actually determined not to poke at it as its basically a time sink. Hopefully time allows me to do some studying, but most like I just do my best with it and that's that.
Getting some of the second tier characters in. Spearman, Halberdier, Swordsman, Bloodsucker. Got some more demon characters and then onto calvary. Feel like I'm getting a little better with the hand painting. Definitely enjoy it a lot.
Also did something like a tilt shift camera. Just light blur on the top and bottom and 25 degree FOV. Not sure if we will go with this but I wanted to test it out. We have a fog of war effect that kind of makes the blur unseeable most the time, so maybe no point in that. I do feel like the narrow FOV helps reinforce the board game aesthetic.
Bloodsucker, Faceless Archer, Beastman, and a ballista.
Few things I learned recently : You got to be careful about too much noise. Makes characters hard to read from game camera distance. Especially a noisy normal map. Got to make sure you are doing a lot of your texturing from a distance, don't waste time getting in close for little details. Big contrast changes to block out the body shapes and also gradients seem to work well with these types of characters. And every one gets a curvature layer to make the edges pop a bit more.
After a bit of play testing, I feel like each unit has a distinct personality, so that's good. The Imps tend to pop out from the woods and feel really impish when they do so. The archers are a pain in the ass and I feel they look the part. Always look like they're ready to get you. I think the action oriented poses helps with that. In a game without animations, I think it would be more boring if the units were in resting poses.
We are just about to MVP status and should be ready to go into polish stage next week. A bit behind so we will be chopping the content pre-ordained.
I tried to not let the Crusader look like a klansmen, but it's impossible to avoid entirely as that's where those pricks pull their reference from, but for the record we ain't cool with the klan.
Got the last of the essential characters done. Crossbowmen and Yeomen Archer. Also updated some textures on old characters. This leaves room in the schedule for four to five guys more we have deemed important but non-essential. There is some inconsistency between the characters in terms of time period and geography, and some subtle stuff like general level of saturation and style in which I painted them, but I don't think its too awful. Just a symptom of not having clear design from the onset and goin fast.
After all that, polish will include a little bit more texture updates, some small enviro props and texture updates, some voice acting (gonna be fun), and lots of writing. Gonna be a busy two weeks.
Made a title screen scene as well. Needs some more lighting work and better fx but it gets the basic idea across.
Most UI is still placeholder and there's a good bit of polish left, but we've got the first three missions done with army progression working, no real bugs, and enough content right now for about five hours of playtime. We should be able to double that playtime duration by launch time.
I've got the tutorial mission built with dialogue and triggers and all that. Massive amount of fun putting it all together. I spent the afternoon trying to learn to roll R's so I can do proper accents for the voice acting (this is why making your own games is the best.)
UI is coming together, but is still WIP. Most of the particles like the god rays and ground fog are placeholder as well. I think a few more small props like windmills, rocky hills, and maybe on more type of tree will be enough to complete the environment. We will replace the brown hexes (which are meant to be roads) with a spline solution that creates a road mesh. Will look a little nicer.
Most of the work from here though is more game design/level layout related and not art related, so I'll just post some gameplay videos once it's all finished in two weeks.
The first three maps introduce player to a series of increasingly complex encounters. First, cross a river. Next, maneuver marshlands. Players goal is to first amass an army by collecting allies that have been scattered by the Demons Army attack. Later, player must out maneuver the bulk of the Demon Army, reach the last stronghold of Humans left before the Demons do, and defend it. Whoever player keeps alive carries over into next mission. Whoever dies, is lost forever. Player must be conservative and play the long game. Yet there are times when sacrifices must be made.
The fourth mission is where player first learns this lesson. With ally forces trapped in a ravine, will player commit all forces to the rescue, or split, head north, and call upon Lord Raumbolts Castle for reinforcements? The simulation does not stop -- there is no scripted storyline. Only hard decisions and real consequences.
Trying to find a stylized style that I can do relatively quickly and that I like. Even when I was a kid I still didn't like very toony stuff so I find that hard to do. It may be a bit more popular but I think this is a decent balance between realism and simplicity? I post up more iterations as I work on them, but first I gotta rig this and make a few simple animations so I can run around in unity.
Got the pupper rigged! I used Advanced Skeleton. Took me two days to sort of figure it out, but given the power of the toolset it's well worth the time. I dunno why, but I get stupid happy making poses after getting the model rigged.
At some point I gotta do face rig and blendshapes and all, but for now this is enough to build an idle, walk, and run animation so I can get things started in-engine and "feel" my game.
The is Sipsey when he see's a squirrel in a tree.
Oh Sipsey, you shameless beggar!
Oh, and thanks to @nik@NikhilR who sent me this awesome Bison sculpt which I retopo'd and textured this morning. Very cool!
My first animation! Not great but I will be making a lot more so plenty of practice to improve from.
I've
dropped Sipsey into Mikail Gustaffsons low poly enviro package in
unity, which for initial prototype I will simply alter the textures on
before going into production with my art. Reason is because I think I
will do something pretty similar to his style anyway, so reusing his
work can save me a bit of time while trying to discover my style.
Maybe not the purest artistic thing to do, but I always start new work but first seeing if somebody else has done something similar, that way I get a good head start.
Has taken lot of time to figure out my workflow for world building. In the end I have settled for using world machien to make a big terrain, then i split it into tiles inside unity, later i will optimimze the border terrains into low-poly meshes. I use a simple spline generator to make river meshes inside unity. I use vegetation studio to instantiate vegetation and draw billboards. Very fast application and has nice splatmap generator as well.
I've also implemented ramp lighting shaders, as well as some post effects like fog and outline. I headed out for job in a few days and be gone for too long, but before that i try to get a bit of my vegetation filled in with some basic logic. When I come back I do more work on character controller and animations.
Learned how to make "fluffy" trees by editing vertex normals. Thanks to the wiki for that. I got inspiration from an article on 80lvl about soon to come unity asset store package called "the illustrated environment". I was really impressed with it.
I'm just getting a handful of veg assets ready for about three distinct biomes that will exist within this Yukon River scene. Up high we will have birch forest, in lower water collection areas we will have thick underbrush of ferns and alder, and down lower on the flats we will have black spruce muskeg. Of course throughout all these areas we have the iconic white spruce.
Each area isn't there just to look authentic, but to provide distinct gameplay as well. Some plants will create obstacles to dodge or jump over while chasing or being chased, so there needs to be a clear logic to this, while at the same time not looking too gamey.
I wonder if the palette is too much right now, but that's easy enough to tweak later. Something more washy or canvasy might have an appeal, but it makes me feel a bit pretentious too. I'm not a big fan of the indie darlings that take themselves very seriously. I kind of like simple, bold, direct art that doesn't seem like it's throwing every punch in every frame.
If the target audience is younger crowd, maybe it should sort of look like if they made it themselves. Or maybe it should all be a bit sloppy and blobby, as if direct metaphor to the dogs general goofiness. Dunno, just things to experiment and figure out.
More work trying to find art style I like for doggo game. I pretty happy with this. Lots more work to be done as this is mostly just a few basic procedural rulesets to spawn from, but I think I can build from this.
Was supposed to work on some more animations but had too much side business. So I did a bit of tweaking and trying to clean up the look of my current biomes. It seems like using less species of plants to keep a clean color paletteper area is nice. Still lots of stuff to add but I keep this idea of simplifying things down a bit in mind. I have tendency to want to put every conceivable thing in each area.
I like silhouette of the birch trees on hill. Has a fun shape. But hte colors are too contrasted I think. I should try to have a smooth blend like the spruce trees.
This is way too busy. The birch forest should be a relaxing place, so this is an area that needs lots more work.
When dog goes he goes real fast.
In a draw. A trickling creek will run through here. Tall ferns make it a place you could accidentally bump into a bear.
thanks @Ashervisalis . I'm really having a lot of fun experimenting with it. i like how fast and easy it is to work like this with simple models and even simpler textures. I think its a good exercise in breaking art down to the fundamentals. Like, 90% of what i focus on here is just color and composition, which is really nice. I don't gotta think about normal maps or anything complicated at all. Lots of fun.
This will be last update for a hot minute. New job has me back up in alaska with the same mutt you been seeing here. so maybe we find some inspiration for our game.
Two new biomes. Creek and muskeg. Making a river is surprisingly hard.
Been a minute since I done any modeling. I am doing a smaller game project as something like training before diving into the larger one. This will be a military land navigation training sim. Although it seems pretty different from Dog Go Run, the world creation and game balance will depend on similar variables.
To get back in the groove of things I am making a lensatic compass. It will be fully functioning so I keep the moving parts all separated.
Replies
This isn't great artwork, but I had a lot of fun doing it. So I win, okay? I really like the simplicity of 2d, and how flexible and forgiving working in PS is, especially for somebody who can't draw a straight line.
ok, one last photoshop painting exercise to help solidify what i've learned. This one I did on my own, and used similar workflow to what I do in substance painter. Basically i exclusively use fill layers with mask and just paint on the mask. this way i can tweak colors till my hearts content -- very little commitment and lots of flexibility. I also did most of the work with some homemade brushes, which I actually find to be way easier than trying to figure out somebody elses brushes. They are easy to make, so I don't see a reason to purchase pre-made brushes (this is what I am saying after I bought a bunch.)
So, here's my dog-ter, Poppy Jane. Possibly the most prettiest dog that ever was a dog. This makes my third painting I've ever done, and while I like the way the dog looks I didn't really have a clue what to do with the background. So it doesn't tie together very well. A more painterly style similar to the dog would probably be better. Maybe I'll play with it more in the future, but I got to get back to 3d now.
At some point I gotta do face rig and blendshapes and all, but for now this is enough to build an idle, walk, and run animation so I can get things started in-engine and "feel" my game.
To get back in the groove of things I am making a lensatic compass. It will be fully functioning so I keep the moving parts all separated.