Motivation
I'm currently playing around with a game (engine). I like that it has so many different things to play around with. Rendering, physics, ..., art.
Since this is my first post I kinda feel obligated to write a short intro. I intend to keep my posts somewhat concise, I'm not an experienced writer. But my current idea is to show my thought processes and my "struggles". Therefore the posts might get more verbose than I want them to be. Also, they may not contain a solution to the problem I'm currently tackling - at least not in the first version.
So much for the general intro. Now what is this first post about? I'm currently playing around with a game or engine. I'm not sure if a final product/game is actually on the horizon, I just enjoy writing some code, thinking about interesting challenges and do the odd piece of "art"work. Variation is the name of the game.
I recently stumbled upon some art theory stuff, Hue Value Saturation and character design, which made me wonder how my game "art" stacks up. "Maybe", I can figure out how to improve in this department.
So, let's see how my developer art fares.
Silhouette
The silhouette of a character should be recognizable. That means if you remove all detail from your character and just paint it in one solid color, you should still be able to recognise the character. So how does (never thought about naming the player character, until now that is)'s silhouette hold up?
It's you! Whatever you're supposed to be. I'm not sure what to do with this. It looks like a somewhat distinct stick. This might be good enough, considering how few pixels the character is made up from.
But his legs feel a bit long in the silhouette version. Maybe some booty fixes this "issue"?
It looks 🤔 better? How the heck does his backpack stay where it is, it has no straps. But im diverging...
UPDATE, paint the silhouette with colors to make it something completely different. Twitter @strugglingdv, hashtag #jergen
Value
But, let's add back the color. I've always struggled with color. If I, in a rare case, drew something with pencil I actually liked, I never dared to add color in because I usually f***** it up. With colors it's important that they form a harmonic unit and go well together. Another thing that's important is where our eyes are drawn to, e.g. what pops out. There are different color systems like RGB, CYMK, HSV, ... HSV seems to be the one the digital art guys prefer
1.
- Hue is what us layman would call the color itself, so yellow, blue, green and so on.
- Saturation is the intensity of that color (hue)
- Value defines how light or dark the color is.
The following image shows how a change in saturation or value changes a hue/color. A lower saturation makes a color lighter, while a lower value makes a color appear darker.
Every hue has a different inherent darkness. Comparing darkness levels across colors is hard because the hue throws us off. But, we can kind of normalize our color and remove the hue from the equation. This way we can visualize what parts of an image pop out.
Visualizing the Values
In GIMP: add a new layer and choose "HSV Saturation" as its mode. Then fill it with black and make sure the layer is the top most.
- the dark mustard of the hoodie and the dark blue shadow color of the pants seem to have the same value.
- the light mustard color, and the lighter pants blue are also of about the same value.
What to do with this information? The different parts of the character itself seem distinct enough. When looking at the value version there are four to five main parts: pants & hoodie, shoes, hair, backpack and maybe the face. The hair, backpack and shoes really jump out and draw focus.
- The pants and the hoodie are distinct enough when colored.
- Whether the backpack is too prominent, or not might depend on its utility in the game. It doesn't seem too overpowering in the colored version though.
- Im a little inclined to do something about the face. For one, the contrast is probably too low. Faces are important to humans - so I have heard. For another it seems to have a green hue, which I haven't noticed until now.
Summary
Although I still have mostly no clue what I'm doing, the little theory helped. It made me take a closer look at my character and although the actions I took weren't always a direct response to what I wanted to improve (looking at you silhouette booty), overall it helped make my "art" better.
Thanks for reading, and keep on struggling.
References
- Steam: Dota 2 Workshop - Character Art Guide - Silhouette
- Game Developer: How to Reduce Visual Confusion in Your Game
Addendum
Meet Jergen
I've decided to name the player character Jergen - at least for now. The pronunciation is as follows:
- The J is pronounced like the Y in year
- the e is pronounced like the E in v*e*ry
- the r is pronounced with a rolled R
- the g is pronounced like the G in great
- the e is again pronounced like the E in v*e*ry
- the n is pronounced as usual
-
Absolutely no source for this 😉 ↩
Top comments (6)
Nice article, these basics really do matter and you've reminded me of that. I'm in the same boat. A full-stack developer from the Flash years get's cancer and decides to start community driven gaming company, picks up Unreal and Godot and that's how I'm spending my recovery :). I look forward to more of your journey, thank you for sharing. My exploits are VR as the end/ongoing goal, but the courses I'm taking have me delving into 2D, retro platformers and such. I'm also a musician and I've spent far too much time on the music and sound effects and I look forward to your articles on that aspect of game design.
Hi Paul.
Thank you very much for your kind words, you made my day 🙏. I hope I can keep this going and someday reach the point where I have to start with the dreaded topic of sound😉.
Quite a journey you're on. I'm looking forward to reading about your experiences along the path from 2D to VR, and the community driven gaming company. That one sounds really interesting, maybe you could give me a sneak peak on what this is all about?
Wish you an awesome day, all the strength in the world for your recovery and all the best for all your endeavors.
Hey, great to hear from ya. Do you have a VR headset?. Both Godot and Unreal are very XR friendly. I took a long look at the market and my own gaming history and realized 3 things, 1. I'm not going to make AAA titles, 2. Everyone is scrambling for the same carrot 3. If you go back to the atari (via launchbox and some dedicated archivists out there) you will find a foundation for every gaming style we play today... it's all been done, with VR being the only 'frontier' left. Even in that realm, any old 3D game can have a VR camera plopped into it and suddenly be re-released as a VR title. So if I'm going to start any kind of gaming company, what makes sense? I needed a completely new paradigm. Then it hit me, what about community oriented games? interactive games/experiences that are composed of the city I live in (Sarnia, Ontario). If I'm not seeking AAA hits then I'm basically an indie developer, there are 78,000 people in Sarnia. If I were to create games that featured Sarnia, sarnia locations, landmarks, business etc. would the people of Sarnia buy them? It sounds a little crazy, but I think if I take all the game styles I love, plus the interactivity and immersion of VR and start developing entertainment/educational products that focus on the waterfront, border town of Sarnia then I think there's a perfectly sustainable, indie game, market there, with room for employees. Then there's VR's many other uses for education, city planning, tourism and more. Local business could advertise passively within the Sarnia-focused gaming environments creating all kinds of micro profit models. The best part of this is that the next city over and cities all over the world can do the same thing and no one is chasing the same carrot. With the low overhead the games don't have to cost a lot either so there is an accessible price point. So I got pretty excited about the concept and started work on my first prototype called "Climb the Kenwick". The Kenwick building is a very unique and iconic building here in Sarnia, sticks out like a sore thumb. There's decades of stories and lore surrounding it and I thought it would be an excellent candidate for a climbing simulation. I went with Godot because it's lightweight, runs on anything and my dev computer isn't the fastest horse on the track. I have a working demo (complete with sound lol), it's a little simplistic, but a wonderful workout.. and tickles the thrill-seeker in me so I know I'm on to something. I hit a point where I had concepts in mind, but didn't have enough Godot under my belt to flesh out programmatically so, with the incredible help of HumbleBundle ( GO THERE) I was able to snag courses and low poly models to fill in the holes in my modelling and programming. So that's where things are at right now. I do my courses for part of the day, chip away at the "Kenwick Climb" when eureka strikes and create music.
Hi, hope you had a great weekend. Your reasoning sounds really sensible, not falling into the trap of wanting to make the next AAA game. Very grounded, but with an expansion plan already laid out - I like it. And, as you wrote there are so many options like an AR city guide, outdoor museum, ... I can already imagine myself learning about the Blue Water Bridge and then playing some Space Invaders while taking cover under it ;) No, no VR headset on my side - my computer isn't even a horse anymore ;)
I've thought about using Godot, Unity and the likes but I've always come back to rolling my own engine. Not the most sensible decision when it comes to getting something out of the door, but it's more about the journey than the destination for me - at least for now.
Your own engine - that's slick man. I think if my path had taken me deeper into C type languages I would have pondered doing the same, but my experience was largely web related and more recently Unreal and Godot. Once I got into Godot and ended up learning more about C# I realized Godot might be the type of engine I myself would have built. It's logically structured in a way I think my own brain might have approached the task. I don't feel like I'm blindly standing on the shoulders of mystical functionality (Like you do in Unreal lol).. I feel like theres C# then stuff that makes good sense to package up into a function and Godot is simply a big bag of that, good bundles that you are free to unbundle and rebundle if you see fit (and I have in some cases). I was thinking about our chats while working on a save game feature for my little top down driving game, in my head I was saying 'I had no idea the save game story was so thick with scenarios' - so let me say this to you, if 'saving game' is a feature of your game, plan and integrate early!! Luckily I had created custom structs for many of the activities like lap records, car upgrades etc. so saving them wasn't an issue, but stuff like 'no save games', 'new game', 'saving over', 'deleting save games' - this all turned into a lot more paperwork than I had anticipated.. to the point I was reminding myself that I've never done most of this stuff so 'don't beat yourself up too much'. I ended up taking a few days away from "Dot.Dot.Driver" (tentative title, gf thinks it's cute and likes it) and building a clone of another wee game I discovered (and developed a minor addiction to) called "Bullet Bunny" (which I think is a clone of some other bullet-hell thingy.. there's so many), it was built in GDevelop. There was something so basic, so elemental to the game play mechanics that I ended up writing a little 'paper' on it, a breakdown of it so I could build my own (I couldn't validate playing it one more second without doing a personal 'Jam' to build me own). I did build my own and now, when the urge strikes, I play/code my own version of it. Your own engine... that's really cool man, my mind is ablaze. What language, 2d? 3D? are there engines you like already? I'm extremely interested in that journey and hope you'll share that story with us, sign me up as a beta tester when it's that time. If you're interested in playing any of my little games just let me know and I'll pass you an ftp to my domain. I'm planning some mini launches on itch.io in the next 2 months or so, just to get my foot in the water and kick-start the portfolio. My primary machine is a Ryzen 7 1800x, 16gigs ram, AB350 motherboard with an Nvidia 960 6gigs - nothing crazy, but it runs the Godot 4 VR stuff just fine. Oof, sorry bout the book.. I don't really socialize since the surgeries so I'm a little backed up with conversation lol. -- PLAN SAVE GAMES EARLY :) ttyl.
Hey, no worries. Godot sounds interesting the way you describe it, as a bundle of tools. I've always liked the Unix philosophy, a lot of small tools that do exactly one thing but do it well. Thanks for the tip, haven't thought about saving yet ;). Let me return the favor: if you want to make a game, (almost) never write your own engine. The engine is built with C++, OpenGL and GLFW, and I'm currently focusing on 2D. Thanks for the offer, I hope the day comes when I need beta testers ;) Nice machine, I'm on a - I think (can't currently look it up) - second or third gen Core i7 Mac with integrated graphics. So, if your games run on this potato I'm happy to try them out. Btw would love to read your paper on the game play mechanics, maybe turned into your first blog post? And a second one about what you've learned from your save game journey.
Take care