Rambling of a restless mind. That propably gets it. I write about whatever comes to my mind. Granted, that's a lot of more or less useless stuff. So I'll probably spare you of my thoughts of lesser importance or those that are borderline mad. The rest would be science, philosopy, computer graphics and mostly game development.
This article was originally published over on Gamasutra by myself. Go take a look :D
Why should I read on?
When I began freelancing some time ago, I spent a considerable amount of time looking for and making up my own prices for my art services. I found it very hard though, to find any information about pricing. Some time later and some experience richer, I want to share a few of my insights to make it easier for both, artists and clients alike, to get an idea, of what price is to to be expected. This article is not meant to be a means to all ends. I just want to put it out to help everyone somewhat adjust their expectations before going to look for talent or planning their budgets.
Also, this is not a guide on how long a certain piece of pixel art must take, or how much exactly it needs to cost. There are many things to consider here. For one, you can get very cheap prices for pixelart from some offshore countries. Sometimes, when considering individual artists, this might work out. Working with contracting companies from offshore countries comes with its own slew of problems and opportunities. I don't want to write about that here. This topic can easily fill an article of it's own. The other thing is efficiency. I am just going forward from my own work experience and what I witnessed both on me and fellow artists. Each and every artist has it's own pace of work, its own strengths and weaknesses. In the end you are not paying only for the execution of placing pixels next to each other, you are hiring a person and that person's creative prowess, technical expertise and accumulated work experience. So keep in mind, artists charging differently from each other is OK. Find the artist for the job, whose skills you value most, who can contribute in the most positive way to your project. Having that person on board is worth much more than a few dollars per frame.
The Tool for the Job
So let's start with the obvious. Pixelart is said to be less expensive than 3D. That's true, most of the time. But depending on the individual setup, the costs for pixelart can become prohibitively high. Not only in creation but also in asset management. If you have a game that features, multiple characters with different customizable armor parts, multiple weapons, funny hats. All of that will be added upon in later updates. This will demand a very specific and technical setup to stay viable cost wise and management wise. So you need to have an artist that has a very strong technical skill for solving that task more efficiently. And obviously, this is something, that needs to get paid. The more complex the task, the more you'll have to pay for technical expertise in addition for pure execution. Pricing something like this can only be done for the task at hand and isn't really the focus of this article. Just keep in mind, that asking for a quote on "pixelart" depends on what actually needs to be done and designed and if the workflow is already defined. That brings us directly to a common misconception regarding pixelart...
Everyone can do pixel art ...
I am personally an advocate of everybody can LEARN to be good at, well, almost anything. However, I cannot support the notion to believe that something is easy because it LOOKS easy. Pixelart, like any other craft, needs to be learned. Learning art is a passion. One that you don't just master in a weekend course. It involves acquiring a lot of classic art skills as well as a lot of practice and studies. Usually in your spare time. And because we are in the world of computer graphics, a lot of technical skills jump in to complete the equation. When someone talks about how art is a passion, they mean that dedicating your free time to get better at your job, on the expense of real life, needs passion to endure and drive you forward. What they don't mean, is that art doesn't need to cost much. Because the artist is content to have a job that's its passion. That's nonsense. As nice as passion is, you can't eat it.
A word about the job
Sometimes a offer for pixelart will go something like: 100 icons, 64x64 for 120$. Not joking. Real life. Seriously. I was reading this and said to myself... "No one would ever take this". Then I read the comments and my eyes widened in disbelief. Some people were willing to take it. That's crazy! Why? Let's do some math. How long does one icon take? Starting off, do we have concept sketches on any icon? No? So, assuming that you have an idea instantly and bring down a few sketches, I'd say that 10 minutes are needed at least, just to explore some designs. Finished on the sketches and further assuming that you are doing your own sign off (which is basically never the case) I'd say that a finished high quality icon might take about 20 minutes in average (64px square) depending on the contents, which palette is to be used and such. Do I have to get a signoff on the final icon? Let's assume yes. Another 5 minutes for minor changes. And you can end up so much worse than 5 minutes. So, maybe 35 minutes in total if everything is in my hands and I am super inspired and at maximum efficiency. So that's 1,71 icons per hour. Eight hours a day? Definitely not. But let's just assume that we can hold up the crunch. So that's 58,47 hours of work. So that's $2,05 per hour. Seriously. You are better off begging for money.
And a word about the price
So what is a realistic price for a single icon like this? Here's a number:
Something along $38 net.
Working off a concept sketch with feedback loops included $27 net. Again, this is just an average. Depending on the chosen style, things can get cheaper or more expensive.
Summing up, the above equation you'd make about $65 an hour. "Wow, now this is a bit much", you might say.
Give me a second here. Craftsmen charge up to $80 net an hour. Electricians sometimes even more. They do a good job. But one that often requires significantly less dedication to be get good at. Taking your earnings home, taxes on income will shave off a bit. Up to a third in some countries. You need to pay for lodging, insurance, food, new hardware and software every now and then and save up to bridge some idle time where you are not employed. That's regular freelancing business math. In turn, clients don't have to hire someone and pay them even if they are sick or you don't have anything to do right now. That's the tradeoff. If you are sure that you can employ an artist for a certain time full time, stop reading, go hire someone.
Idle time, happy time
I said idle time? Most freelancers aren't in the luxurious position to be booked 5 days a week for 8 hours all year long. One project ends, another one starts, maybe that takes a week. Or a month. Or two. Maybe there are just no jobs around at the moment. And the market is full of competitors. Every self employed businessperson experiences idle time as a part of their work reality. Actually, if an artist gives the customers estimations, talks about the project, maybe does an art test... the artist usually doesn't get paid. But it's part of the work equation. It costs time. All these things wittle down on your $65 an hour to something much less glorious. Maybe half, maybe even a third when starting off as a freelancer. So here's a shout out to all artists. Check your ledgers and check your work time. Then you'll have a pretty clear picture of what you need to charge to arrive at somewhat viable earnings. $25 the hour, $1 per frame, all this might work while living in your mom's basement, but it won't work out later in your life. Especially not with a family. So maybe reconsider and understand that you cannot retain your clients by saying "Oh, sorry, I have a flat now, so I need to charge $7 additionally per frame in the future". Start out at a realistic price. This price may of course vary in your regional economy, but whatever it is, start from there.
Scale and pixels and the good ol' time
Good, I wrote about how that icon is priced, but what about a general rule of thumb. How can a client save on budget? Just scale it down? Or if the budget is there, double resolution is double the price? Costs in pixelart actually rack up pretty fast. Think twice, going from 8x8 to 16x16 is actually four times the resolution. Contrary to regular digital painting(which has other challenges), pixelart often forces the artist to go in and touch the individual pixels. Or even start by placing pixels one by one from the start. Where a 8x8 tile can be completed pretty fast, a 64x64 tile takes some time to make, a 256x256 pixel art takes a considerable amount of time. A fullscreen HD pixel art piece is taking forever and a day. I have done a few larger pixel art pieces and seriously underestimated the effort in the beginning. Scars make you wiser. Maybe. So choosing a lower resolution will make everything come cheap? Cheaper, yes. But certain tasks like concepting won't go away. You'll have to think about how a 8x8 orc sprite should look like and animate. Even if it's just that handful of pixels. For even the lowest resolutions, as hard as I try, going under 5 minutes per frame almost never works for me. Unless on very, VERY mundane tasks. So I'd really recommend to set 5 minutes as your very lowest unit for pixelart execution, reserved for especially easy tasks. Meaning that something like $5,50 net per frame is your lowest rate for execution.
I like to move it
Up to now I wrote about frames, price oper frame. But what about animation? Is there any difference between let's say, a static environment prop and a game character? Of course there is. Animation makes a huge difference in effort. As a rule of thumb you could say: The more complex the motion and the more complex the shading, the higher the effort. This holds especially true for effect animations like smoke, fire or lightning. But this also means, that giving a rate on something like "20 Spell FX" is nigh impossible to give without knowing more. If a client refuses to give more information or show some concepts, negotiating for an hourly rate would probably be an artist’s best bet here.
The basic premise for animation is like this: You have a large number of shapes and/or particles that are all transforming and moving. I usually go ahead and animate flat color shapes first before putting down the shading. During shading I often notice some problems in the animation flow and need to go back and forth to adjust to the changes. So I usually put down shading in steps until I arrive at the shading complexity that is needed. I usually try to stay as low on shading as possible for FX animations. I also try to let the client in after the first rough animation draft. Depending on the client I sometimes produce one fully shaded hero frame to drive the concept home. It puts everybody on the same page and gives the client the ability to provide early feedback before time and money gets wasted.
Character animations are often times simpler than fx in that the individual motion paths are shorter and affect a much smaller part of the body. But the more different kinds of movement a character needs to support, the more difficult it gets. Fighting games for example have a very diverse moveset for each character that sometimes bleeds into FX (smoke, dust, fire, Kame Hame Ha, whatever).
So how does all this affect the price? Depending on what is actually needed (complexity, shading) I will usually add up to 30% to the bill for animation. Again, the impact is lower, the lower the target resolution. As for 2D rig animations, this is an entirely new beast that needs to be covered separately. The pixel art for the individual character parts is pretty simple, however. It doesn't deviate much from the regular pricing per frame.
Here, take my +1 Mace
So, where to go from here. Please remember that this is not the definitive guide for pricing. Prices haven't been extrapolated from my experience in EU and US markets. They make sure that the artist can dedicate enough time for a project to deliver the best. Prices may vary, depending on your local economy. But while offshoring may be an option, it comes with it's own problems and... well, prices are already catching up. Accepting $1 per frame as an offer/payment means either starving the artist or or hiring someone with subpar skills that may eventually bail out. You might get the occasional exception, where someone is doing absolutely stunning work and charges next to nothing. Sometimes those artists desperately want to break into the industry and charge deceptively low. From an artist's perspective it pays off tenfold to do your personal stuff instead of accepting underpaid commissions. From the client's perspective you are burning through exceptionally valuable human resources. If your budget cannot support the art that you are aiming at, get back to the drawing board. And for everyone who's just trying to get their feet wet, find some student or hobby projects where everyone on the team works for free and the product gets released for free. But it's no good getting scars while having nothing to eat.
Over and out
I wrote this article because of countless job offers that read "For Free" or "Pay later when..." or "Pay nothing but it's awesome for your portfolio". I really think that the indie scene has a lot of power, but I see so much potential go to waste by burning out artists by offers that feel like a scam. So I thought I'd push out some numbers for everyone to read to help artists and clients talk to each other on the same level of understanding. On both sides. I'd love to be hearing your thoughts on this article and would love to be hearing about your experiences. Your game doesn't have a large art budget? Shoot me a mail with a description and some screens and I'll give you my thoughts. To my fellow artists... resist the urge to accept the low offer. There will be a better one. If you need some advice on freelancing, feel free to nudge me any time.
The last update on development has been some time ago, hasn't it? Bad news first: I had some commissioned artworks to do. Yeah well, "bad" news. Not on a strictly personal level. More on a "Crap I couldn't write devblog" level. Good news: I just hadn't time to write a devblog, but I had the time to push PMap along. So, relax.
Where should I start, a lot has happened...
Performance optimization is a big thing for procedural content creation. You don't want to wait like forever for the level to load. So, since the last blog entry I did some groundwork for getting the framework do it's thing faster. This week, I took a VERY close look on the generation modules, the walkers. A walker is basically a Sampler/Placer for grid values with a specific algorithm. Each walker does something else. Some Walkers have a very low performance footprint. They complete in under 4 mas even with map sizes of a 1000 square. Which means one million cells. One bloody million. Needless to state, that some other walkers are WAY more expensive when it comes to them doing their thing. Like the Mazer. This walker creates mazes. Ranging from perfect mazes to variations with loops and such. And even with small maps, this comes with a price. its like 300ms average for 50 square grids. But that's only after my first optimization pass.
HI everybody. I have just recently finished a artwork that has been on my art backlog for quite some time. This time I just added layer after layer while editing this thing in Photoshop. I usually don't do that. Don't really know why I did it this time, but this gave me the opportunity to create a small image showing How I went about creating this piece. Enjoy!
There's that point in software development, where the features are totally great, but the execution is still a "bit" shoddy. And colossally slow. And you try to crank the map size up to a thousand units square and you are like "Yeah, that'll be so awesome" and Unity is totally like... dead. Just why? Hwhyyyyyyy!?!
Being an artist I didn't ever want to need to think about performance optimization. And memory management. I didn't want to know what the heap is. And what the stack. And why there's a difference between the both. And quite frankly, I'm not sure If I got that into my brain already. But anyway. I was fortunate enough to being able to get my hands dirty with scripting a few years ago. So that's when I started to get into programming. That's when slowly the beauty of concepts like performance optimization unfolded. So Monday was that day where I thought to myself "You really need to start to clean up that mess". Allright. I tackled the small things first. Merged a few classes and cleaned up the errors when they arose. Until all went fine and dandy again. So, that's where the messy part started...