Posts Tagged outsourcing

With a concept artist in tow, it was time to get some in-game art.

I’ve always been a huge fan of pixel art, and not just due to nostalgia. I really like the style and its use of colours and textures, and its overall ability to convey visual information. To put it simply, I think it’s charming, and it would’ve made a great fit for Tribes of Mexica.

Over the span of the last decade, I’ve worked with about a dozen pixel artists at two professional game companies, and met countless others while pursuing hobbyist projects. I even enjoyed my own time with DPaint; I was decent enough with tilesets, but horrible at animations. Unfortunately, all these artists were either too busy or were working under non-competitive clauses that prevented them from helping me out.

As such, I had to find someone new for the job.

A group of 2D games that were never meant to be. The above list contains mocked-up screenshots from various pitches we worked on at my last company (art by Eric Vedder, Ben Henry and Joe Pendon).

Before that could happen, though, I was determined to tick off a few goals that would hopefully aid me in my search. First, I wanted to hire a concept artist to do some work that would serve as a reference and a token of my dedication to the project. Next, I wanted to code a “tech demo” that would not only be tangible proof of ToM, but would also help me figure out some of the art requirements (such as dimensions and animations). I also wanted to get a single music track for the demo and make a video out of it as it’d be a useful tool for showing off the game.

Finally, I needed to do some research.

One of the reasons I thought pixel art would be a good fit for ToM were the sizes of the in-game entities. They ranged from 36×36 all the way to 200×200, making them very suitable for pixel art. However, the backgrounds weren’t. My prototype ran at a 1280×720 resolution, which was huge even without any scrolling. I certainly didn’t need a tilemap, but this would still be a potential problem. I discussed it with some artists I knew, though, and they assured me that it was easy enough to draw and scale a background, and then paint over it with a somewhat limited palette while avoiding smooth gradients, blotchy brush strokes and anti-aliasing. This would give the background a somewhat pixelated look that would blend in fairly well with the entities.

With all that out of the way, I posted a job ad on the Pixelation, Pixel Joint and IndieGamer forums.

My initial request consisted of the following:

• 4 backgrounds, 1740×720, with some decorative objects that could be moved around.
• One “unit” type consisting of 16 angles. Well, 9 angles, really, as the other 7 could be flipped/mirrored without any extra work. Each angle would contain 4 animations, 2 of them single frames, and the other 2 comprising 2-3 frames. Each unit would be 36×36 in size and would need to be provided in 4 coloured versions via palette swapping.
• 6 enemy types with 4 animations each. Once again, 2 of the animations would consist of single frames, and the other 2 would be 2-3 frames each. Enemies ranged in size from 36×36 to 200×200.

Konjak made a name for himself with Noitu Love 2 and various other hobbyist and professional games. Although we never talked before, and he was quite busy when I sent him an e-mail, he was also nice enough to quickly respond in a cordial manner.

I didn’t post a specific fee for these as neither I nor the other Incubator Games guys were sure of what the price ranges should be. We didn’t have a specific budget set aside for the in-game art either, so we decided to roll with the punches. If the offers we received involved sums that we could immediately afford, then great. If not, we’d either have to scale down or look to pool together some more money.

The lack of a specific fee in my post probably meant that we wouldn’t get as many offers as we did via my concept art ad (at least not initially), but I was OK with that. I also noted that if artists were only interested in doing the backgrounds or the animations, they should still apply. Obviously I would’ve preferred to have a single artist do all the work, but I knew that some would be experts in only one area and would want to avoid the other.

What quickly became apparent is that most pixel artists judged the budget of a work strictly by the amount of pixels it involved. What this meant is that the price tag of an image grew exponentially with its sizes. For example, if a 50×50 image cost $10, a 100×100 image would cost $40. This lead to quite a few misunderstandings over the backgrounds. I would’ve been crazy to expect artists to create these large images one pixel at a time, but that seemed to be the overall impression I had made. I was forced to repeatedly explain that I did not want these to be pixel art, but rather drawn/painted, and then touched up to blend in with the pixel work.

Following the first batch of offers we received, I decided to scale down my request. It wasn’t so much an issue of money as of time. Despite the 5-week deadline, some artists expressed concern with the amount of work involved. To accommodate for this, I changed my requirements to include only 1 background and 4 enemies. Ultimately, I also took down the background size to 1280×720 and the amount of angles for the units from 16 to 8 (for a total of 5 angles per unit if you don’t count the flipped versions).

Of all the artists who expressed interest in the project, I was particularly fond of Jim Jansen's work.

Since I posted on multiple forums, it was a bit difficult to track all the responses via Private Messages and e-mail. All in all, though, I received roughly 25 offers via Pixelation, 15 via Pixel Joint, and 5 via IndieGamer. I also personally contacted about 20 artists, although only 5 of them responded. The monetary estimates in the responses ranged from $400 to $1100 at rates of $15 to $40 an hour, with one or two that were significantly lower. The breakdown for my requested assets was generally the same: the units cost the least, followed by the enemies (the 200×200 enemy was more than all the units combined), and finally the background. Based on these estimates (and feedback from artists I knew saying that the work shouldn’t take more than two weeks to complete), I edited my original posts to include an $800 price point.

Adam Saltsman, the admin of Pixelation and a somewhat prominent figure in the indie games community, did not agree.  When I added $800 to my ad, he responded by saying “I could smell your cheap, exploitative rates from a mile away…” following which he moved my post to the “unpaid” forum. Pixelation is his to run as he sees fits so I didn’t complain too much, but this is my blog so I’ll briefly state the issues I had with this:

• Right off the bat, Adam claimed that this work was worth $3000, or even 2-4 times that amount. This wasn’t even close to the estimates I received, so it’s hard not to think of it as an inflated number. I even discussed this with some of the artists I’ve worked with, some of whom Adam knows personally, I think, and they all seemed to agree.
• Despite being rather friendly and self-effacing on his blog, Adam’s comments on my post were pretty snarky. I’m a big boy so they didn’t really phase me, but they did seem rather unwarranted.
• The $800 was called exploitative and not a competitive figure, and yet it was based purely on what the artists themselves offered. It wasn’t a lowballed amount either as $15-$40 an hour isn’t a bad rate, but Adam ignored all of this. Considering this number was derived from competing artists, I’d say “competitive” was a very apt term for it. I’d also go so far as to say that the process by which I settled on this amount was very fair and didn’t exploit the employee or the employer.
• $800 is long way from $0, so I wouldn’t consider the offer to be “unpaid.”

Anyway, in the end we had about 10 or so artists that we thought could be a good fit for ToM. This list included Jim Jansen, Miguel Angel Perez, Joshua Astorian and various others, but we didn’t pick any of them. The reason for this is that the commission was for only one part of the whole game, and we wanted to hire someone that would end up doing a lot more work. The pixel stuff was doable for a demo, but ultimately it would require too much time and money for what we had in mind.

Instead, we decided to go with a different approach, which I’ll talk about more in the next post.

Next Tribes of Mexica post.

Previous Tribes of Mexica post.