Board Game: Tavern Wars (and AI Art)

About 7 years ago I spoke with my friends at Attic Door Media about creating a board game. We sketched out some basic concepts, and they started looking for artists and developing a kickstarter plan while I designed the game. We printed out proxies, and playtested, then I made edits, edited proxies, playtested again, etc. Finally, I felt ready for us to try putting it on Kickstarter...  

But getting the art was actually much harder than we expected. I'd designed over 150 unique cards, and that would cost tens of thousands of dollars to get art for in any sort of style we really liked, and while some mega hits definitely can recoup that sort of cost, the vast majority of board games do not make nearly that many sales.

I cut the card list down, but it was no use; the amount of time and energy it would take to find an artist we could afford sapped me of my will to keep working on the project, particularly once I started to dislike one of the central mechanics.  I've fiddled with it here and there over the years, but the looming specter of getting the art for the project kept me from being truly motivated, and I had plenty of other projects to work on meanwhile.

Over the years I kept an eye on the rise of AI generated art, wondering if it was a potential solution... but knew I would have to accept very abstract pictures if I did, which might be fine but also would feel off for some of the cards in particular.  Finally, I learned something that a lot of others also seem surprised to learn: you can now spot-edit parts of images created by Midjourney, and in my opinion it's an absolute gamechanger for making specific pieces of art.

You don't have to keep rerolling for the perfect combination of hair color/background/finger count/etc. You can get a picture that's 80% of what you want, then just outline and reroll the parts that it got wrong until it gets them right.

This picture in particular was hard to get right. Midjourney *really* didn't get what a "quest board" or even "bulletin board" was, or maybe it was something about setting it in a tavern that made it hard, and then there were all the complications that come from having a person in the picture, not to mention a bunch of random brickabrack on the table.

But thanks to spot editing, this picture took me just ~15 minutes of fiddling around to get to where I like it, and that makes my board game project (which requires over 100 pieces of unique art) actually achievable with my already packed schedule, instead of an extremely time and/or money expensive pipe dream.

I'm not ready to make any announcements about it yet, and it's pretty different from my other projects here, but I thought I'd let you guys know, since I wrote about this game years ago when I started my Patreon.