Hi all, I'm starting to work on a new personal project. Been awhile since I work on that and got really inspired with the recent creatively directed movies, like Arcane, The Witness, Spiderverse, TMNT, etc. With that being said, I decided to remake my childhood game, Final Fantasy IX.
One of the most memorable scene personally was the end of Disc 1, where Zidane and the party got wiped out by Beatrix in Burmecia. So I decided to search for related concept art online and found these:
I think these are all great references to start with for the environment. So now I had to find some references for the characters. I found a mix of both official key art and their concept art.
Now that I've laid out most of my references, I think it's a good time to find a style guide. As I mentioned before, I was very inspired by these recent shows, something that looks 2D and moody, which I believe fits the storybook style of FF IX.
Aside from the overall look style guide, I tried to look for a character style guide too. I've always been a huge fan of Akihiko Yoshida who made the artworks for Final Fantasy Tactics, Bravely Default, etc.
Fortunately he also made some artworks on FF IX, thru a collab done for a mobile game FF Brave Exvius - War of the Visions.
Another thing I did was to find some real life references in how these lighting setup works:
Not many references I can find that has such a moody look in the lighting. But in general the heavy fog situation diffuses any sense of light direction and made foreground elements in shadow. I like going back to photographic references to start with so I have a strong base on how the light bounces should look like.
Before ending this first post, I wanted to show some stuff I played around with. With all the AI craze, I thought it would be good for me to try use them as a way to create a concept art of how I envisioned these characters would look in the style guide I made. So I went into Stable Diffusion and try to recreate some of the official 3D key arts into a more 2D, and the results are horrible:
I'm only showing the more "okay" results here. In general. for a more normal human characters, SD seems to be able to generate a much normal proportions and look. When it gets to characters like Vivi or Quina, as you can see it fails to create any real relevance to the character. Granted, I came in expecting AI would do anything with a "touch of a button and a few prompts". Unfortunately (fortunately for us artists?) that's just far from the truth.
I still don't believe AI will take over our roles as an artist and a creative individual. It's definitely an interesting tool that can help out in terms of brainstorming, but I can't see it replace us 100%. At least, not in any way that I would operate in this industry. Not including the amount of R rated content over flowing the internet. As of now, I believe we can try using and implementing them into our workflow. I'm still a huge advocate of reducing unnecessary tedious work that can be replaced with tools, and AI is great at that. Give them tons of stuff to crunch through, and they'll thrive. Give them the "liberty of creativity", you'll have a hard time.
Ok that’s it for this week’s post, see ya on the next one!