Dragon Quest project progress
I was surprised to see the theme of Google's April's fools, because I was just noodling around on my Dragon Quest project.
skip to main |
skip to sidebar
Updates
- January (1)
- September (1)
- May (1)
- April (1)
- March (1)
- January (1)
- November (1)
- September (1)
- August (1)
- May (2)
- April (3)
- March (2)
- December (3)
- November (2)
- July (1)
- May (1)
- March (5)
- February (1)
- January (2)
- December (3)
- October (1)
- August (1)
- June (1)
- April (1)
- March (1)
- January (3)
- October (1)
- June (1)
- March (1)
- February (4)
- September (1)
- August (1)
- July (2)
- June (1)
- April (2)
- March (3)
- January (1)
- December (5)
- September (1)
- August (1)
- June (1)
5 comments:
exciting! can't wait to see more
Looks like I have to do tax declaration now though. Also, spring fly caught me.
I have some ideas for plopping down some coastal tiles, be it beaches or mountain tiles. A beach tile would be handy, as it would allow for placement of some interesting amphibian monsters.
I have no clue how to handle border tiles (e.g. transition between ocean and land). DQ1 didn't of course, but if I do highrez painted stuff I might need some border solution. Perhaps an overlaying dimension which stores contrasts, then draws sticks between the tiles. I dunno.
you could look at how games like Final Fantasy VI and Seiken Densetsu 3 did it. of course it's a bigger scale than overworld tiles, but i guess the tiled principle still applies somehow.
best of luck (and hope that this project sticks around a bit longer)
If you don't mind sharing I'd be interested in more technical details of how you implemented the map generation. I was really inspired by this article by Amit Patel: Polygon Map Generation
It uses Perlin noise so there are obviously some aspects about it that you might disagree with but I thought his results were pretty good. It also uses a lot of other math to make the final result good for games and his use-case.
Mostly I am curious to hear the details of your "clouds" that rain down land. I am also curious to hear more about what looks unnatural to you about using Perlin noise to generate maps.
Have you thought about posting some of your code on a place like github or open sourcing the stuff you work on?
Regarding tile edges, the problem is that I don't know anything about commonly occurring adjacency, since the maps are random. This means that I have to cover all possible combinations, which means that I probably have to split the edges up into 4 or 8 neighbor images and use transparency or something. Then there's the question of what overlaps what. Are the transition on the inside or outside of the tiles, or in between?
Regarding the polygon map gen and perlin noise gen, I don't think it would suit DQ's map style which have a certain hand drawn aesthetic to it (like some artist just decided to throw down a nonsense mountain range maze). I think perlin noise, unless processed in some special way, often looks samey, and very distributed, lacking interesting oddities. It's what the name implies, noise. The landforms it produces often look unnatural in their jaggedness, spacing and I'm often able to tell right away if perlin noise was used. There are probably ways around it, but I felt my cloud method was more straight forward... just slap blobs down and slap blobs down on the blobs. I can't really sum it up what's going on because the code does a lot of little things which are necessary for the effect. I'd have to upload the source which I don't dare to do before I've commented it.
Post a Comment