New MC tileset, community project!
#16
Posted 21 March 2011 - 07:53 PM
Also, you're going to have to make extensive use of 8-bit mode. That's part of the reason you'd have to think carefully about what colors go where. It's also why I recommend focusing on gradients: you want to avoid redundant colors wherever possible.
When I started DoR I spent several weeks planning out palettes. I think you should do that for this project, and when it comes to planning palettes, too many cooks will spoil the soup (or however that phrase goes).
#17
Posted 21 March 2011 - 08:01 PM
#18
Posted 21 March 2011 - 08:07 PM
I dunno. I suppose I could help with the palette-planning part. It's a pretty big endeavor. And honestly, I don't think you're going to get very far with the "no color loss" goal unless you create a totally new palette structure designed for 8-bit mode.
But that's not a bad thing. No one's created a tileset designed specifically for 8-bit before, and some would say that it's about time. Minish Cap seems a pretty good reason to try it.
In any case, palette planning has to come before people go on a recoloring spree. After all, how else are they going to know what they are recoloring it into?...
#19
Posted 21 March 2011 - 08:10 PM
Having you to help with the pallettes and anyone else with great talents like your's would be very beneficial for sure.
#20
Posted 21 March 2011 - 08:17 PM
Okay. Great.
So hey... I can tell you what the first step to creating a palette structure would be. You don't have to go out and do this right now, but I'd like to throw this out there.
In order to prepare for palette planning, there are two ways to go about it:
1. The "rainbow" approach: create a palette that is designed to handle as many color styles as possible, without specifically determining what MC will need. I imagine this would be easier, but probably less effective.
2. The analytical approach: go through as many Minish Cap graphics as possible and look at the palettes for each sprite and background graphic. Try to create major categories for the types of color gradients that will be required, and note how many are likely to be required. For instance (I'm pulling this out of thin air):
"Moblin:
5 brownish greys (including white)
4 reds
3 yellows"
After awhile it'll become apparent what types of color gradients are most needed, and which can be incorporated into others. For instance, if there are tons and tons of sprites that require brownish greys, then some of the sprites that use "regular" greys might look good in a "brownish grey" gradient. In turn, that would free up space for other gradients... say, reds.
Another thing that Linkus was suggesting awhile ago is having color gradients that weave into each other. If you want an example, look at CSet 8 in Pure or DoR. The reds and oranges in that CSet are sometimes used as a "mix and match" set when CSet 8 is employed to display fire graphics.
EDIT:
Y'know, I used a Moblin for my hypothetical example, but it occurred to me that enemies (and sprites in general) are going to be a huge problem in this tileset, if it happens...
Yes, we have big Link now, and it's technically capable of taking full-sized MC Link tiles (I think...haven't tried). And NPCs can be any size since there's no actual standard NPC function in ZQuest, unless you count "Guys."
The big problem will be enemy size limitations. Currently, there's no simple way to enlarge enemies, and MC enemies are almost all bigger than their ZC equivalents. That means that either the MC tileset can't have actual MC enemies, or every single enemy that we want included in the tileset will have to be edited for size. Not pleasant.
Also keep in mind that the use of "Level" palettes will have to be limited. Area graphics (dungeon walls, house interiors, etc.) will require the use of 8-bit mode in order to access some of the sprite colors. What will the level palettes be most needed for, then? That's the question. Dungeon/house walls, for starters. As for the rest, time will tell.
Edited by Radien, 21 March 2011 - 08:27 PM.
#21
Posted 21 March 2011 - 08:27 PM
#22
Posted 21 March 2011 - 08:28 PM
No problem. And I can offer more input, too. Mostly I want to do it because I think it will make it easier on everyone else down the road, and palette structure will ultimately determine many of the limitations of the tileset.
#23
Posted 21 March 2011 - 08:35 PM
#24
Posted 21 March 2011 - 08:38 PM
Aside from the sprite size limitations, the main problem is that you have to view individual 16x16 blocks in the pixel editor. It helps enormously if you have an image manipulation program, and know how to use it in conjunction with ZC, but not many people do.
#25
Posted 21 March 2011 - 08:40 PM
#26
Posted 21 March 2011 - 08:42 PM
Hey, thanks a lot. I certainly appreciate that arranging tiles and combos is a lot of work in its own right, especially while making sure that ZQuest's known bugs are avoided (since some of them can mess up a tileset horribly, or at least they could in 1.92). In fact, I think combo organization is far more important than most people recognize.
#27
Posted 21 March 2011 - 08:59 PM
#28
Posted 21 March 2011 - 09:03 PM
I think the MC tileset would be a useful source of tiles, but whoever did it (Linkus, I think?) expressed quite a bit of frustration with ZC's capabilities, and I think that's the main reason we'd need to plan the palettes carefully. Using the palettes of the MC tileset would probably leave us in the same bind as him.
I guess, strictly speaking, it might be okay to start off with his QST file as a base... but all that'd do would be to save the trouble of ripping some of the tiles and creating some of the combos.
By the way, if everything is done JUST right, I have some image manipulation features that just might save a lot of time when it comes to recoloring tiles from the original MC tileset. If we're lucky, I might even be able to use the "Batch Conversion" feature. Just like it sounds, it can save a LOT of time.
#29
Posted 21 March 2011 - 09:05 PM
#30
Posted 21 March 2011 - 09:10 PM
Well, like I was hinting before, I recommend starting on palettes. Get a program which allows you to at least view the palettes of various MC graphics, and figure out what kinds of colors it uses most. If you want to simplify it to extremes, make a list like this:
Pale blues
Bluish purples
Orange-yellows
(etc.)
Note that each of the colors I listed would represent a group of similar colors in varying levels of darkness.
Given that this is MC, I expect it makes more use of pastels and neon colors than most games.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users