Maybe a thread should be made specifically for the guide so more people will see it?

I paint everything in 24-bit because I know the palette is pretty big at 256 colors. With so many colors, it's unlikely that the image will be mangled badly when indexed automatically. Doing doing stuff by hand is always better, but I don't think it's so much better that it warrants the extra time it takes.

Also, it's good to have 24-bit source material. I want to rewrite my image converters to be a bit more user friendly. Preferably it should be a 1 button push kind of deal. Source material in a 24-bit/ folder structure, then the program just targets and overwrites the game files.


Sometimes I keep a palette in another window and color pick from that palette, even though I'm working in 24-bit. This ensures that I use colors close to the safe ones. Here's an example of a simple texture for Quake 1 that I painted this way. One side has been indexed.



If I were doing 16 or 8 color sprites, I would work in 24bit at 100% opacity until I feel that I have used 16 colors, then I would index and continue.

I have attempted to remake the CC palette several times. I'll get it right one day. I just started on my 5th or 6th palette tool. Taking the gradient/ramp approach this time.

Playing around in my new program. I haven't tried the palette on an image yet so I don't know how good the coverage is. I put most of the focus on the mid values, because it's hard for the eye to tell much difference between dark/bright colors, especially when most of the game has a mid value. I might be able to merge some of the desaturated ramps, as the difference is smaller between those. Could free up 2 new ramps by doing that.

After some quick experimentation with that palette shows that there are a lot of near duplicate colours (mostly whites and greys) and while it gives better coverage it also takes a little bit of the contrast out, maybe make some more extreme tints for things like yellows and reds? the skin ramp is very nice though.

Excuse the nievity, but would changing the pallet break all of the current sprites? That seems pretty drastic, so I'm probably wrong in thinking so.

Yes, it will. Every sprite in the game will have to be put back into the new palette. I'm assuming Arne has his larger source images stored somewhere in 24bit.

Indeed. I wrote a program for when I made the old CC palette which highlighted HSV and RGB pythagora distances. It allowed me to optimize away the dupes. Also, since the old CC palette doesn't use ramps, it can cover more space.

I just tried this palette on a few images and it's pretty much useless. I'll try to nudge a few of the ramps towards CC colors and see how it fits after that.

EDIT: Even after making gradients/ramps from color picked CC colors (the 24 bit material), I get a worse fit than with my old CC palette! It's not really too surprising. I made that palette by directly indexing a lot of my 24 bit material, along with a whole bunch of general images from the web (nature stuff, animals, toys, humans, naked humans... etc.). Then I merged some of the close colors and used the freed up space to add some colors of my choice.

Why is it that you are changing the pallete? I don't see any immediate problems with it, but then again I've only ever drawn grey/black stuff.

It's got an okay grayscale palette, but the rest is pretty bad.
Hey Arne how about a bit more hue shifted palette? Stuff like graypurple and redpurple could be mixed together, with some colors being more gray etc.

A more versatile palette can't hurt.

Good thing i also store my sprites in 24bit, hope you will release in public your converting tool it really takes me a lot of time to convert all of them by hand. Already have like 600 sprites and the mod isnt finished yet.

I wanted some better ramps in the old CC palette, but it works OK otherwise. The new one that I did doesn't work at all for me, so I'll drop it for now.

I have put the program here (source included). Maybe it's useful for something else. Use at your own peril.

http://androidarts.com/PalEd/


I had an urge to write a conversion tool some days ago, but ended up staring at 10 lines of code for an hour and eventually lost the will. Programming a public tool is different from making an internal one, because it has to be stable and user friendly. Quite a bit of more work.

Now the promo art that I promised to link here.

http://androidarts.com/HIB2/

Nice arts prom, liked the one with baloons.

Oh the day that we have martial arts in this game!
[Yes, via slowmo, and limb dragging, the problem about this is how this would pan out in multiplayer]
Whatever, great promo art, I think they should at least be tempted to use it.
[I am also curious about any unreleased dev art]

I love the balloon promotion art. Just the fact of the coalition solider trying to help them up is epic.

Very little sleep, candy/snack diet, snatched away from the computer, for days. I feel blunt.

Anyways, this guy is a leftover from one of the HIB images (new one at the bottom, btw.).



I might make a little icon/decal or something out of it.