Like a challenge he would issue, actually, as you've quoted a post from July of 2015 and linked to one from seven months later.Myask wrote:Like a challenge you've issued, in fact.dougeff wrote:I noticed on the Wikipedia page of 'list of video game console palettes' it says they can't show an example of their colorful parrot with the NES palette, due to system constraints...that sounds like a challenge to me.
https://en.m.wikipedia.org/wiki/List_of ... e_palettes
NES with color Palettes of different systems
Moderator: Moderators
Re: NES with color Palettes of different systems
Re: NES with color Palettes of different systems
I decided to give this concept a shot. The .pal files in the attached zip were generated using a CIEDE2000 color difference algorithm that compared Nestopia's YUV palette against palettes that I found in other emulators, on Wikipedia and other web sites. The monochrome palettes were generated by comparing normalized luminance components. The zip includes palettes for the following systems: Apple II, Atari 2600, CGA 1, CGA 2, ColecoVision, Commodore 64, Game Boy, Grayscale, Intellivision, Master System, Monochrome Amber, Monochrome Green, MSX and Tandy.
- Attachments
-
- palettes.zip
- (2.7 KiB) Downloaded 624 times
- BioMechanical Dude
- Formerly AlienX
- Posts: 137
- Joined: Fri Apr 18, 2014 7:41 am
- Location: Bulgaria
Re: NES with color Palettes of different systems
This technical approach is very interesting. In some cases the results are spectacular (Castlevania with the C64 palette, for instance)! In other cases, it's a bit disappointing (In Metroid, Samus is completely yellow with the Apple II palette). The good thing is that the color ramps could be more accurate in terms of lightness. The downside is that some of the color replacements simply don't work. But anyway, this is very cool!
Greetings! I'm That Bio Mechanical Dude and I like creating various stuff like movies, games and of course chiptunes!
You can check out my YouTube Channel.
You can also follow me on Twitter.
You can check out my YouTube Channel.
You can also follow me on Twitter.
Re: NES with color Palettes of different systems
I noticed similar issues with the algorithm, especially with the Tandy palette (EGA) since it does not contain unsaturated colors. The algorithm could be modified to optimize colors over a set of player sprites. I.e., given a large set of sprites, it could find the optimal colors that retain all their looks. And, after that, it would optimize the remaining colors.BioMechanical Dude wrote:This technical approach is very interesting. In some cases the results are spectacular (Castlevania with the C64 palette, for instance)! In other cases, it's a bit disappointing (In Metroid, Samus is completely yellow with the Apple II palette). The good thing is that the color ramps could be more accurate in terms of lightness. The downside is that some of the color replacements simply don't work. But anyway, this is very cool!
Re: NES with color Palettes of different systems
I made 2 EGA palettes, the standard 16 color one used by Tandy graphics, and the full 64 color EGA palette.
[Edited markup to make viewing the images less confusing --MOD]
- View image: Castlevania 3 with regular RGBI colors
- View image: Castlevania 3 with the full EGA palette
[Edited markup to make viewing the images less confusing --MOD]
Re: NES with color Palettes of different systems
The images aren't showing up. You can attach images here, no need to use external hosting.
Re: NES with color Palettes of different systems
Coming back to this, years later...Sik wrote: ↑Wed Jul 08, 2015 7:34 amIt just uses what's set by the hardware. Windows (in general) only attempts to modify the palette when there are 256 colors or more.lidnariq wrote:Windows 3 on an EGA card can't do that; it only has a 6 bit (2 bits per channel) DAC. (in the monitor, in fact) ... but I can't find a reference of what values they use instead.
I thought to set up Windows 3.11 in dosbox with an emulated EGA card and found that its default palette is very odd. Eight of the colors are the standard 3-bit boolean RGB. Dark grey is added. The remaining seven all have extra blue. (This isn't the obvious mistake: it's "secondary green" that's interpreted as "intensity" on an CGA monitor, not "secondary blue".)
It's conceivable it was a typo: if two of the colors (that magenta and seagreen) have the 0 brightness component increased to 1 (to mahogany(?) and fern(?)) then they're consistently the subset 3-bit RGB palette cube in the center.
And the dithering routines haven't been updated for these true colors, and they assume the colors are what they would be on VGA. (i.e. color "#000080" displays as #5555AA) Color "#404040" is special cased to mean #555555 and is otherwise not included in dithering.
Still, less crazy than running Windows 3.11 on 4-color Tandy: It dithers between black, white, CGA RED, and CGA BLUE.