Fix color translation tables to not rely on player map arrow colors.

[JadingTsunami]

A suggested fix for kraflab/dsda-doom#130 which does not use the player's map arrow colors as a basis for color translations (which seems like an odd choice to me).

This would bring the behavior of PrBoom+ for color translation in line with other ports.

Comment is invited.

[kraflab]

Makes sense to me

[fabiangreffrath]

Me too! What a weird choice to bind marine suite color to their automap arrow color in the first place. Thanks for figuring this out!

[fabiangreffrath]

Could you also change the automap player color as suggested in the original bug report? And I believe there is also a header somewhere which declares extern byte mapcolor_plyr[] which is now probably obsolete.

[JadingTsunami]

Could you also change the automap player color as suggested in the original bug report? And I believe there is also a header somewhere which declares extern byte mapcolor_plyr[] which is now probably obsolete.

Sure. Updated. This should bring it in line with the original:

https://github.com/id-Software/DOOM/blob/master/linuxdoom-1.10/am_map.c#L1244

#define REDS		(256-5*16)
[...]
#define GREENS		(7*16)
[...]
#define GRAYS		(6*16)
[...]
#define BROWNS		(4*16)
[...]

If someone could test/verify this I'd appreciate it as I do not have access to any networking to try it out. Otherwise I'll have to create a test harness later.

[kraflab]

You can play back a coop demo to test instead of creating a network game.

[JadingTsunami]

You can play back a coop demo to test instead of creating a network game.

Ah, thanks for the tip. I was able to play back a 4-player demo and the arrows look the same between chocolate and this patch now, plus the remaps look the same as well.

I think perhaps the original idea here was to allow customizable player colors by changing the player arrow color, but it seems it was never implemented. I see no reason not to merge this now and get rid of this presumed-buggy behavior.

[fabiangreffrath]

And I believe there is also a header somewhere which declares extern byte mapcolor_plyr[] which is now probably obsolete.

Never mind, it is still used in g_game.c for some reason.

Sure. Updated. This should bring it in line with the original:

Thank you!