This page is a mirror of Tepples' nesdev forum mirror (URL TBD).
Last updated on Oct-18-2019 Download

Questions about the NES's blacks

Questions about the NES's blacks
by on (#26423)
I was wondering about the blacks in the NES:
    In NTSC PPUs, is $1d supposed to be the same as xE/xF? According to the PPU voltage threads in this board, $1d is the same as xE/xF (black level) when color emphasis is off. Some palettes including Nestopia's contradict this, where $1d and $3e are very dark grays brighter than xE/xF and $0D. Does that mean they inaccurate, or do these colors vary with NTSC NES models or TVs?

    From kevtris's PPU Voltage thread, when color deemphasis is on, $1D gets darker. It also says that xE/xF are always the black level, regardless of emphasis settings. Does this mean that $1d is going to be darker than xE/xF when emphasis is on?

    Besides the Game Genie, has anyone here ever see TVs glitching because of other games that use color $0D? Maniac Mansion (the American version, not the Japanese version), Skate or Die 2, and The Immortal use this color, so maybe those owning a PowerPak could test these games on their TVs, especially if those TVs glitch for Game Genie.

by on (#26424)
MegaMan2 uses 0F as black. Of course, this is software though...

by on (#26425)
In fact, I think color emphasis is reducting the voltage with some analog circuitery, and maybe for $0d it actually increase the voltage (considering the voltage is originally negative), making it less harmful. That's why some games did this without too much problem.
Anyway, as already mentioned a hundred of times, no new game should use that color (nor $1d, $2d and $3d wich are pretty close to other exising colors anyways) no matter what (because they will fade to $0d).
I was told that $1d = $0f, $2d = $00 and $3d=$10 by someone some time ago, but I haven't verified on a NTSC TV, since I have none. I don't know how those colors look on a PAL screen neither.

by on (#26429)
Quote:
In fact, I think color emphasis is reducting the voltage with some analog circuitery, and maybe for $0d it actually increase the voltage (considering the voltage is originally negative), making it less harmful. That's why some games did this without too much problem.

The PPU Voltage thread says that $0D's voltage also gets lower with color emphasis on (it's "L0 lo" in the table), so if what you are saying is correct, then maybe this analog circuitry you speak of may not have been taken account for when this info was found.
Quote:
no new game should use that color (nor $1d, $2d and $3d wich are pretty close to other exising colors anyways) no matter what (because they will fade to $0d).

What do you mean by fading to $0d? Do you mean that when $1d, $2d, and $3d are rendered, they will hit the $0d voltage at some points? Even if they do hit the $0d voltage, I don't think it may be long enough to cause glitching on TVs; the $01-$0c colors don't seem to access the $0d voltage long enough to cause glitching. Or are you saying that games would fade those colors to $0d? In that case, those games could simply redirect the $xd to $0f or some other safe black when they are fading.

by on (#26430)
When a game fades to black (during scene changes), it typically does so by repeatedly subtracting $10 from each value in the color palette, setting the color to $0F when the subtraction goes negative. If one of the colors is $2D, for example, that color would go $2D -> $1D -> $0D -> $0F with this fading algorithm.

Color $1D is not distinguishable from $xE / $xF with color emphasis off. This has been confirmed. With emphasis on, it is slightly darker, but I don't think it's enough to cause sync issues. Colors $2D and $3D are unique AFAIK. Nestopia's NTSC filter should show these colors as they really are.

As I said before, I did once have an old TV that didn't like the Game Genie code entry screen. That TV is no longer in my household, and the one I currently use with the NES has no problems with GG. I suspect that very few TVs will have problems with color $0D, but such TVs definitely exist, so I'd stay away from $0D at all costs.

by on (#26437)
dvdmth wrote:
Color $1D is not distinguishable from $xE / $xF with color emphasis off. This has been confirmed. With emphasis on, it is slightly darker, but I don't think it's enough to cause sync issues.

Actually, now that I look at the table in the PPU Voltage thread, it looks like $1D with emphasis on is only very slightly brighter than $0d without emphasis, so maybe $1d with emphasis on can also cause glitches in certain TVs as well with $0d.