[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: RGB colors



Knut wrote:

I'm wondering about that the gray "colors" are the same. All the emulators do Grey 1 darker than Grey 2 but when I look at NTSC I agree with your calculation, they are same. Is this correct?

Most definitely, yes. For Grey 1 the NTSC video circuits produce "0101 0101 0101"...., for Grey 2 it's "1010 1010 1010".... i.e. in both cases zeroes and ones alternate. It's not difficult to see that the average value (which determines luminance) is 0.5 in both cases. The only difference between the two greys is the way they interact with other colors to their immediate left or right -- the color fringe effects will be somewhat different.

Is the difference something that came with RGB cards and RGB output of IIc and/or IIgs?

Yes.

I think there was a difference on PAL but I might be mistaking there

There wouldn't be a difference if the resistor values in the NTSC-to-PAL conversion matrix were perfect, and if the two color balance pots were perfectly adjusted. Since Apple used normal 5% tolerance resistors in this place and since the pots might be slightly deadjusted, a slight difference between the two greys, not in brigtness, but in hue and/or saturation, is possible.

(it's a pity I never grabbed an image there, the frame grabber does both NTSC and PAL). Or is there a difference between II, II+ or IIe.

Shouldn't be. They work pretty much the same way when it comes to NTSC or PAL video output.

I was toying with the idea to make a summing-network for XRGB (RGBI) to create the proper colors on an analog monitor (or TV with RGB in scart). I would use a PAL (actually GAL) to control a couple of resistors for each color but I need to know what to make... To mimic the KEGS emulator might be a good idea.

Depends on what you want to achieve, IIgs look-alike or NTSC look-alike. Both should certainly be doable.

I should note that the saturation of the colors in my "calculated" colors might be a bit lower than on a real Apple. I assumed perfectly rectangular output. A real logic chip of course doesn't produce perfectly rectangular output; the flanks will be less steep in reality, which means that the result will be more sine-wave-like than a perfect rectangular wave. This higher "sine-ness" changes the first Fourier coefficient and thus increases the saturation.

--
Linards Ticmanis