6

On early color computers, it was possible to add a feature by which you could turn off the color burst output, restricting the display to black and white but making it significantly sharper and clearer; this was useful for text. I don't know how much it cost to provide this feature, but I am inclined to think it would be useful for any machine intended for serious use rather than just games; its historical rarity suggests either the cost was nontrivial or there is another issue I'm not taking into account.

The Apple II had it from early on (including a useful mode where you could have most of a screen of color graphics, with four lines of crisp black and white text at the bottom).

The Tandy CoCo tried to provide it, but the first version allowed the color burst to leak through, and a revised chipset that would have fixed this, was rejected in testing because it broke games that had been written to exploit the artifact colors in what was supposed to be black and white mode.

Did any other early color computers (that output RF or composite video) allow the color burst to be turned off?

  • 1
    Not a real answer, but in principle a demo coder could achieve it in effect on an Atari 2600 because vertical sync is programmatic and is combined with horizontal sync via XOR. So with a very, very delicately-timed loop you could turn off the real horizontal sync and put your own somewhere else. Which wouldn't remove the colour burst, but would move it so that it wasn't where it should be, so was no longer a colour burst. I would dare imagine other platforms might offer similar demo-coder exploits even where it's not an intentional feature? – Tommy Aug 29 '18 at 19:53
  • 2
    One of the benefits of separated video (luminance+chromiance) on early machines like the C64 and Atari 800 is the luminance signal can be used by itself to drive a B&W monitor with no color interference. This is because the luminance signal includes the H-sync and V-sync as well. – Brian H Aug 29 '18 at 21:27
  • 1
    Some monitors allow switching between green and color. Maybe someone can answer, do they do a good enough job of removing the color burst signal, or is it significantly better to not send the signal to the monitor in the first place? – traal Aug 29 '18 at 21:50
  • 1
    The Atari 2600 doesn't allow enough control over vertical timing to produce a usable signal without colorburst, but there wouldn't be much point since it can directly produce 8-level grayscale, and its resolution is low enough that a grayscale picture won't tend to pick up unwanted color fringes. – supercat Aug 31 '18 at 19:09
5

The V9938 video chip in the MSX2 home computer can be programmed to disable the color burst but I don't know if any utilities or applications provided such an option to the user. That would have certainly been beneficial for the 80 column text mode and 2-color high resolution bitmap mode.

Some MSX2 variants had a V9958 instead which has no composite video output and no color burst control function so it depends on the particular manufacturer.

2

An Apple II did not turn off color burst for the 4 lines of black and white text at the bottom. Instead, it output wide enough (low-res) pixels for an NTSC television or monitor to interpret the IQ phase (with respect to color burst phase) as equivalent to white or black.

  • 1
    I don't think this is right. The Apple II definitely turned off the color burst in text mode (no color fringes on the text), and I don't see why it shouldn't do that in mixed mode when essentially switching to text mode at the bottom. Yes, wide pixels would be white or black, but they still had the color fringes in hi-res mode. In doubt, one could look at the schematics. – dirkt Aug 30 '18 at 5:35
  • 1
    Jim Sather in archive.org/stream/understanding_the_apple_ii#page/n199 says that the color burst was left on in mixed mode, and doesn't mention pixel stretching at all and explicitly says that text will be green & purple. – PeterI Aug 31 '18 at 16:40
  • 2
    The effect of turning color burst on and off mid-frame is apt to depend on enough factors as to be essentially unpredictable. If one were to turn on colorburst for some area in the middle of the display (possible with a well-timed switch to text mode 40% of the way down each frame, followed by a switch to graphics mode 60% of the way down) most monitors would continue to behave as though the colorburst were present during most or all of the portion of the screen where it was off, but the color phase would drift). Depending upon how much the phase drifted, they may then... – supercat Aug 31 '18 at 19:06
  • 2
    I suspect a physical switch was used because the effect of taking what's supposed to be a monochrome image, decoding NTSC color, and re-encoding PAL color, would be far worse than the effect of feeding such a signal to an NTSC monitor. If the Apple ][ had a soft-switch combination that would yield hires without colorburst (I'm not sure if that would have required any extra circuitry) then PAL systems could have used that to control the decode/re-encode process, but no such feature exists in the NTSC machines. – supercat Sep 1 '18 at 18:34
  • 2
    @PeterI: Thinking about it, perhaps what would have been most useful might have been a potentiometer to control the blend between transcoded chroma and delayed (accurately to be in sync with chroma) luma, since different programs might benefit from different trade-offs between luma clarity versus colorfulness. – supercat Sep 1 '18 at 22:43

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.