Jo22 wrote on 2023-03-21, 12:52:
I'm wondering if it was possible to record pure monochrome signals (luma) via VHS.
By pure, I mean without recording any chroma information on tape.
Of course it was technically possible to skip recording the color signal. Every VHS recorder had to separate chroma and luma from the CVBS (composite) signal, modulate them in different ways and mix the modulation results together. You could just skip mixing the color signal before recording.
Jo22 wrote on 2023-03-21, 12:52:
Possibly with a higher resolution than standatd VHS?
Maybe slightly higher, but in no way comparable to the quality of pure-luma video signals as seen on the the Apple II for example. VHS uses a FM scheme to store luma, with a center frequency of 4.3 MHz, which limits the video bandwidth to an absolute maximum of 4 MHz, because otherwise the lower sideband would cause artifacts due to aliasing. In practice, they used a bandwidth of around 3MHz, so there is 1.3MHz space below the luma signal. This space was used to store a severly band limited chroma signal. This scheme is called "chroma-under". While you might reclaim the low sideband space used for color signalling as extra bandwidth for the video signal, it is unlikely that the video tape is good enough to also provide an extra megahertz bandwidth above the regular luma band. While it is possible to reconstruct signals that do not have to full side bands (search terms: Single Side Band modulation (SSB), Vestigal Side Band modulation (VSB)), video recorders would likely not be able to hit the desired price point. As most people agree, hitting the price point was one of the main contributing factors for the success of VHS. To allow a higher-bandwidth luma-only signal, the low-pass filter for the luma signal in the VHS signal conditioning would need to be adjusted to allow this higher bandwidth. The same would be true for playback: The filter that removes the low 1.3MHz from the luma circuit would need to be disabled.
Pure-luma video signals often had bandwidths around 6 to 8 MHz, which is way higher than the 4 MHz you might get from a luma-only VHS recording process. For example, the base frequency of a alternating black/white pattern in CGA high-res mode is around 7.16MHz. The IBM font tried to avoid requiring a bandwidth that high, though, by making all regular vertical lines two pixels wide, such that you get a mimum bandwidth requirement of around 5MHz for readable b/w text in 80-column mode. That caclulation is based on a two-pixel wide foreground color segments separated by one pixel of background color. The only place where this frequency is exceeded in latin letters is the lowercase "m" which often had a single pixel wide central stem. So the point I want to get across: You might get better than chroma/luma VHS, but still not good enough for 80-character text.
