Hugo, a question about video timing - Was there ever an accepted timing standard for computer video output for 15KHz monitors, that were's PAL or NTSC, but either mono or color with separated signals, whether analog or digital?
I mainly worked with analog video, PAL, NTSC and the timing standards for these are well known. I'm more familiar with these than computer signals. The Chroma sub-carrier is easily switched off if access to the circuit that generates the composite signal is there. You can also switch of just the burst and the set's color killer will deploy the TV to monochrome, but there will be patterning from the high frequency luma signal with the chroma, so it is not as good. Also the sync generators for items like PAL & NTSC color cameras, video test and other generators, and off air TV signals have proper interlacing with equalization pulses too. You can also combine RGB signals to a good quality luma signal, or just use the green if desperate.
However, when it comes to computer generated signals, especially from vintage computers, it appeared to be "anything goes" looking at the circuit designs. If the picture can come into lock, even with slightly non-standard signals for the VDU, and has reasonable resolution it was "good enough"
Initially many signals were not interlaced and the fields merely scanned on top of each other. In the Dazzler for example, it was a fairly poor attempt to generate NTSC standard signals, it doesn't even use a suppressed carrier modulation system for the chroma, but, most color VDU's were very forgiving. It seems to have taken a while before early computers were designed with more specification compatible signals for NTSC and PAL VDU's.
But then, generally it was as though the manufacturers wanted to set their own standards for computer video. RGB style systems rather than composite video of TV systems, higher scan rates and better bandwidth in the video amplifiers and finer dot pitch color CRT's than what any common garden TV PAL or NTSC VDU could produce, in the interests of detailed graphics of course. Looking at the color rendition on a composite signal from a IBM CGA card that gives a different result on a color NTSC VDU than a CGA VDU, like the 5153. It is as though they didn't really care about a color composite signal, but cared a lot about the CGA signals.
I think when early computers came out, the costs were a barrier (except perhaps for those buying IBM computers that came with recommended VDU's like the 5151, 5153 and 5154), so the makers provided composite video outputs so the buyer could deploy an existing TV set or VDU, like those used in CCTV systems, even though it would not have been as good as a VDU specifically designed for the task with an RGB system of one kind or another or a high reso dedicated monochrome VDU. But the IBM 5151 was a good choice, though they cut back on the circuitry in it, making its H output stage vulnerable to damage from incorrect signals.The 5153 and 5154 color VDU's don't have that issue.
Interestingly the same sort of thing happened when aftermarket CDI's for cars. When they came out in the mid 1960's the manufacturers claimed that you did not need to change the existing Kettering style ignition coil. (the extra cost would have been a barrier). However, this was not entirely true because a custom transformer style coil with a Ferrite core, more suited to the task of CDI, would have yielded at leas 4 times the spark energy with the same drive unit, talk about telling a Porky. But it would have made the whole box of tricks a lot more expensive and put off a buyer.
One monochrome VDU which I think is an interesting case in point, is the one in the IBM-5155 computer. It was not a re-purposed TV signal standard VDU, as some small VDU's used in early computers were. I was designed by Zenith with a wide-band computer composite signal in mind. It has a CRT with a very small focused spot size. Geometry correcting magnets on its yoke for a perfectly rectangular raster scan, a video output stage with near twice the bandwidth of that seen or required for a TV VDU. And it can produce highly detailed graphics as a result. The only error IBM made with its deployment was to feed it with the composite color signal from the CGA card (hijacked from a connector that was intended initially to feed an RF modulator). The color carrier beats aggressively with the higher luminance frequencies and due to the excellent response in the video amplifier in the VDU, this results in objectionable patterning on graphics. The fix is to disconnect the resistor that mixes in the chroma to the video output transistor on the CGA card. These particular VDU's "turned my head" and I became unable to resist their power over me:
www.worldphaco.com/uploads/THE_AMBER_COMPUTER_VDU_PROJECT.pdf
One other interesting feature of these VDU's, not common ( Though I have seen it done in mil spec Radar VDU's, using a UJT) Zenith created the vertical scan oscillator from two transistors wired as a PUT (programmable unijunction transistor) the method avoids the need or requirement for a vertical hold control. Makes me think that the designer probably did stuff for some of Zenith's mil spec systems.
