Wow, that’s a fantastic post! I don’t know if my missing pixel is anything to do with that, precisely, but it’s certainly worth keeping in mind that what’s in the character ROM may not be exactly what’s being displayed.
At any rate, I’ve got a bit of programming to do at some point to start figuring this out. It would also help to have a monochrome monitor so I could see what’s going on “behind the aperture grill,” as it were, but sadly I neither have one nor know anyone who does. But I guess I can at least trace a signal with my 'scope and get some sense of it there.
The Service Manual doesn’t mention anything about the phosphor, but I doubt it’s different between the two models. (Both are covered by that same service manual.) The primary difference between the 9042Q and 9045Q seems to be just the aperture grille
(0.5 mm for the former, 0.25 mm for the latter).
Thanks! It’s actually just a combination of luck and persistence (pun not intended); it’s not unusual for me to take a half dozen or more shots of each screen until I get one that’s acceptable. However, I got so irritated with this that I went looking again for my proper digital cameras (an old DSLR and a Four Thirds) and found them this time, so now, having manual settings available, I should be able to get what I need with less work. (A macro lens would also be nice for this sort of stuff, but unfortunately all my macro lenses are for my film cameras. And I’m not feeling like going back to film for this; apparently I even have limits when it comes to just how retro one can go!)
Colour TV transmissions started in 1974 in Norway (I know someone who bought a set in 1972 just to be prepared). At my home we switched to a colour TV set in 1976. I’ve never heard about a “tiring effect”. If anything I remember it as easier, not harder, to watch. Norway uses 50Hz refresh rate (interlaced), so I don’t think 50Hz vs 60Hz matters. But of course we had a Tandberg CTV2… as did many others, or they had a Radionette (same technology). There were some other brands of TV with, let’s say, a slightly worse viewer experience. Later I learned how to repair TVs (until I continued with an engineering education) and learned a lot about TV design, good and bad… we started reparing sets while at school, and I did for many years after, also after starting an engineering job. A well-adjusted, high-quality colour TV is not the same as a sloppy designed, badly adjusted one.
I never saw a 50 Hz broadcast until I was (nearly) an adult (probably 20ish?), when I visited England for the first time. I found it exhausting and headache-inducing. I was, however, fascinated by teletext, which almost made up for it.
I assume the problem was indeed flicker, but I also know that the color spectrum and resolution are different between NTSC and PAL.
I can’t speak to color TVs, but I know over time (particularly as I started to need glasses), I would change things to be more comfortable.
When I switched to a Windows PC for a terminal, I found that white characters on black background relieved eye strain.
Several years later, I had one of the then new flat panel displays. It was a 17" model, but it was far more comfortable on my eyes than the monitor (which was larger, as I recall) the it replaced.
This is funny especially today with much be discussed about the effects (or not) of Blue Light. LCDs apparently emit more blue light than normal, with assorted affects including eyestrain.
My current computer glasses have a yellow tint in them to block much of that out.
While we were discussing horizontal resolution, I wonder, was the VIC-20’s resolution of 22 columns tied to requirements to display Japanese Katakana with single-pixel strokes? Mind the origins in the VIC-1001 exclusively sold in Japan and featuring a Katakana in character ROM 2. (For these characters, the C64 way of just having a bold font wasn’t an option. Also, at 22 cols and 23 rows pixels are distinctively more square.)
Compare: VIC-1001 Kana (Japanese VIC-20 Characters)
I vaguely recall reading something that stated that the VIC chip was originally conceived of as a cheap display chip for kiosks. The large text would be friendlier and easier to read than small text (especially if the kiosk itself had a small display - such as for a retail checkout). And the display would require less RAM simply because it had fewer characters to store.
I don’t think katakana factored into the decision at all. I mean, the VIC-II chip was also first used for a Japanese product - the Commodore Max. It had a 40 column display.
I didn’t know of the background in a kiosk display. This would certainly make sense.
(The weak point in my argument is that the chip design predated any marketing decisions rather significantly. While it could have played a role in the decision of which chip to pick, in actuality, it was more the other way round and the VIC-20 was designed around the VIC chip. Apparently, there had been an inhouse 80 cols alternative, but this would have required lots of expensive SRAM. This one was intended to be used in the TOI prototype, but I’ve no idea, what kind of display had been intended to be used with this.)
This video maybe interesting in context, regarding home computers and RF video output. To me, this looks much like the luma and chroma signals were deliberately softened (blurred) in RF modulators, probably to deal with some of the issues referred to here.
(As I understand it, the RF modulators used were pretty generic devices and not specific to a particular computer. So these were also addressing more generally problems with image generators and TVs found in the real world, I guess.)
The oscilloscope trace of the signal of an early PAL VIC II, shown a bit later in the video, shows also that signal flanks were also an issue, at least with some video chips.
