[grishka]

I grew quite an appreciation of color TV recently when I made a software-defined SECAM decoder. It's mind-boggling that they were able to do this with 60s technology, to be honest. But then what do I know, analog electronics is witchcraft to me :D

https://github.com/grishka/miscellaneous/blob/master/AVDecod...

edit: I watched the video. It only took them until the 13th minute of a 15-minute video to show something resembling a real video waveform, lol. That's, uh, not how you explain how TV works.

[pavlov]

SECAM was pretty crazy because it required a delay line: a memory that would hold the previous scanline so it can be combined with the current one.

Without digital circuits, the delay line was a piece of glass. You’d convert the video signal to a sound wave, send it through the glass, and (hopefully) get it back exactly 64 us later so it aligns with the next scanline.

Here’s a picture: https://www.eevblog.com/forum/projects/glass-ultrasonic-dela...

[grishka]

Yes I did read about these while doing my research. Also fascinating. Most (or maybe all?) PAL TVs also have a delay line to correct for phase errors. That ability is what differentiates PAL from NTSC, apart from timing.

[lbourdages]

Fascinating! So it's a bit like a plate reverb (used in the olden days in audio engineering), but in the MHz range.