Something I've wondered for a long time about VHS tapes, is that because of the nature of the FM luma signal, do bright colors have a higher resolution than dark colors, because bright colors have a higher frequency than dark colors?
Not bright colors necessarily, but it's possible that a less saturated color (like one approaching white, occupying more of the spectrum) appears brighter. As a less saturated tone will occur at more points along the subcarrier's rotation through phase, it can be better represented as the line marches on and the subcarrier continues.
I meant bright as in light, not bright as in saturated.
I was looking into how to demodulate an FM way, and one of the methods is to use a Hilbert filter to shift the phase by 90 degrees, and then find the arctan of the two signals. An exact Hilbert filter would be extremely long, so I calculated a shorter phase-shift filter that would approximate a 90 degrees phase shift from 1/3 to 2/3 Niquist frequency.
{-20, 0, -148, 0, 148, 0, 20}
I derived these numbers using trigonometry and algebra.
148 = 256/sqrt(3)
20 = 148 - 128
The (NTSC) VHS luma frequency carrier is between 3.4 Mhz for sync, and 4.4 Mhz for white level, so you sample a VHS signal at 13.5 Mhz, the phase shift will be accurate from 2.25 Mhz to 4.5 Mhz. If you want to emulate a VHS tape of a Sega Genesis game, you can sample it at 2x the pixel clock, because it's very close to 13.5.