Okay, re-routed csync_o to pin 9 (with 330 ohm resistor). Lifted the nearby 75ohm resistor, which prior to lifting it, tested continuous to pin 9.
elgato capturing is stable at all lines, and pre- and post-coast adjustments no longer have any effect.
I also re-located csync_i connection to S-RGB pin 7 per marqs’ recommendation. I didn’t notice any difference between the two, but if marqs recommends it, I do it. 🙂
Have I done anything else foolish or uninformed? Please don’t critique the assembly work too much. I’ve done quite a few QFP’s but not many SMD resistors and capacitors prior to this. 🙂
And a quick video capture @ line4x, 8:7 ratio: