Hi! I'll try to be concise.
Both ports seem to work properly, using my single 2-block EDID: I have a Quantum Data 882 with an EDID parser pattern, showing me what I want to see on both ports. However, if I swap my two front-end hardware chains before the 7612 (as is required for our final product), there is some communication failure on the ADV7612 DDCB lines (pins 93 and 94). On the scope, the clock is constant and the data line is never quiet. I'm imagining that it's constant polling/querying, like some handshake just fails, repeatedly. In contrast, when I have an image (non-failed state), I only see very brief communication every few seconds. I have tried countless software and hardware configurations (eg: changing hardware, or swapping channels downstream in my software), and I can only produce this failure on ADV7612 port B.
Any suggestions that might help resolve this are very welcome.
In the meantime, I'd like to set up port B specific EDID, to see if that solves the issue. I'm looking at page 34 of ADV7612_Hardware_Users_Guide_Rev0.pdf, at a section called Structure of Internal E-EDID for Port B. It's rather more complex than the setup for Port A. Any notes or tips would be greatly appreciated.
Thank you for your time!