One other thought here ( apologize, I am just brainstorming here at work ):
According to this
http://ocliteracy.com/techtips/ide-cable-select.html which includes among other things this:
"Pin 28 on the IDE ribbon cable is used for cable select. If a drive sees this signal as grounded, it's the master, but if it floats (unconnected), it is the slave" .
That "hole" i was talking about in the ATV cable is in fact the 28th pin which would indicate that it would be a "Slave" in the atv configuration in other words there is no ground signal from it. When we hook up directly to the logic board, that pin is *not* interrupted as it is with that cable. So even though the atv drive has no problem with this, the sata to pata bridge might be since that pin is connected when we connect right to the board and according to the ide pin out that would be a ground.
As well, there is also the matter of the number 20 pin ( I errantly called it 19 earlier) which is blocked off on the ribbon cable but has a pin in the logic board.
So other than the possible voltage drop issue supplying the pata to sata bridge, the only other inconsistencies I can see are these two pins which are blocked in the atv cable but not in our tests right to the board. Hmmm...
Edit: That number 20 pin may not even have any signal in the logic board, but there is a pin for it in the logic boards connector anyway.