If I had the energy to further debate valvetrain geometry, it would be a discussion better suited for a "religion, politics & ford vs. chevy" sub-forum. After dumping another grand on a set of rockers and having bronze guide inserts installed, I'm patently sick of the subject at the moment.
I discussed the subject in detail with Lloyd last year. All the research we could gather indicated that the XFI lobe design requires a lot more spring pressure than Comp specifies to control the valves above 6000 rpm. So if I go with softer springs the cam has to go. It's a really unique grind - basically capable of 480 crank hp in a 350/5.7 while having an idle lope that's less "conspicuous" than the GMPP hotcam.
Anyway, back to the DIY-LTCC discussion - I don't think I snapped any pictures to share but when I built the prototype controller (the one I'll be more or less permanently installing in the car) I added another analog input to the harness for MAP. I'm hoping with the removal of all the floating point math from the code, the mcu will have enough overhead to monitor both system voltage and MAP (for accelleration compensation of dwell). The ADC section of the arduinos is a bit finnicky so I'm not very confident it will work well, but I thought it was worth a shot.