I was trained a long time ago - PS (pre-shearwater lol)This is how I dive. If I've cut my tables right, it shouldn't be too different from what the computer spits out. I like to pull up the shearwater generated schedule at my first stop and compare it to my wetnotes to make sure they're aligned. If there are meaningful multi level segments (P3 comes to mind), those are factored in to the tables. Average depth usually works though.
If we're down a computer, I'd compare the schedule on the working computer to the tables and (gas volume permitting) go with the more conservative (more deco) schedule if they're reasonably close. If they differ significantly, try to determine why (shallower average depth, shorter BT, etc) and go with the one that makes more sense. If I can't determine why, go with the more conservative schedule. Either way, deco out as a team. 99% of the time, a bit of extra deco won't hurt you.
I consider my buddy's computer/depth gauge to be my backup. I consider his/her brain as a backup to mine.
I would go with the schedule that makes the most sense. This would likely be recalculating my table plan (assuming it didn't already account for what we've done up until the point of failure) based on a shallower than max depth and seeing how that matches the remaining computer. I would expect my buddy to hang with me even if his/her computer had cleared a stop. The further away our obligations are from one another (for whatever reason) the MORE we need to stick together since things are already going sideways.
Last summer I ended up doing almost 25 extra minutes of deco with a buddy due to his computer problem - so our 95 min runtime dive got extended up at 120mins. Fortunately I wasn't on OC. If I had been, the time glitch never would have happened as it was a CCR specific issue I think.