I would also like to add that using two dive computers does not really mean full redundancy.
Complete methodologies and classes of devices can be rendered useless in a heartbeat. Having one, two or ten computers does not help then.
Example 1: Confusion
I was diving at 100 ft using EAN28. Visibility was incredible. I was warm in my drysuit. I felt relaxed and I had eaten and slept well. Suddenly the line tender gave me a return signal. I did not understand why. My computer also beeped and I could not understand why (I had forgotten that I had a preprogrammed depth warning for 75 ft from a previous dive). I was utterly confused. Then I looked at my computer. There were numbers of different sizes in the four corners. That was all I could understand. I had just forgotten how to read the thing. Even if I had carried two or more computers, they would all have been useless. It took me half a minute to gather myself. An analog depth gauge would have been understandable. In an emergency I might not be able to think clearly, either.
Example 2: Loss of mask
I allways carry a spare mask, unless I have forgotten or lost it. But it happens, rarely, that I dive with a single mask. If I lose my only mask, then I cannot focus my sight on a dive computer. I have tried. It does not work. It does not matter how many dive computers I have. None can be read. A large analog depth gauge might be readable. It is yet to be tested. Bubbles might be visible without a mask. Have not tried. The suggested monitoring of pressure equalization in ears is another interesting idea.
Anyway... it is clear that we need several independent and completely different ways of monitoring critical data such as depth, time, ascent rate, air, battery and return path. Hence, discussing alternative methods should be healthy, even though they "will never be needed". Titanic had redundant waterproof walls, remember?