Posted by Kendall Raine on February 15, 2005 at 15:36:49:
In Reply to: Broken record posted by Ross-O on February 14, 2005 at 11:14:27:
Hi Ross;
If you don't mind I'd like to share some thoughts raised in this dead horse debate.
It's actually kind of scary how little validation goes into dive computers, either the algorithm or the electronics. Remember the Aladdin nitrox computer which assumed the surface interval breathing gas was the same percentage O2 as the bottom (nitrox) gas? Oops. That little boo boo took years to find and a lawsuit to bring into the public domain. How about the Abyss trimix computer which worked great until you took it into the water whence it flooded and stopped working-kind of like the manufacturer itself. If memory serves even the venerable Edge had less than 200 man dives on it before it was released and then required a series of software upgrades (5) to keep it from telling you things like pull 85 minutes of deco at 2 feet.
I find the "take my course and pay me money and I'll explain the meaning of life" crap as silly as you. Consider the source. But the fact is that many computers sold today use deco concepts decades old and out of date. They certainly haven't stopped people from bending themselves and that's the ultimate test. Worse, they can end up being a substitute for understanding what's involved with decompression. People regularly do six to eight dives a day during bug season because their computers say they can and then wonder why they get bent. Compounding that, a programming error like the Aladin situation is pretty tough to spot, particularly if you're into blindly trusting the computer.
The best line of defense against the bends is don't go under water. The next best is actually learn something about decompression, physiology and inert gas dynamics and put that knowledge to use in how you dive. If you do that, a dive computer ends up as no more than a security blanket for your brain, anyway. After 11 years my daughter still has the shreds of her security blanket. I see nothing wrong with that because it doesn't substitute for other things.
The assertion that not having a computer means you devote all your intellectual bandwidth to monitoring depth and time is hyperbole. You can do that and still have lots of attention span left over to enjoy the surroundings. There is very little computation involved. Adjusting for deviations from plan are as simple as applying a couple of simple depth and time coefficients. The critical pieces of information are time and time adjusted average depth. You need to know your time with precision and your average depth generally. Even saw tooth profiles are trivial with a little practice and situational awareness. The deco curve is far more sensitive to minutes of exposure than feet of exposure. To know the time you need to be conscious and not much more. Average depth requires a bit more awareness, but only a bit. Once you know that, and you observe some concepts about how the ascent curve is shaped, the rest is execution.
Repetative dives pose additional complications, to be sure. Here again, however, knowing something about physiology and decompression concepts is important. Conservatism in profile and gas selection, SIT, exertion, etc. is the key. A Doppler is really good, too. Rest assured any computer model dived to its limit repetatively will result in progressively higher bubble counts across dives. Physiologic response to repets is really hard to model because of the chaotic nature of the body's response to hyperbaria and the error terms compound. RGBM, designed to provide a big improvement in repetative deco safety, requires some major simplifications (e.g. perfusion/diffusion limit assumptions) in order to adapt to dive computer implementation. Even in its full up form, it's still only a "model." As with bounce dives, relying blindly on a computer which tells you you're OK to do that one last dive to 80 feet is playing Russian Roulette.
As you might have guessed, of all the reasons and psuedo reasons (things are not always better by the dozen) given against computers, the "idiot light" argument is to me the best.