Charlie99
Contributor
- Messages
- 7,966
- Reaction score
- 169
- # of dives
- 500 - 999
The short answer: It doesn't matter much to the typical rec diver.JeffMandell once bubbled...
I'm a little confused by this thread (and some others like it). It seems that the underlying premise to the RGBM computer preference is that other computers are not conservative enough and are, therefore, risky.
I'd like to know if this really makes a difference to recreational divers who rarely stray beyond 100 feet, or is this needed "conservatism" really only an issue in tech dives? I'm not aware of any non-anecdotal data suggesting that non-RGBM computers are causing increased DCS in the casual diving range I'm asking about.
Longer Answer: If you have a poor model, then you need to add extra fudge factors for all profiles in order to keep all profiles safe.
Theoretically, diving a good algorithm/model to its limits would alway subject you to the same DCS risk, independent of which profile gets you to the limit.
A better algorithm that more closely models reality can also lead to improved profiles (deep stops, better allocation of time to the various stops, etc).
We could easily dive with poor quality depth gauges that had random errors of +/-10' ---- we'd just add in a bit of fudge factor to compensate. The same with decompression models. A better model means that your fudge factors can be lower. However, even the BEST decompression algorithms have huge uncertainties since there is a large variation in susceptibility to DCS from diver-to-diver, and also from day-to-day with the same diver. RGBM could be reducing DCS rates by a factor or 2 or 4 and we wouldn't have enough data to be able to really know that.
As a single tank diver, if you want to increase your safety, then avoid yo-yo profiles, have reasonably long total ascent times, do a couple minutes of deep stops, and add a couple minutes of hang at 10-15'. These actions make more sense than diving ANY model right up to the limit. With a single AL80, diving not much past NDL, the algorithm choice doesn't make that much difference.