But Ron, it wouldn't have been true. In fact, for many users, "full RGBM" wouldn't really be true either, in as much as they'd never invoke it's use :mooner:
Sorry if I wan't clear. By that I meant that it would have been easier, from a design standpoint as well as potentially better for marketing, to use full RGBM at all depths. In fact, we did so in prototypes back as long ago as 2003. BW said we should not use this approach- and he provided the algorithm specifications for Atomic. We are not the designers of the algorithm, so I can't provide as much detail as you would like about why this approach was specified. We are, however, responsible for implementing it accurately in a real time system. I know there has been considerable verification and testing that has led to changes in RGBM conservatism and settings over the years, these are reflected in the Cobalt implementation. As it stands, we don't mention "full" or "dual phase" in marketing descriptions. As you observe, most users will not use this recreational computer in the range where full RGBM would kick in. Perhaps at some point there will be a full RGBM version that BW recommends for use at shallower depths- but if there is, I don't think divers in the recreational zone will notice any difference, mapping settings to the same statistical risk levels would yield essentially the same profiles and NDL's we have now.
As someone who has been using a dual phase algo exclusively for all dives, NDL or deco, shallow or deep, long or short, successfully for the last 3 years, I'd have to say it scales well for any sort of diving.
As tissue saturation is a part of a dual phase model, I'm surprised to hear that the older, tried & true models are considered to do this any better than the RGBM model.
What algorithm have you been using? I assume this is in desktop dive planning software? It would be quite easy to compare the Cobalt's output to desktop software, including for repetitive dive series, since the Cobalt includes a "desktop" style graphic planner on the computer itself.
How well full RGBM scales and at what statistical risk levels compared to a folded approach would be a question for BW. I said "more efficiently" rather than better, and by that I'm referring to computational efficiency in the context of a dive computer. Fully iterative RGBM is very computationally intensive. It's one thing to run an iterative algorithm like RGBM in desktop dive planning software, quite another to implement it in a real time system with far less processing power, concerns about battery life, and the need to account for divers doing unexpected things, while recomputing continually on the fly. Desktop software needs to only run the algorithm once per profile, a dive computer has to do it repeatedly and provide quick updates to the diver. In that case, you want the most efficient approach that yields correct results. For deep diving, full, iterative, dual phase RGBM has some very real advantages. That does not necessarily mean it has an advantage in the shallower zone.
