It's no more of a fudge than Buhlmann-GF divers lowering there GF-Hi to be more conservative. From what I've read there isn't much evidence on how safe any algorithm is when used for multi-dives/day or multiple days in a row of diving. The fudge as you call it is simply a feature of RGBM's bubble model keeping very slow tissue's from feeding existing bubbles that could get close to the critical size on subsequent dives. It could also be said that RGBM recalibrates the algorithm for these multiple dives.
Hi EFX. I'm not so sure about that. It's completely clear to everyone how a Buhlmann algorithm works and what affect of gradient factors have on the dive. Buhlmann doesn't work any differently on a repetitive dive than it does on the first one. The principles are the same, the calculations are the same and the parameters are the same. The only thing that changes is the residual tissue loading at the beginning of the dive.
Being proprietary, the working of the RGBM algorithm has to be treated as a black box. Nobody really knows for sure what it's doing under the hood, or how.
I would be convinced that it does the right amount of "fudging" if in a comparative "black box" test the algorithm were run through a long series of various types of repetitive scenarios and the adjusted NDL's were sufficiently similar to the calculations Buhlmann gives. This might not prove that RGBM is right because taking Buhlmann for "correct" might not be right, but such a test would at least convince me that if something is broken that they are both similarly broken and therefore one is not better than the other in this context.
As it is, what we CAN see is that bubble models tend to overload slow tissues in favour of making fast tissues very clean during the ascent. This is an inherent characteristic of the algorithmic approach that is shared by all bubble models. We also know that in a situation where you are making decompression dives, that this overloading can already reach the point during a sufficiently extreme single dive where there is an increased incidence of DCS as a result of this algorithmic error.
We also know that repetitive diving is mostly a problem with respect to slow tissues. Years ago (I don't know how old you are so this could have been before your time) PADI adjusted the shallow NDL's, drastically reducing them, because divers were getting bent using the tables on a long series of shallow repetitive dives. The problem was the handling of slow tissues. Ergo, since we know that slow tissues are the issue with respect to repetitive dives, then I would instinctively distrust an algorithm that has an inherent tendency to overload slow tissues.
Anyway, that's my logic. I'm not on an anti RGBM crusade but I will admit that I distrust it due to the complete and persistent lack of scientific verification. All the science that HAS been done (and it's not a lot) seems to point to it being broken.
So linking this to my previousl point, I think companies like Mares must know this but it's hard to know what they have done to "fudge" the algorithm so that recreational divers remain safe on repetitive dives. Not knowing what they've done makes me personally uncomfortable although I actually do trust Mares to have the interests of divers at heart and therefore do believe that they have done something to address this. Mares hasn't said that they did, but I've used the Mares Puck a lot in the last year or two because of teaching and I strongly suspect that what I'm seeing it do isn't pure RGBM, despite the marketing claims.
I guess you can put it this way. If you wanted to buy a car and you had a choice of (a) a car for which the exact workings of the motor were fully documented, you had complete instructions about how to adjust it and this type of motor was scientifically proven to be reliable or (b) a car for which there was no documentation at all about the workings of the motor the hood was welded shut and there was no testing to know if it was reliable or not. .... which car would you be more inclined to buy?
R..