Which decompression algo does GUE use?

[PfcAJ]

I think rgbm is a proprietary code and hasn't been thoroughly tested or verified.
id always be wary of the codes creater claiming it to be super safe. If it was the case, everyone would jump the buhlmann/vpm ship and use it.

Buhlmann 20/85 works. Vpm +2works, but maybe not as well. Both are conservative enough so that most people won't get bent on most dives. Fiddle with the gradient factors or critical radii and things might get interesting for you. I would stick with the tried and true stuff before experimenting.

 

[decidedlyodd]

In 'Deco for Divers' they talk about how RGBM is proprietary. There are some early papers that discuss the principles of it, but the actual constants used are proprietary and the algorithm must be licensed by software developers or dive computer manufacturers. It's a bubble model like VPM, but nobody really can publicly say exactly what's in it due to the licensing. VPM, on the other hand, is based on a series of published papers. People tend to pay for the software because it's fairly complicated implementing decompression software from formulas and constants in a set of academic papers, but technically everything is open and available for anybody to use.

My impression is that all these algorithms have good safety records--RGBM, VPM, variants of Buhlmann with conservative adjustments like gradient factors and even Ratio Deco. I don't think there is any academic literature definitively showing that any of the specific algorithms above result in fewer cases of DCS or fewer deaths. These are all just approximations for a biological process whose behavior is well understood but not its underlying causes.

 

[rivers]

I agree. I also feel that taking a tool meant to be a handy heuristic (i.e. ratio deco) and using it to plan dives exclusively is...err....a bit worrying to me. Especially when it stops working at extremes of the range.

There seems to be an emphasis on simplification over precision in the UTD philosophy. I think that's great for things that don't really matter, but for calculating deco schedules and things like rock bottom, I'm not really comfortable with over simplification.

---------- Post added May 28th, 2014 at 09:09 AM ----------



Thanks for the references to the reading materials, it's just what I was looking for.

I'm still a bit confused as to how Baker's work relates to Buhlmann's work. Is Baker's work derived from Buhlmann?

I was introduced to the various gas models in UTD class. Buhlmann's model in particular was knocked on pretty often. E.g. we were taught that his assumptions that gasses stayed dissolved at depth was wrong, that he assumed similar rates for on-gassing and off-gassing, which was incorrect. So I am a bit surprised to learn that deco planner still refers to Buhlmann's model as a reference? I was given the impression it was obsolete and unsafe to follow, because it does not take into account getting rid of bubbles at depth before ascending.

From what little research I have done, I have read of VPM bending people also. It seems RGBM has 100k trials of real dives with no reported bends (according to the creator who posted on Scubaboard some time back), and the model can be safely extrapolated, within limits.

My knowledge in this area is very limited, but I am just curious why GUE has not adopted RGBM as the deco model of choice. Can anyone explain?

Pure Bulhmann (100/100 GF) is designed to get you shallow, quickly and keep you there for quite awhile. What gradient factors do is manipulate Buhlmann's model and insert stops. Where those stops occur and how long they are depend on what you've set your GF to, if that makes sense. Both Mark Powell's book and Erik Baker's paper go into detail as to what lo and hi GF mean, etc.
All gradient factors and VPM conservatism are is a manipulation of an existing deco model. It is important that you understand how you are manipulating it. Deco theory is just that, a theory. What works for you one day, may not work for you the next.

 

[WhiteSands]

I think rgbm is a proprietary code and hasn't been thoroughly tested or verified.
id always be wary of the codes creater claiming it to be super safe. If it was the case, everyone would jump the buhlmann/vpm ship and use it.

Buhlmann 20/85 works. Vpm +2works, but maybe not as well. Both are conservative enough so that most people won't get bent on most dives. Fiddle with the gradient factors or critical radii and things might get interesting for you. I would stick with the tried and true stuff before experimenting.

This was the thread I was referring to btw. Would love to hear your thoughts (any everyone else's).

http://www.scubaboard.com/forums/ask-dr-decompression/41678-rgbm-vs-vpm-differences.html

 

[rjack321]

This was the thread I was referring to btw. Would love to hear your thoughts (any everyone else's).

http://www.scubaboard.com/forums/ask-dr-decompression/41678-rgbm-vs-vpm-differences.html

1) GAP is stupid expensive
2) 100,00 incident free dives is complete hokum, well take that back its actually plausible because...
3) It'll take you forever to exit the water using RGBM

RBW's arrogance about the "superiority" of RGBM is disturbing, nobody really knows how individual physiology interacts with dissolved gasses. And people are most definitely different.

I use Ratio deco augmented with VPM. Bulmann requires such a low GF high to work well for me that I might as well skip using it. A GF high of 85 is waaaay to high for me after 3 back surgeries with various scar tissues in there and being 45yo.

If you haven't figured it out by now there is no one "official" GUE algorithm. Decoplanner (their software) includes Buhlman and VPM for comparison, but I don't know any GUE divers using that output directly.

 

[rongoodman]

As far as I know, the only dive computer which has RGBM available for it is the Liquivsion Xeo, marketed by a third party. I've never seen any comments by anyone actually using it. (I think there was a beta from GAP for the X1 at one time.)

 

[Lorenzoid]

As far as I know, the only dive computer which has RGBM available for it is the Liquivsion Xeo, marketed by a third party. I've never seen any comments by anyone actually using it. (I think there was a beta from GAP for the X1 at one time.)

Suunto uses some sort of proprietary implementation of RGBM. Has been characterized as RGBM with extra conservative margins added.

 

[rongoodman]

Since it's proprietary, nobody knows what it is, but all the speculation I've seen is that it's a Haldanean model with some RGBM style modifications, aka '"folded RGBM". It's not likely the cheaper recreational computers would have the computational horsepower to run the full model.

 

[Lorenzoid]

Since it's proprietary, nobody knows what it is, but all the speculation I've seen is that it's a Haldanean model with some RGBM style modifications, aka '"folded RGBM". It's not likely the cheaper recreational computers would have the computational horsepower to run the full model.

Interesting. Suunto sure has no problem referring to it as RGBM in their manuals. It sounds reasonable to me that modeling bubbles would take more "computational horsepower" than the way I understand Haldanean models work.

 

[boulderjohn]

Yes, Suunto uses the term RGBM and even invokes Weinke's name, but my understanding is that it is a limited version of the full program.