dive computers and reverse profiles

[stepfen]

Hi,
I finished yesterday SSI's "Science of Diving" online course (which BTW is offered for free these days) and in it in the section describing decompression (Section 3) I fount the following remarks about dive computers:
(bold and red are used by me to highlight)

Section 3/Dive computers:
"...
Dive computers do have several limitations:
...
- The mathematical model works only with the correct dive profile, which is a multi-level profile with the deepest depth first, followed by subsequently shallower depths. "

And a couple of paragraphs later:
" Going deep, then shallow, then deep again will yield unreliable results. Since a dive computer is an actual computer, an old axiom in the computer business is applicable here: “garbage in equals garbage out.” In other words, the computer model must be supplied with the correct input to calculate decompression status properly. Since the computer gathers its input using a depth gauge and a watch, if a diver dives improperly the computer will calculate improperly (see Multi-Level Diving). This illustrates the third limitation, which is training. "

Then again,
Section 3/Different Dive Profiles That Impact Decompression Theory:
"... Dive computers were designed for multi-level diving and are an excellent tool if they are used properly and conservatively. Use the computer with deep-to-shallow profiles, not deep-shallow-deep profiles. The computer continuously computes nitrogen absorption based on the actual depth and time actually spent there.

Therefore, if divers go deep then shallow, their nitrogen absorption is computed at a slower rate at the shallow depth. However, on descending again to greater depth, the theoretical nitrogen absorption is increased accordingly. As indicated earlier, the residual nitrogen affects that process, yet the computer does not accurately account for this affect. Therefore, the computer is not calculating absorption the same way your body is, which is potentially hazardous. "

So SSI tells us that computers (or to be more precise their algorithm(s) ) cannot accurately account for residual nitrogen !!!???

That's something totally new to me. And makes me wonder, how on earth can then computers safely be used eg for consecutive dives ?? Even 2 square profile dives with a typical surface interval (say 1-2 hours) break the rule "not to be used for deep-shallow-deep profiles" (in this case shallow being the surface).

BTW I am very well aware of the fact that all algorithms are theoretical models that just try to estimate (not always successfully) what is going on in our bodies. I thought though that they can track theoretical nitrogen levels pretty accurately (but of course not 100% compared to reality) irrespective of the profile.

Any comments?? Am I missing something?

Thanks

 

[tbone1004]

I can say so many things about this, but basically ignore everything you quoted. It's all wrong.

In computers running Buhlmann, i.e. basically everything except Suunto. The computer knows exactly what you're doing and really couldn't care less about what your profile looks like. The algorithm plots your current experienced pressure *depth*, and plots that against a theoretical tissue loading. It either adds or subtracts some value to each tissue compartment.
BUT! some computers have idiotic safety factors that they won't tell you about, so some of them may penalize you for doing reverse or sawtooth profiles.

@Lake Hickory Scuba tell your people that this book needs to be rewritten because whoever wrote it was CLEARLY drunk as a skunk to write blatantly wrong information, but even worse, wrong information that contradicts other wrong information...

 

[loosenit2]

I finished the same course a couple of days ago and like you I found that section tethered to old ideas. The course was really pushing the old school methodology of dive deep first and then get progressively more shallow in subsequent dives. I think that idea, which was the prevalent thought for a lot of years, has been shown to be inaccurate by more recent research.

 

[tbone1004]

I finished the same course a couple of days ago and like you I found that section tethered to old ideas. The course was really pushing the old school methodology of dive deep first and then get progressively more shallow in subsequent dives. I think that idea, which was the prevalent thought for a lot of years, has been shown to be inaccurate by more recent research.

That idea was based on how the dive tables worked. If you look at the USN/NAUI tables, then it makes a lot of sense for why you won't want to dive reverse profiles. It was a limitation of how they were made more than anything. Unfortunately ignorance is very prevalent in our industry which leads to all sorts of nonsense like what was quoted in that manual from a leading training agency that has absolutely 0 founding in science.

 

[stepfen]

I suspected that the info given might be old/wrong but in another section of the same manual some innovative(-ish) ideas about dive computers and algorithms are (briefly) mentioned. Things like computers that account for heart rate (metabolism?), temperature and other sensors. I know that all these are still ideas or at best ongoing experiments and nothing related has been proven reliable enough to be usable yet, but based on these I though this manual was somewhat up to date.
Thanks anyway

 

[tbone1004]

I suspected that the info given might be old/wrong but in another section of the same manual some innovative ideas about dive computers and algorithms are (briefly) mentioned. Things like computers that account for heart rate (metabolism?), temperature and other sensors. I know that all these are still ideas or at best ongoing experiments and nothing related has been proven reliable enough to be usable yet, but based on these I though this manual was somewhat up to date.
Thanks anyway

I am not aware of any actual research being done to validate anything to do with heart rate. Plenty have been done on temperature, but skin temperature is not really useful to measure and was not the focus of those studies, nor has anything been determined other than "hot" means you ongas and offgas faster, and "cold" means you ongas and offgas slower. I.e. you want to be cool at the bottom, and warm on deco. Any mention of those two that make it sound like those variables are useful, ESPECIALLY when the manufacturers don't actually tell you how they are used to adjust the algorithm, needs to stay out of any sort of instructional materials. Any hypothetical nonsense like that *note, heart rate and skin temperature affecting decompression is not theoretical, it is hypothetical*, really has no place being put into print. Instructional manuals need to be printed with current state of the art, factual information only.

 

[dmaziuk]

I suspected that the info given might be old/wrong but in another section of the same manual some innovative(-ish) ideas about dive computers and algorithms are (briefly) mentioned. Things like computers that account for heart rate (metabolism?), temperature and other sensors.

-ish is the keyword here: Uwatec's "ADT" is at least two decades old by now.

 

[tursiops]

Reverse profiles used to be bad, because it was thought the tables couldn't track them and no validation had been done on the use of tables with them.
Then this report came out twenty years ago:
http://archive.rubicon-foundation.o...6789/4244/AAUS_ReverseProfiles.pdf?sequence=1
The findings and conclusions were:

The reason for the "depth differentials" caveat was they could not find much data outside that interval to validate any conclusion.

 

[dmaziuk]

In computers running Buhlmann, i.e. basically everything except Suunto. The computer knows exactly what you're doing and really couldn't care less about what your profile looks like. The algorithm plots your current experienced pressure *depth*, and plots that against a theoretical tissue loading. It either adds or subtracts some value to each tissue compartment.
BUT! some computers have idiotic safety factors that they won't tell you about, so some of them may penalize you for doing reverse or sawtooth profiles.

VPM (Shearwater, OSTC) (optional): deep-first may give you longer no-stop time or shorter deco.
RGBM: penalizes for "bad profiles" (all kinds). Apparently worse in Mares, Cressi, and low-end Suuntos running the "folded", aka dumbed-down version of the model.
Pelagic "Z+": unknown.
Buhlmann: not part of the model.
DSAT: apparently not part of the model.

Nitpick: note that all of them track exactly what you're doing and calculate tissue loading based on current pressure and time. The difference is really only how/where they draw the "safe" line on the way up.

 

[tbone1004]

@dmaziuk
VPM-disproven by current research, don't use it
RGBM-disproven by current research, don't use it
Z+ and DSAT- who the hell knows since no one will release any information
Buhlmann-currently the "least bad" model to work with

Agreed on the last point, they definitely all track accurate, but it's what they do with the information. Computers running unadulterated Buhlmann just track it and continue on with their day. I have a lot of issues discussing the validity of computers that have all of these idiotic "Safety factors" built in that is not based on any sort of research, heaven forbid when it is compounded by their refusal to disclose what it is they're doing to try to "keep you safe". The point remains that NONE of that should be discussed in the academic information provided by a training agency though with SSI and Mares being buddy buddy, it's unsurprising they're trying to bring some of that BS into their training materials.