diving with two algorithms

[scubadada]

Just did some diving in Pramuka, since I came for a short weekend stop.

I dive with two DC, a Mares Punk Pro and a Suunto D4i ( christmas give ) did 4 dives, two each day, on my first day and my second dive the Punk Pro was very concervative as it kept me very tight on the NDL, even as I started ascending slowly and stay at a shallower depth to stay in my NDL but eventually I enter in Deco, while the Suunto, for my second dive it lock me in to do it with Nitrox, did not even show me the Air option so I had to change the EAN % to 21 in Nitrox mode, the Suunto never enter in Deco, aldo at max depth they were tracking the same on NDL, but as I ascended the Suunto begin to add more time to the NDL more and faster than the Punk Pro.

At the end I followed my Punk Pro Deco time and surfaced, second day the Suunto still don't even show me the Air option, lock on Nitrox, and there was about 13minutes difference on NDL in some point between both DC's

Honestly my Punk Pro is my choise and is my first DC, the other is may daily watch and was a present from Christmas, I start diving with both just in case one go south on me I still have another one to count on.

Did you think I could had the small or remote possiblity to got the niggles with the Suunto ???, taken that it is a very popular DC, and probably the most sold DC in the marked, is there a trend of who gets more bends or niggles diving with a particular brand DC taken different Algorithms ???

---------- Post added February 7th, 2015 at 10:39 PM ----------

To add a note, the DM was using a Mares Smart, but I never bother to ask him if he enter in Deco as well, I just follow my equipment.

Your post makes me wonder whether you know how to set your FO2 prior to your dives. Both of these computers run a conservative decompression algorithms. If you dive two computers with different algorithms, the more conservative will control your dive by limiting NDL or by requiring deco clearance to avoid violation

 

[guyharrisonphoto]

I will offer you a slightly different perspective, from a recreational diving standpoint.

I have tried to research the topic of algorithm safety extensively, including the internet, DAN data, and several threads on this Board, and, as of now, for the popular recreational computer brands, there is no objective, (i.e. observational or statistical-based) study to show that one computer algorithm is “safer” than another, or, put conversely, that one algorithm puts you “more” at risk for DCS for either single or repetitive dives. Thus, labels such as “more conservative” or “more liberal” are, in reality, meaningless as objective terms for measuring the risk of following one computer over another.

As TS&M noted in her post above, computer algorithms, collectively, have so greatly increased diving safety that there is very, very little incidence of DCS. This, in part, is the problem with comparing algorithms; there are just not enough cases of DCS to create a statistical basis to show that divers using a particular algorithm/brand of computer get DCS “more” or “less” than divers using another algorithm/brand.

So, using two different computers does not put you more at risk than using two of the same. You can follow either one. So long as one algorithm shows that you are still within NDL, you can choose to ignore the computer that says you are in deco, and there is no statistical or objective evidence to show you are more at risk of DCS.

Of course, if you “violate” the computer that is in deco, then that computer will lock out and not provide any information, and you will essentially be required to use the other for the duration of your repetitive dives. You cannot go back to using both computers when the violated computer clears (usually 24 hours or more), because it has not tracked and calculated, during the lock-out period, the dive time underwater that the other computer continues to track.Only when both computers clear completely can you resume using them simultaneously.

Thus, as a “practical” matter, if you are committed to diving two different computers for repetitive diving, you will ultimately be forced to follow the “more conservative” one (either by shorter NDL, or by incurring deco stops), because if you violate it, it locks and becomes useless as a second computer. This does not mean, however, that you are objectively safer from DCS than if you followed the more liberal computer.

So, for this very practical reason (and not a safety one), it is better to choose the algorithm you are more comfortable with, and use two of the same computers, because you will never have to worry about conflict.

From your posts, you seem to feel most comfortable following the more “conservative” computer and even adding yet more time, for yet more conservatism. But, you also like the option of seeing what a more liberal algorithm will give you, in case you want to walk on the wild side and give yourself a longer NDL. That is perfectly fine for a single dive. But, I don’t see how this could possibly work on repetitive dives, because if you follow the more liberal computer to its NDL limit on any given dive, you will violate the conservative one, and it will stop working as a deco computer for your next dive.

So, get two of the same.

As for what brand/algorithm to pick, on the theoretical level, deco algorithm modeling is based on the principle that shorter NDLs or longer safety (or even deco) stops create less risk of DCS, because you don’t on-gas as much (short NDL) or you have more time to off-gas (deco stops). Thus, from a theoretical modeling view, a more “conservative” algorithm is “safer.” On the other hand, other people use a more “liberal” algorithm to maximize their time underwater, being willing to accept a theoretically increased risk of DCS. You can choose your algorithm (and then the brand using that algorithm) accordingly, based on your psychological comfort level. However, there is no statistical evidence to show any differing degree of safety between Suunto’s proprietary bubble model, or the more common Buhlman (and its variants Z-8, Z-16 etc), or the D-Sat Pelagic, that are used in many of the most popular brands.

This, of course, only discusses computers and not things like age, weight, muscle mass, hydration, fatigue, exertion during the dive, or the many other factors that might influence how you use your computer, once you are more fully familiar with deco theory.

 

[Steve_C]

Very good post. However, I question one point

But, you also like the option of seeing what a more liberal algorithm will give you, in case you want to walk on the wild side and give yourself a longer NDL. That is perfectly fine for a single dive. But, I don’t see how this could possibly work on repetitive dives, because if you follow the more liberal computer to its NDL limit on any given dive, you will violate the conservative one, and it will stop working as a deco computer for your next dive.

Say I am at 80 ft and computer S reaches its NDL but computer A says I have another 10 minutes. If I reach the anchor at 0 NDL for S, S will go into deco because it gives credit for going up very slowly at first. Or perhaps I reach the anchor at -2 minutes for S. Now as I go up I make sure that I clear S so that it is happy back on the boat. Basically a 10 minute required stop and then the usual safety stop. On dive 2 I have both S and A happy and can go diving. Actually the point is not more NDL. By the time you get to the surface both computers are showing hours of NDL. The point is more bottom time. Also note that I am not talking about riding the NDL of A. We are riding the NDL, sometimes of S. Personally I often ride the NDL of S but I am always a few minutes inside the NDL of A. If I were to actually ride the NDL of A I would likely incur a more serious deco obligation on S and that I avoid unless an emergency caused by some one else arises. (Hypothetical - never done it or had to do it yet).

If this happens on Dive 2 on a 2 dive day, then there is no impact on dives the next day, if there are any. Second dives of the day with their shorter NDL at depth are times when I am most likely to want to use all the NDL of S at depth.

S= Zoop
A=Aeris

 

[Agility]

for an overwiew of diving algorithms see A sense of algorithm - Divernet

 

[Redshift]

I don't get some of the harsh replies given to the OP. Especially when after several posts those same people ended up saying "no-one knows".
Some of the OP's questions were very valid and can be confusing for some divers.
Thankfully TS&M and guyharrisonphoto tried to explain instead of bashing.

Dive computers can't track our inert gas loads, but they have algorithms that have been developed based on mathematical models, lab experiments and real dive studies. So while we know they are not really accurate, we expect them to be fairly consistent and safe. And looking at the DCS incidence for divers within NDL, the models are very safe. So why do two dive computers who are both using RGBM (although the details, at least for Suunto, are not publicly known) can show a 13min NDL difference? And if doing so, what does it mean when one computer's model says "you cannot ascend directly to the surface or you may get bent" and the other is saying "no problems, stay here for 13 min and enjoy"? How can the second be safe as well with such a large difference? These are legitimate questions and I don't see the need for the personal attacks in some of the posts here, especially when some of them were not even focusing on these questions but extrapolating to some other ideas of what the OP would be doing on his dives.

 

[Lorenzoid]

Yes, Redshift, they are good questions to ponder as an academic exercise from the safety of one's armchair, and at least a couple of the replies have helpfully focused on those questions. (I especially liked Guy's comprehensive explanation.) It's good to understand why one computer might show a deco obligation when another shows 13 mins. more no-deco time and that there is no one correct "bright line" to follow. Knowledge is good! While I believe that kind of understanding is good to have from an academic perspective, I also believe taking that bit of knowledge into the water can pave a path to trouble. What is the saying about "a little knowledge can be a dangerous thing" or something like that? In practice, a recreationally trained/equipped diver--especially a less experienced one--should pick ONE computer, and follow it religiously just as though there really were a bright line between getting bent and staying safe, even though the diver may know that is not really true. If a buddy team has different computers, follow the most conservative one.

To put the theoretical knowledge into practice, deciding for ourselves what kind of stops are or are not obligatory for us under the circumstances, we need to get the right training and equipment. It's all too easy to deal with with a couple of minutes of deco obligation because we know every computer is inherently just showing a "best guess." Next time, you figure you can handle three minutes. Then four .... Pretty soon, you don't have the skills and equipment (e.g., gas) to handle contingencies.

Playing with two computers near no-deco limits for comparison indeed could be fun--the data could be interesting, as in that link Agility posted above--but I wouldn't do that until I am sure I can get myself out no matter what unforeseen events might happen.

 

[guyharrisonphoto]

Hi Steve, you are correct in that if you clear "S" then you can dive both of them on the next dive, and that is what I meant when I said that you are essentially forced to follow the more conservative computer if you want to continue using both on the next dive. However, there is nothing unsound at all about you (or the OP) also carrying the more liberal computer, so that if you encounter a situation where you stay down longer than "S" would allow, you still have a statistically safe alternative for how much more time you can spend on the bottom without deco (if you have to deal with an emergency at depth and then ascent directly without stops, for example).

Redshift and Lorenzoid, the whole concept of "how can they both be safe?" is what me started on my research, and my surprising conclusion that, as of now, the evidence shows they all are safe. This results in some interesting conclusions. For example, the old saw about "always dive the more conservative computer in a buddy team" has no objective support that this a safer approach. In fact, when a buddy team uses different computers, they are just as safe (statistically) using the more liberal one. The true danger is in relying on a computer that is not on your own wrist and so not tracking your precise dive parameters (depths, ascent or descent rates, gas mix). I do agree that you should always follow your own computers, not someone else's. Thus, as a practical matter, in a buddy team the most conservative computer "wins" not because it is "safer," but only because it triggers the first ascent for the individual member of the team wearing that computer, which all buddies (if they are a true team) then follow so no one is left alone.

For your personal computer(s), there is no objective evidence to support using anything but the most liberal algorithm. Although, different algorithms can behave very differently at different depths or with aggressive repetitive dives (think liveaboard trips) and a "liberal" algorithm can become quite conservative over repetitive dives, where other algorithms (Buhlman variants) do not change so much.

Because of that, I picked a moderate algorithm that does not seem to vary much from traditional tables over multiple dives (Buhlman) and make no adjustment for additional conservatism in my rec diving.

I am also glad that my comments were useful. This is, absolutely, an area where additional study and some uniformity among algorithms would be useful, and, also, all manufacturers should disclose the basis for their algorithms and how their conservatism adjustments operate so that the divers know exactly what calculations are being applied to them.

Still, they are all basically safe, so these would be nice but not essential.

 

[Nasser]

Forget about putting one computer into gauge mode - this is not practical for doing several multilevel NDL dives over several days. It's a good option if you're already cutting your own tables and following run times but a pain in the butt if you're not diving that way to start with.

Pick your favorite DC out of the two you already have and sell the other. Then buy the same computer as the one you like as a backup. You want to enjoy your dives and not spend time worrying about which computer is being more conservative than the other... or if one is going to put you into deco sooner... or if one is going to lock you out... etc..

Keep it simple and use two similar computers with the same algorithm and be done with it.

 

[Andrew Chunis]

It is better to always have a backup

 

[tech_diver]

I think that most OW instructors do teach the concept of the 'grey zone' of deco safety. Two computers might put up different boundaries within that same grey safe zone. If the diver understands that, he could apply the context of the current dive (cold, rough, warm, mellow) and decide where in that grey zone he wants to be. But I can understand how such a large 'grey zone' might alarm a new diver.

Tables won't address the concern. If you have two different tables, they can produce different results. The PADI table say you can do 55 minutes at 60 feet and Navy tables say 60 minutes at 60 feet. Over a series of repet dives the difference do become more pronounced but not as pronounced as differences seen in a series generated by multilevel computer profiles.

Look at all your instruments and understand where the numbers come from then consider the context of the dive.


There is an ancient Chinese saying which tells us, "A man with two deco computers never knows his real deco status." The truth is that even with one computer you really don't know and have to make a judgement call based on all the knowledge and info you have.