conservative versus liberal algorithms

[diatom]

As a newbie, I have been reading quite a bit on the board regarding computers/ tables, and which should be used. I have also seen discussion of computers as being more or less conservative depending on the algorithm used. My understanding of computers is that they take into account changes of depth during the dive, as opposed to assuming a square profile, when calculating ndl's. My question is how conservative or liberal are the different algorithms compared to the tables/wheel?

In other words, if you took the algorithm used to generate the tables, and modified it to account for multilevel diving, would the ndl's be more or less conservative than (or identical to) those generated by the various algorithms used in computers?

It seems to me that computers essentially take similar information that is used to generate the tables and integrate it over changes in depth. If that is true, then diving a computer is essentially the same as diving the tables. The only difference would be the algorithm used. So, again if you accounted for multilevel diving using the algorithms used in generating the tables and wheels, would the ndl's be more or less conservative than the various ndl's generated by computer algorithms?

I hope that question makes sense, and feel free to let me know if I am thinking about this totally wrong.

 

[vicky]

I think that most of them were based on Haldane or Buhlman models, other modified according to some observations of bubbles (detected with doppler ultrasound) and so on.
You can do your own tables based on whatever model you want- just make a simulation with a square profile, set the dive time and time between dives and you can print for yourself any table (if you trust yourself that you are sure what you are doing )

Tables are a very particular case of what a computer does- because the time between dives is divided into rather big groups, and depths are in 10feet jumps, ascent rate is fixed etc. etc. IMHO a computer is more accurate and also calculates more accurate data, so for recreatinal divers who can change their diving profile according to reef structure, or because there is a whale shark just a few meters above you etc- better to use a computer. Modern computers incorporate better models than standard tables (VPM, Abyss, RGBM etc.). By the way, there are some softwares that help you print your own personalized tables based on these algorithms, but again- a computer will tolerate any changes in your diving profile, a table wont.

 

[zboss]

Myself, my sister, and my brother just back from a dive trip to the BVI's in which I was diving with an Aeris Atmos 2, my sister was diving an Oceanic Versa, and my brother was diving guages. In every instance we used dives tables from YMCA and the dive times / surface interval times from the computer. In every instance we all descended at the same exact time and ascended within a minute or two of each other and all dove extremely close profiles. In every instance we all started our dives with almost exactly 2800 PSI and ended between 500-1000 PSI.

In every instance - diving times based on the tables (even repetitive dives) resulted in my dive computer leaving one green bar segment available and my sisters as having gone to the end of the green segments (the Versa has four segments and the Atmnos 2 has five).

In reality - I think we found the best dive computer value vrs. tables EXCEPT in the instance where we dove to depths greater than 60 feet. In that instance I think the time spent getting down to 60 feet had a strong negative impact on available dive time based on the tables, while the computers were able to better calculate the available time. In all instances having planned the dives on tables, and dove the plan on tables, gave me a better appreciation for the reality of using computers vrs. tables, this reality being that above 60 feet your dive time is limited by available gas (mostly), and below, the tables infringe upon available dive time. However - incorporating both technologies into our diving resulted in an extremely favorable diving plan.

 

[Charlie99]

diatom once bubbled...
My understanding of computers is that they take into account changes of depth during the dive, as opposed to assuming a square profile, when calculating ndl's. My question is how conservative or liberal are the different algorithms compared to the tables/wheel?

Oceanic and other computers made by Pelagic are a special case in that they use the same DSAT model used for the PADI RDP. If you are diving square profiles, then the computer and the tables have identical calculations (except for minor variances in ascent/descent times), and if your are approaching NDL on the table, your computer will also show that you are approaching NDL. Other computers, such as Suunto, has more conservative (shorter NDLs).

If you are diving multi-level profiles, but staying within the limits of simple (not multilevel table or wheel) table calculations, then you have built in a large amount of conservatism compared to a cmputer.

For example, if you do a 30 minute (excluding safety stop time) dive with the max depth of 71', but most of the dive actually at 45', you would be right at the 80'/30 minute NDL limit of the PADI RDP table. The computer, even if using the same deco model as the table, will show lots of NDL time because it doesn't round up to 80', and it takes into account that most of the dive was shallower than 71'. If you did a detailed running calculation of the dive with the PADI wheel, the answer would be very close to the computer's calculation.

Why many divers use computers is that relatively mild profiles, such as 75'/10 minutes + 55'/10 minutes + 35'/20 minutes is beyond table limits (max depth 75' requires the use of the 80' column, dive time before ascent of 40 minutes exceeds 30 minute NDL for even a 1st dive of the day). In this case, the PADI RDP is unreasonably conservative --- the result of treating the entire dive as if it were at the deepest point.

While computers and simple tables are not the only alternatives, they are by far the most common. Other alternatives include 1) using a deco program such as GAP, Decoplanner, etc to plan the dive 2) preplanning the multilevel depths and times by a multilevel planning device such as the PADI wheel (I use the RDP in a similar fashion, although that method is no longer taught), and 3) mental calculation of NDL on-the-fly during the dive (you will have to ask others about this method, I have not seen a clear description of this method).

I hope this is clear,

Charlie

 

[Braunbehrens]

Rodale's has a nice chart of various computers and how conservative - liberal they are.

 

[diatom]

It sounds like my thoughts on computers/tables are at least in the ballpark. From what you said, zboss, the algorithms for the tables and the Versa and Atmos 2 sound as though they are in fact the same (or very close to it). So, you were essentially diving with a "fluid version" of the same dive table.

If I'm not mistaken, those 2 computers are considered liberal in their ndl's, while some of the computers using RGBM et al. are considered moderate to conservative. The most important question in my mind is, are the liberal algorithms actually conservative. i.e. Is there a sufficient cushion between their ndl's and the actual times until dcs becomes highly probable. I have read at least one opinion that the more liberal computers are frighteningly liberal when compared to other computers. In order to determine whether or not they are frighteningly liberal, one would have to compare the ndl's from the computers to actual data collected on instances of dcs/tissue loading rates. It seems to me that providing comparisons of the ndl's for the various algorithms to such data would be extremely useful. Then you could make a good determination on whether or not a computer is truly liberal, or simply liberal relative to the other computers. Does anyone have any idea as to where you could find that info? That is assuming it is publicly available.

 

[zboss]

your thinking too hard and that will cause your brain to burst. I used to worry about liberal vs. conservative but it all comes down to dive planning. If you plan your dive and think very carefully about what your dive should turn out like and then you stick to that plan then everything should be fine. Pick a dive gas rule, share it with your buddies, and then use the same dive rule throughout your diving vacation (unless the DM dictates otherwise).

The basics are this: assuming your first dive of the day is above 50 feet and you are diving aluminum 80's and are a moderate breather - you will run out of gas before you run out of NDL - computer or not.

Between 50-70 it is likely that your NDL will match or come close to your available gas down to 500 psi - computer or not.

Below 70 (diving tables) it becomes increasingly likely that you will run out a gas before running out of NDL if diving a computer but not necessarily if diving a table.

On subsequent dives it becomes increasingly likely you will run out of NDL before you run out of gas - computer or not.

In fact - I am becoming increasingly convinced that computers - while they do improve bottom time - are not clearly worthwhile except to use as accurate timers or where you are doing multiple sucessive dives each day.

 

[diatom]

I hope my brain doesn't burst. I could use it a little while yet. I guess I probably am thinking about it too much, but I just thought it would be useful to know how conservative the most liberal computers are.

At any rate, all your posts were helpful. I appreciate it.

 

[Braunbehrens]

"Liberal" and "conservative" only refer to the NDL times. There are two problems with this. One, is that there is no "no decompression limit". All dives require decompression. Once you understand that, then you arrive at the second problem. Computers don't help you deco in the proper way. If you take your computer down to 120 fsw on air, right to the limit, it will ask you to ascend to 15 ft and then stop for 3 mn. My contention is that it would be much safer to do 1 mn stops starting at 80 fsw or so and then spend a couple minutes at 20, a couple minutes at 10 and then do a really slow ascent from there.

I'm just pulling these numbers out of my hat, but you get the idea. If you try to do this with a computer, it will start yelling at you to "hurry up and go to 15 fsw". No thanks. That is not conservative, or liberal, it's stupid.

 

[diverbrian]

Again, as in many of the threads I have observed, the topic of conversation appears to be drifting.

The last post is not about tables or computers, but about diving theory. I have read the comment about NDL on other sites. But, my tables are written to accomodate a 30 ft/min ascent rate with a 3-5 minute (preferably 5) safety stop at 15-20 ft. It took my Advanced Nitrox instructor a little bit to explain to me that 30 ft/min does not mean 5-10 ft per minute (about what I was trying to do and still try to do as a recreational assistant instructor) or 50-60 ft per minute (for obvious reasons). It means 30 ft/min. That was by diving tables. By comparing to the Deco that my computer had to do, you could see where his point was. The longer that I stayed at depth below my deco stop, the more nitrogen that I was taking in (by table and computer models), and therefore the more deco time I was actually occuring.

All NDL means is that if a diver gets into serious trouble, he/she can go to the surface at a reasonable rate as a last resort with a very small chance of getting bent instead of a high liklihood.

BTW, one of the latest articles that I read in a diving magazine basically agreed with you. I will bet in a couple of years, it will be protocol to do a safety stop at the halfway point of your ascent below 40 ft to further reduce the micro-bubbles. Remember, the safety stop that we take forgranted now is a relatively new concept to mainstream diving. It used to be considered just a good idea.

Let me ask this of the basic question, is the function of the computer strictly to give you more bottom time or as an aid to allow you to dive and stay out of a recompression chamber? I appreciate by conservative RGBM algorithm. There is a lot of water out there. I plan to dive to see it for many years. If the computer is more conservative, maybe I am not pushing my limits quite so far as decomression theory is just that, THEORY and hypothetical limits. Yes, I engage in some riskier activity that my original agency doesn't endorse (diving requiring deco stop vice staying within NDL limits to be precise). I had to go to another agency to get my training in that. No, I have no intention of visiting the recompression chamber if I can help it. I will have other dives to see what I wish to see, if my tables or computer give me a couple of minutes or a couple of feet of safety margin, that may be what keeps me out of trouble. So obviously, I am in favor of the slightly more conservative computers, just as I am sure that the divers advocating a 3-5 minute "safety stop" were thought of in the early days of diving as being a "little conservative."

This is just my opinion here, but maybe it will spark a little more debate.