Understanding Algorithms

[LavaSurfer]

Looking for a place to get more info on the different Algorithms that are used in computers. Any Ideas where to start?

This curiosity comes from my experience with several computers I own.
I have an Oceanic ATOM 2 and VT Pro and also dive with a Diverite Nitek He

I have notices several times a massive difference between the oceanic algorithm and the Nitek. My latest experience was diving with several friends in Roatan on a nice deep dive to Hole in the Wall at 130 FSW. The agreement was at 2 min NDL time remaining we would move up the wall to the reef and hang out around 50 feet then 30 feet. After several minutes I noticed I was at 3 NDL and the other two were still showing 6 on there Oceanics and Sunto. I was pretty much happy with the idea of letting mine go into deco mode as I had plenty of gas and experience tells me that the Nitek does well in DECO mode. as we started up I went into deco mode and knowing the profile I did a 1 min stop at 70 feet then another at 60 and rested on the reef around 50 feet for some time looking at sponges. we then proceeded to 30 feet where I made several other stops to satisfy the profile and after a few minutes at 30 feet the Nitek cleared from DECO mode and was happy. I still completed a 3 min 20 foot and 3 minute 10 foot stop for my own personal satisfaction and played with a Cuda and grouper that were watching us.

What bothered me is I know we really pushed the limits and neither of the other chaps went into DECO mode and had I not insisted on some deep stops they would have skipped them and went right to a 3 minute safety stop.

I calculated the dive in vplanner and the profile required several stops. One at 70 for (1) minute and one at 40 (1), 20 (2) and 10 (3). These liberal algorithms skipped these all together. Course the time at 30 feet and 50 feet excluded the need for deco as off gassing started at 89 feet.

I love diving the Nitek because it is conservative and when it goes into DECO mode, it knows what to do. How well do other computers handle DECO? I hear the Oceanic / Aeris computers revert to navy tables, is this true? I would never trust my ATOM in Deco mode because its way to liberal anyway. Any thoughts?

 

[Meng_Tze]

Look at the manuals for the computers and go from there. Be careful not to compare apples to cars.........

 

[Jim Lapenta]

IF you really want to get into different comp algorithms come on over to The Deco Stop and post in one of the comp or deco forums. You'll get alot of info with little BS.

 

[divenutny]

LavaSurfer - You stepped into rather difficult topic/area. I posed a similar question to the Dive Rite quarom and I received more rhetoric that assistance. As stated above by Meng_Tzes, look to your manuals for the algorithms each computer uses. The Dive Rite follows Bulhmann, I think the Oceanics follow Modified Haldanean. These tables/calculations are different, giving different dive profiles. I know the Nitek is conservative, I have the NiTek Duo. I understand the Oceanic is liberal compared to other Modified Haldanean. The consensus I got from my thread was nether is calculation is right or wrong, the calculations are different. You need to make a personal choice to follow one or the other.

One thing I am surprised about is the first stop at 70'. I know V Planner will look for this. I understand that the NiTek may not calculate this stop as off gasing and extend your other stops. You may have a lot of research ahead of you to get the answers you want.

Good Luck - Tom

 

[CompuDude]

... and if you take the convo to DS, do us a favor and post the URL here so we can follow it. I'm curious to read the answer from those who have real knowledge of the algorithms. (I'd also be curious to see how Uwatec's algo's read into the mix)

 

[LavaSurfer]

One thing I am surprised about is the first stop at 70'. I know V Planner will look for this. I understand that the NiTek may not calculate this stop as off gasing and extend your other stops. You may have a lot of research ahead of you to get the answers you want.

Good Luck - Tom

The Nitek did not calculate the 70 foot stop. I calculated, in vplanner, a 130 dive previously and knew the stops needed and just followed them. By the time I was ready to do stops required by the nitek, the Nitek had cleared and already started off gassing. I had something like 200 min NDL when I hit 30 feet. Being the good boy I am and conservative, I chose to continue following the vplanner stops that I had calc'd earlier.

 

[LavaSurfer]

Look at the manuals for the computers and go from there. Be careful not to compare apples to cars.........

Ya, I get you there. I knew the models the Oceanic and Nitek use I am just amazed by the difference. The part that makes me curious is although neither is wrong nor right, there are probably applications where one is better than the other. I have been told that the Bulhmann algorithm is better for DECO diving but thats something I heard not something I know. I know vplanner and the vr3 use a vpm model which is also different.

I was also surprised when someone told me the Oceanic ion DECO mode reverts to Navy Tables. I wonder if that is true. I will have to research that. If so, how does that apply after SI.

 

[battles2a5]

Here's an awesome explanation of VPM (and its permutations). VPM picks up where Buhlman left off (one of the more refined tissue-models). V-Planner uses VPM-B so if you plan of using it, this is a good one to read.

VPM For Dummies - Rebreather World Forums

The encyclopedia of diving has a decent section about the development of the Haldanean models. If you go the deco stop, you'll see a good amount of info in the deco models section.

 

[Meng_Tze]

Ya, I get you there. I knew the models the Oceanic and Nitek use I am just amazed by the difference. The part that makes me curious is although neither is wrong nor right, there are probably applications where one is better than the other. I have been told that the Bulhmann algorithm is better for DECO diving but thats something I heard not something I know. I know vplanner and the vr3 use a vpm model which is also different.

I was also surprised when someone told me the Oceanic ion DECO mode reverts to Navy Tables. I wonder if that is true. I will have to research that. If so, how does that apply after SI.

You have micro bubble algorithms, dissolved gas models, gradient factors rolled over and some proprietary, even empirical models. Comparing stop times in one model to a different one only tells you the difference. Understanding where these differences come from is a completely different issue. It all depends on the approach the algorithm takes towards on- and offgassing.

I would not agree that Buhlman is better for DECO, nor VPM, Haldane, RGBM, or any other model is better for deco in general. It is all about what you as a diver feel comfortable in diving with given a certain profile.

You will not find any absolute answers on this one except that there is no ONE solution.

 

[LavaSurfer]

Here's an awesome explanation of VPM (and its permutations). VPM picks up where Buhlman left off (one of the more refined tissue-models). V-Planner uses VPM-B so if you plan of using it, this is a good one to read.

VPM For Dummies - Rebreather World Forums

The encyclopedia of diving has a decent section about the development of the Haldanean models. If you go the deco stop, you'll see a good amount of info in the deco models section.

Thats awesome, thanks.
I have it printing and will read it.