NEDU Study

[boulderjohn]

When this thread started, a lot of posts were of the "Oh, no! Not again!" variety, but repeated threads like this have a real value.

The PADI trimix curriculum, which like most curricula is several years old, has two features that are in play here.

1. It was written during the heyday of deep stops, and it requires instruction in deep stops. Students are required to plan and execute some (not all) of their dives using deep stops.
2. It has a well-written section on the need to keep track of the latest developments in dive theory and practice and then make good judgments about making appropriate changes in your diving.

Last year I wrote to PADI about a potential conflict--what if the instructor's research on the latest thinking on deep stops leads him or her to decide not to use them? I got a reply almost immediately telling me that, yes, they knew what was going on with current thinking, and, yes, as an instructor I was serving my students best by teaching them about these issues and coming to a team decision about the issue. I was not bound by the deep stops requirement, and I could safely disregard those standards. A few months later, in the last training bulletin issued to instructors, PADI formally announced that trimix instructors were no longer required to include deep stops in dive planning.

This is a further indication of the way the research winds are blowing--PADI is usually very cautious about making changes like this. It also shows the problem with making such changes. Having had many conversations with technical divers about this over the last year or two, I know that many and perhaps most of them are blissfully unaware of any changes in thinking over the past few years. I would say most of the ones with whom I have spoken believe we are still in the heyday of deep stop thinking. They do not know these changes are raging, just as the OP in this thread thought he was starting a new discussion on a new issue.

Most divers are not active on social media sites like this one, and most of them do not go out of their way to find out what the latest thinking is on any issue related to dive theory. The more threads we have like this one, the more people will learn about controversies like this and rethink their diving plans.

 

[rossh]

Ross,
I tend to refrain from entering these threads, tends to be futile, but even if I am not a physiologist nor a decompression scholar, I understand math and physics.

GF=100% when refered to a tissue or a diver means that that person has one tissue or the leading tissue at the maximum allowable supersaturation.
Therefore surfacig with GF>100 is not good.

Hi,

Well.... no, that is not a hard rule. You are implying that ZHL limits are set in stone, and absolute, and the max possible, which is just not so. The USN tables will readily exceed ZHL limits.

The other part of this problem, is people here trying to use GF as a hard standard. But its not a consistent measure. What one does with the Lo value, changes how the how the Hi value reads - that is how GF is implimented onto ZHL - a little bit of the tail wags the dog.

Also when we discuss efficiency, is how much bang for your buck. So how good is your decompression strategy compared to your cost (in water time).

Yes... efficiency. In decompression terms "efficiency" has a specific meaning. The least amount of time to achieve an error free result. Now none of the models we do (ZHL, VPM-B, etc), are as strictly "efficient". The navy tables are often more efficient. The objective of general use models is for a more gentle procedure - a balance.

This said, to see which strategy is more effiicent you decompress with equal run times. The profile with the lesser GF wins. Yours loose. But that is only on the clculations that you keep pushing to us as if the models were reality.

That does not test efficiency, because you don't know where the actual error limit exists - neither encounter an error. So from an efficiency perspective, its a draw.




Also as I said before... if one wants to test a model - don't mess with it.... One cannot tell a model how to do its job, by forcing it into some artificial time constraints.

What the NEDU study did, was actually determine pDCS with a deeper and a shallower decompression strategy. The deeper strategy lost.

They did not test efficiency either. The profiles were deliberately made slow and long and easy. They then restored stress with excessive thermal stresses. It was a test of shallow profiles against thermal stresses, but that causes more problems with validation - how to quantify the right amount of thermal stress, when thermal and profile stress work in opposing directions.


Anyway... the nedu test says nothing about real tech profiles, and merely confirms what we already knew - gas kinetics work.

How does this change our world? None - all real models already follow proper gas kinetics rules - there is no problem to solve here.

But it does affect the fiddlers - those who ignored the basic gas rules and added on their own deco strategies - the RD people and the silly science justifications that get passed around the DIR classroom.

Ross, as somebody else stated before me in this thread and myself in others, this hurts you more than you think.
GF actually do do not alter the model, but only what is the maximum allowable supersaturation in a linear way.

That is the theory of GF, but the implementation is not done that way. The current design that is used out here, has GF patched onto the end of a calculation. this leads to inconsistencies and compounding problems.

I'd love to remake this ZHL/GF, done properly. Make the adjustment intrinsic to the calculation method. If these guys think they are going to fix their world with GF alone, they are chasing their tail.

 

[fsardone]

Hi,

Well.... no, that is not a hard rule. You are implying that ZHL limits are set in stone, and absolute, and the max possible, which is just not so. The USN tables will readily exceed ZHL limits.

True, but more gas or supersaturation is present the higher is th GF ... so comparing two profiles the one with higher GF is riskier for DCS ...

Hi,

That is the theory of GF, but the implementation is not done that way. The current design that is used out here, has GF patched onto the end of a calculation. this leads to inconsistencies and compounding problems.

I'd love to remake this ZHL/GF, done properly. Make the adjustment intrinsic to the calculation method.

Do it I believe you are more than capable of doing it, release it as open source, write a paper on why you believe this implementation has an advantage on the existin one and you will get validation (or not).

Cheers

 

[Patoux01]

Yes... efficiency. In decompression terms "efficiency" has a specific meaning. The least amount of time to achieve an error free result. Now none of the models we do (ZHL, VPM-B, etc), are as strictly "efficient". The navy tables are often more efficient. The objective of general use models is for a more gentle procedure - a balance.

I laughed

Also as I said before... if one wants to test a model - don't mess with it.... One cannot tell a model how to do its job, by forcing it into some artificial time constraints.

"The least amount of time to achieve an error free result". Going over that critical time (let's name it T, the perfect time for 0 DCS according to the perfect algorithm) will mean we're showing low efficiency, but with also no errors. Going below T will increase DCS necessarily. The more an algorithm increases DCS once we're below T, the less efficient that algorithm is. (should be clear to anyone, but I can make drawings should it really be required....)

IOW, please be consistent. If we're testing efficiency, it is required to have the same time. Unless you think having algorithm A plan 10hours of deco and come out DCS free while algorithm B comes out in 4hours of deco without DCS means they have same efficiency...

They did not test efficiency either. The profiles were deliberately made slow and long and easy. They then restored stress with excessive thermal stresses.

And again, both profiles were subject to "excessive thermal stresses".




@boulderjohn I disagree with you btw. As Simon said, "if it works for you, you don't have to change". So this is only relevant to people interested in this kind of stuff, and I hope those are able to type "scuba deep stops" in google. I don't think it requires redoing the whole thing over again, just let the first page point to the other threads and let this thing die in peace...

 

[boulderjohn]

@boulderjohn I disagree with you btw. As Simon said, "if it works for you, you don't have to change". So this is only relevant to people interested in this kind of stuff, and I hope those are able to type "scuba deep stops" in google. I don't think it requires redoing the whole thing over again, just let the first page point to the other threads and let this thing die in peace...

There's the problem. They aren't doing that. I used the phrase "blissfully unaware." If you are blissfully unaware of something, you don't out of your way to seek information about it. You find out about it by running into the information unexpectedly. That could come by stopping in at a site like this and blundering into a thread. Or it may come the way I saw it happen after a dive yesterday. It was a planned recreational dive, but three people who were also tech divers chose to do it in doubles. They did not know each other previously, and they naturally ended up buddying. I ended up joining them. At one point on the boat, the conversation drifted to these threads. One of them was a ScubaBoard lurker, and he talked about it. The other two knew nothing about it. Of course, I chimed in. So that is how two people learned that thinking is changing on deep stops. They did not say, "I wonder if thinking is changing on deep stops" and google it on their own.

 

[UWSojourner]

Yes... efficiency. In decompression terms "efficiency" has a specific meaning. The least amount of time to achieve an error free result.

Your description of "efficiency" is not correct.

Decompression Method A is more efficient than Method B if for the same dive:

1) Method A's recommended stops produce a lower decompression risk than Method B's given the same stop time, or

2) Method A can achieve the same DCS risk as Method B using less stop time.

Efficient decompression has nothing to do with "an error free result", whatever that means.

Based on the NEDU study, bubble count studies, etc., it is highly likely that all current bubble models (e.g. VPM-B) are "inefficient" in comparison to another decompression method that uses the same stop time but redistributes the time toward shallower stops.

That is why it's critical to compare profiles using the same runtime -- we are trying to compare how decompression methods like VPM-B and GF perform given the same time to work with. Ross's insistence on littering the web with meaningless charts showing dives of vastly different runtimes is just a fistful of powder being flung into the air to obscure the clear implications of current research.

 

[Dr Simon Mitchell]

VPM does NOT make any claim about efficiency

Yes it does! EVERY time a diver plans a dive using VPM the model prescribes a decompression stop regimen that the model assumes is the best way of decompressing from that dive. The model clearly implies that the prescribed stop pattern is the best use of the total decompression time. You would have to agree with that, right?

But what if redistributing some of the stops to different depths (eg shallower) within the same total decompression time lowered the risk of DCS? That would be a more efficient use of that decompression time. That is the point of the current research. It is suggesting that bubble models do not make the most efficient use of the decompression time they prescribe.

You cannot claim that VPM makes no claim about efficiency. It implies optimal efficiency EVERY TIME a diver uses it.

Simon M

 

[rossh]

Your description of "efficiency" is not correct.

Decompression Method A is more efficient than Method B if for the same dive:

1) Method A's recommended stops produce a lower decompression risk than Method B's given the same stop time, or

2) Method A can achieve the same DCS risk as Method B using less stop time.

Efficient decompression has nothing to do with "an error free result", whatever that means.

Based on the NEDU study, bubble count studies, etc., it is highly likely that all current bubble models (e.g. VPM-B) are "inefficient" in comparison to another decompression method that uses the same stop time but redistributes the time toward shallower stops.

That is why it's critical to compare profiles using the same runtime -- we are trying to compare how decompression methods like VPM-B and GF perform given the same time to work with. Ross's insistence on littering the web with meaningless charts showing dives of vastly different runtimes is just a fistful of powder being flung into the air to obscure the clear implications of current research.

You are not measuring efficiency...... you are comparing two successful dives..... success vs success.

Your argument is success A is better than success B, based on some internal differences in dimensions. But in both cases, neither dimension caused a problem.....

You assume your interpretations of those dimensions is a measure of risk, but you have no valid formula to verify with, or to make that assumption.

Your comparison is between the fixed dimensions of a real model, vs the vague, ever changing, inconsistent and undefined modifications, that you add onto some other model.... Not a real comparison.


*****************

You guys are spinning your wheels, trying to make up justifications for inflating plans, beyond any reasonable or required amounts, based on some invalid or distorted science measure, and poorly interpreted report results and contexts.....

If you want to focus on the bad practices, then please point a finger where it belongs - the RD and some DIR classroom theory - this is where you will find people ignoring the gas kinetic rules and short-cutting the shallow parts.

Please stop using VPM-B as a proxy for your problem. There is 15 years of success behind us, that does not need you lot fiddling with it.

.

 

[fsardone]

You are not measuring efficiency...... you are comparing two successful dives..... success vs success.

Your argument is success A is better than success B, based on some internal differences in dimensions. But in both cases, neither dimension caused a problem.....

Ross please ...
Successful outcome is a prerequisite see wiki:

Efficiency is the (often measurable) ability to avoid wasting materials, energy, efforts, money, and time in doing something or in producing a desired result. In a more general sense, it is the ability to do things well, successfully, and without waste.[1][2][3][4][5] In more mathematical or scientific terms, it is a measure of the extent to which input is well used for an intended task or function (output). It often specifically comprises the capability of a specific application of effort to produce a specific outcome with a minimum amount or quantity of waste, expense, or unnecessary effort.

In our case the resource is deco time and the money to keep the surface support team at sea.
What they did at NEDU was assessing how many people would get bent distributing the available and fixed deco time in deeper or shallower stops. They kept bending people untill there was statistica significance. They stopped early because they were bending many more divers with one distribution than the other.

The deeper stops showed being less efficient.
Therefore at same deco time less safe. Safety intended as the probability of getting bent.
[my conclusion]
a deeper start of deco as prescribed by VPM requires me to increase the shallower stop time beyond what VPM would require, in order to achieve the same probability of getting bent.

 

[Diver0001]

They kept bending people untill there was statistica significance. .

Yeah, about that. Isn't it an interesting sign of the times we live in that bending goats is no longer legal but bending navy divers is just fine!

R..