Diving too carefully?

[Mark Chase]

Hi Mark,

You have taken a ZHL plan, and tried to convert it into a VPM-B plan. And as you readily admit, they do not align perfectly. That is correct. GF is good for dramatically changing the ZHL, but why are you surprised that it is not perfect replication tool?

But you make it worse, by saying - my ZHL with conservatism, does not align with VPM-B with conservatism. It never will Mark. You are asking too much of GF.


I'm happy to see you using MultiDeco for supersaturation as an overall measure. Supersaturation is an independent measure.

Ross

I have not done anything of the sort

I have done what every GF divers does and applied GFs to a base line 100/100GF profile in order to make the deco less agressive

Then I have compaired the profile with a VPMB profile

Using the super saturation graph was your idea and in doing so we can see the initial super sat levels are now much closer but the final saturation levels on surfacing are 60KPA using VPMB2 and 40KPA using VPMB2

Does this not indicate ending the dive with a higher supersaturation level running VPMB2 or am I missing something?

 

[Mark Chase]

To get the VPMB2 and GF graphs closer together i have dropped the deep GF to 20 from 35.

Can you explain the significance of ending the dive on a 60KPA as apposed to a 40 KPA

Now we have the folowing comparison

GF 20/85

VPMB2

However in doing so I now have a run time for GF 20/85 of 204mins with 27mins at 6m and 43mins at 3m

As a reminder VPMB2 is showing a run time of 169mins and 20mins at 6 31mins at 3m

In total to match the deep stop phase of the dive I need to do 35mins more deco running GFs than i do running VPMB2. This always seemed obvious to me as id be on gassing some tissues during the deep stops an have greatly reduced presure gradients on other tissues so off gassing those more slowley

Only way to fix that was extending the shalow stops

Its little wonder so many swaped over to VPMB at the time with it reducing deco by 35mins but as i was saying at the time to get those shalow stop times down I'd need to be running 20/125+ GF

No doubt I will now be told theres no comparison between the two ascent profiles, which is fine, but I am reading now that theres little or no time advantage with VPMB and that was not its primary marketing tool. OK so if we are to accept VPMB2 is just agressive deco, what setting do you recomend?

The three profiles are

Depth.....VPMB2...... GF 20/85.......GF 20/100

48.... 0.33.... 0.33.... 0.33
45.... 1.... 1.... 1
42.... 1.... 2.... 2
39.... 2.... 2.... 1
36.... 3.... 3.... 3
33.... 3.... 3.... 3
30.... 3.... 4.... 3
27.... 4.... 5.... 4
24.... 5.... 5.... 5
21.... 6.... 8.... 6
18.... 7.... 8.... 8
15.... 9.... 11... 9
12.... 11... 15... 12
9..... 15... 19... 17
6..... 20... 27... 23
3..... 31... 43... 36


SO if we compair GF 20/100 with VPMB2 during the deep stop phase we discover the only diferance on the 48m stops to the 21m stops is 1min more at 42m 1min less at 39m

THATS IT???

But once we get into the shalow stops 18m & up we see extra time creeping in to the GF100 deco

So what is it about swaping the 1mins at the 42 and 39m stops that makes this reduced shallow deco possable?

We have a 1min reduction at 18 a 1min reduction at 12m and then 2 at 9m 3 at 6m and 5 at 3m?

Whats the betting if I could access 100GF+ on multideco, I could get these to line up even closer running 20/125GF?

 

[rossh]

Ross

I have not done anything of the sort

I have done what every GF divers does and applied GFs to a base line 100/100GF profile in order to make the deco less agressive

Then I have compaired the profile with a VPMB profile

Using the super saturation graph was your idea and in doing so we can see the initial super sat levels are now much closer but the final saturation levels on surfacing are 60KPA using VPMB2 and 40KPA using VPMB2

Does this not indicate ending the dive with a higher supersaturation level running VPMB2 or am I missing something?

Mark,

I repeat: You have taken a ZHL plan and severely modified it. In the process, you have abandoned the concepts of ZHL, and tried to emulate the attributes of a deep stop model. And you realize that your GF if not fully up to the job. Well - yes. And you keep posting more examples of this same issue.

The relationship between GF ratio reduction, and underlying model performance, is not a 1:1 linear relationship. GF was never made to know that level of detail. But you try to read this much information from it anyway. GF is not some universal model conversion tool. GF is not a superior or independent method.


Now look a bit further beyond this one dive example. Take a look at deco dive with just a few mins of deco - say OC 40m 10min air. See how the supersaturation is 140 kpa. The end pressure is dependent on the profile, and all the gas and time components it's made of. There is no correct value for supersaturation kPa for any one profile set.

 

[Mark Chase]

OK we all know GF is not a deco moddle in its own right but rather a way of setting a limit for presures against a theoretical 100/100GF M value.

Fine but to write it off compleetly seems disengenious

Lets just asume the deep stops are a multilevel dive

Correct me if I am wrong but you apear to be saying Bhulman deco cant cope with a multi level dive?


Can you explain what area of the math breaks down when devaiting from pure Bhulman by incorperating GFs?


From my humble understanding the controling tisues are taken to a specific presure gradient which at 100GF we call the M value The math keeps the diver at or below the M value during the ascent by decreasing depth at a rate which balances the reduction in presure gradient due to off gassing.

All GFs does is move the M value up or down

I am strugling to see how the math falls down by altering the theoretical value of M?

Could you explain?

ATB

Mark

 

[rossh]

OK we all know GF is not a deco moddle in its own right but rather a way of setting a limit for presures against a theoretical 100/100GF M value.

Fine but to write it off compleetly seems disengenious

Lets just asume the deep stops are a multilevel dive

Correct me if I am wrong but you apear to be saying Bhulman deco cant cope with a multi level dive?


Can you explain what area of the math breaks down when devaiting from pure Bhulman by incorperating GFs?


From my humble understanding the controling tisues are taken to a specific presure gradient which at 100GF we call the M value The math keeps the diver at or below the M value during the ascent by decreasing depth at a rate which balances the reduction in presure gradient due to off gassing.

All GFs does is move the M value up or down

I am strugling to see how the math falls down by altering the theoretical value of M?

Could you explain?

ATB

Mark

Answers in sequence above :

Yes GF is an addition to modify and extrapolate from ZHL, and severely if you so choose.

Not righting it off.

No, Haldane / Schriener can do multilevel just fine.

n/a.

Yes.

Yes.

The math does not fail.

It's the model concepts and limits you abandon. The observations and testing that created the ZHL limits, were made under a specific set of conditions. By adding excessive GF, you have extrapolated way beyond the original work. You apply so much GF, that the underlying ZHL model attributes is gone. And yet you expect the original ZHL limits to hold true and constant in such a precise manner. That's asking too much of GF.

 

[steady1570]

It's the model concepts and limits you abandon. The observations and testing that created the ZHL limits, were made under a specific set of conditions. By adding excessive GF, you have extrapolated way beyond the original work. You apply so much GF, that the underlying ZHL model attributes is gone. And yet you expect the original ZHL limits to hold true and constant in such a precise manner. That's asking too much of GF.

So you are saying that 20/85 is contorting the model too much for it still to be a valid approach? The body doesn***8217;t care how you came to your decision on where to stop, only the actual stops performed. Looking at Mark***8217;s tables, there are only 3 mins difference to 21m and 8mins to 12m on 20/85 vs VPMB2, which on deco times of 120-150mins is ~5%, vs ~25% difference in overall deco time. That seems small to me. If you have to hit these stops to such a degree of accuracy (including ascent rate, exact stop level etc) to make it work, then is the model of any practical use in the real world, where inaccuracies do happen? Are the extra 27 mins shallow asked for by 20/85 simply a relic of a stretched model and so not necessary?

What limits would you suggest on GFs to keep the model consistent? And what limits on conservatism on VPM?

thanks
-Mark

 

[Mark Chase]

It's the model concepts and limits you abandon. The observations and testing that created the ZHL limits, were made under a specific set of conditions. By adding excessive GF, you have extrapolated way beyond the original work. You apply so much GF, that the underlying ZHL model attributes is gone. And yet you expect the original ZHL limits to hold true and constant in such a precise manner. That's asking too much of GF.

I am having dificulty in understanding why its too much for GFs?

If we run 80/80GF we start with a KPA up arround 160 and it drops throughout decompresion down to 40 on the shallow stop.

If we feel that 160 is too high / too stssful then we can reduce that down to the 80 odd VPMB2 starts with by lowering the GF to 20?

When we do this the model recalculates the off gassing and on gassing of the deep stops which it would surley now treat as a multilevel dive as these stops were not necessary or were extended beyound the minimum

I cant see how this invalidates anything? The moddle just deels with what its presented with in the same way as a VPMB dive computer would need take adjustmet for an incorect ascent profile or simila multi level prfile?

What I am seeing is a dive to the same depth for the same time on the same gas yet a 35min quicker ascent running VPMB2 than we see running GF 20/85 dispite a very similer deep stop profile.

Now to me the only way we can reduce shalow stop time is to exceed the GF100 / M value. Either that or the M Value (IE Bhulmans math) is wrong

The concept of controling the critical tension of the bubbles must therefore be more eficient than the existing Bhulman system or it is simply more agressive.

Which is why i find it odd your aparently not claiming the VPMB moddle is more efficient?

ATB

Mark

 

[rossh]

So you are saying that 20/85 is contorting the model too much for it still to be a valid approach? The body doesn***8217;t care how you came to your decision on where to stop, only the actual stops performed. Looking at Mark***8217;s tables, there are only 3 mins difference to 21m and 8mins to 12m on 20/85 vs VPMB2, which on deco times of 120-150mins is ~5%, vs ~25% difference in overall deco time. That seems small to me. If you have to hit these stops to such a degree of accuracy (including ascent rate, exact stop level etc) to make it work, then is the model of any practical use in the real world, where inaccuracies do happen? Are the extra 27 mins shallow asked for by 20/85 simply a relic of a stretched model and so not necessary?

What limits would you suggest on GFs to keep the model consistent? And what limits on conservatism on VPM?

thanks
-Mark

The whole point of this double ended (hi/Lo) GF approach, is to distort the ZHL model and its plans, into something else. Yes it use ZHL as a basis, but the practice today (like your 20/85 example) is to throw away 80% of the ZHL information, and replace it with your own and very different model attributes.

If you take a VPM-B plan and add conservatism, you always get a plan with all the VPM-B attributes. But GF is so wide ranging, it can abandon the ZHL model attributes. The obvious example is that GF can almost emulate a deep stop model plan, or a military shallow model.

Mark is making a distorted / stretched ZHL plan this way, and then calling it a benchmark to judge other models with. You have made the same comments above - 2 mins here - 3 mins there. Sorry - that is not valid. You have extrapolated too far to call it true representation of the conditions that created the original ZHL. GF doesn't have that level of integrity back to ZHL. The differences you bring up, are a product of the extrapolation from the underlying model, and artifacts that creep into this and grow.

 

[rossh]

I am having dificulty in understanding why its too much for GFs?

If we run 80/80GF we start with a KPA up arround 160 and it drops throughout decompresion down to 40 on the shallow stop.

If we feel that 160 is too high / too stssful then we can reduce that down to the 80 odd VPMB2 starts with by lowering the GF to 20?

When we do this the model recalculates the off gassing and on gassing of the deep stops which it would surley now treat as a multilevel dive as these stops were not necessary or were extended beyound the minimum

I cant see how this invalidates anything? The moddle just deels with what its presented with in the same way as a VPMB dive computer would need take adjustmet for an incorect ascent profile or simila multi level prfile?

What I am seeing is a dive to the same depth for the same time on the same gas yet a 35min quicker ascent running VPMB2 than we see running GF 20/85 dispite a very similer deep stop profile.

Now to me the only way we can reduce shalow stop time is to exceed the GF100 / M value. Either that or the M Value (IE Bhulmans math) is wrong

The concept of controling the critical tension of the bubbles must therefore be more eficient than the existing Bhulman system or it is simply more agressive.

Which is why i find it odd your aparently not claiming the VPMB moddle is more efficient?

ATB

Mark

I did not say anything about efficiency. That would depend on the objectives, which have not been stated.

There is lot of assumptions in your post above.

A decompression comprises of all the stops and all the surface time, as one whole unit. But instead your looking at an individual piece of that ascent, and making it superior to all others. Sorry - not valid. You must consider the whole decompression.

What makes you think that 80% (GF 80) has a linear relation to risk reduction. Most things in decompression are based on exponential measures and values.

 

[Mark Chase]

What makes you think that 80% (GF 80) has a linear relation to risk reduction. Most things in decompression are based on exponential measures and values.

I beleive that 100% GF represents the theoritical maximum over presure we can tolerate before dammage (IE DCI) occures. Obviously this is not a line in the sand as there are too many variables and I have no doubt the 100% GF M Value has some level of conservitisum built in

But the purpose of the moddle was to calculate the minimum generaly safe decompresion time

If we accept we can reduce risk simply by increasing the decompresion time relative to depth, then surely resetting the max GF down to 80% represents a greater chance of DCI avoidance.

As I have said in the past this seems easy to acheive by simply reducing the GF to 80% of max across the board so 80/80GF

However since the start of the 00's we have beleived deep stops are benificial to decompresion. Part of this beleife is built on the advent of bubble models and part on empirical evidance from Richard Pyle & the WKPP

As a result we have split the ascent profile into deep stops and shallow stops and reduced the low GF down as far as 5% but in most cases 20% has been the GF low of choice

Thinking these deep stops were infact makeing the decompresion more effective we have gone more agressive on the shallow stops. I personaly went as far as 30/95 I know others running 10/125 in the beleif the deep stops could reduce the shalow stop time past what we would normaly see as the limit of 100%GF

After the deep stop test results came out sugesting we had it wrong we in the know seem to have taken a big step tawards 80/80 profiles.

The good news is we are keeping to or reducing our previous deco in water time but we yet again beleive we now have more efficient deco

So if we accept the NDEU tests for deep stop profiles were not representative of VPMB we must also accept that something is going on with VPMB which allowes the extended deep stops but also allows very short shalow stops yet somehow it still provides a good decompresion profile?


Is it fair to conclude that folowing a VPMB level 4 dive total run 189mins, we would have a higher bubble count (VGF) than a GF 80/80dive total188mins but that you feel the higher VGE count is irelevent?

ATB

Mark

 

[Cameron_R]

So having come to the party very very late; In the last 2 days I've read > 300 pages of forum debate, on this site and on RBW, in 4 separate threads about bubble models short comings, the NEDU study, VGE etc. Reading ~3 years worth of debate in just 2 days gives you a pretty concise view I think.

The only comment I'll make about the debate itself is that I'm astounded that in all of those > 300 pages, rossh did not once produce relevant data or scientific study and despite a raft of post by world experts arguing against him, continued to try and convince us the world is flat. That's what I see.


Now to the point of my post - a question to Dr Simon: In December 2012 you let us know what settings you were using at that time for your diving.

"Hello Drew,

I knew someone would ask this and I most definitely do not want to be seen to be providing recommendations.

However, I am happy to share aspects of my personal story. I was like Wayne Gerth in the early 2000s (as beautifully articulated by Bruce Partridge in his post). Everyone got caught up in the theoretical attraction of deep stop approaches to decompression. I used VPM for a period of time, and then gradient factors configured to mimic the bubble model approach to decompression (GFs ~ 20-90). I did over 500 hours on rebreathers using these approaches, including quite a number of dives over 100m and suffered no problems other than two episodes of musculoskeletal "niggles". In recent years as the deep stop issue has become controversial and the evidence we have been debating has emerged, I have altered my gradient factor approach and now use 35 - 75 on a Shearwater, or settings on Kev Gurr's VGM that produce similar profiles depending on which rebreather I am wearing. These changes have not been driven by perceptions of how I feel, but I have had no problems over a further 300 - 400 hours. I don't know whether this is right, but it seems to me to be an appropriate response to the current state of the available evidence. My decompressions are quite conservative, but I am often the only physician in very remote locations!

Simon M "

I am not looking for a recommendation, in fact even if you gave one I couldn't use it, but I am curious to see how your diving has evolved over these 3 years given any advances in the science.

Any chance you could give us an update on what settings you dive now?

 

[Mark Chase]

Is it fair to ask someone with so much authority / respect in the diving world what deco profiles they use?

Is it not our job as qualified decompression divers to assess the information available and make our own decisions?

Being just another diver with no expertise in decompression past the usual diver qualifications and a bit of internet surfing, I can tell you I progressed in increments from 35/95 when this whole debate kicked off to 70/80 which is where I am today. I did 4-5 dives at each progressive stage most were 60-90m with a total run time of 180ish mins regardless of depth.

I am pleased to report I have felt pretty good and I am 51, moderate drinker, light smoker and 1.78m tall 105kg

That said I have been diving 26 years and only suffered niggles so I may be unbendable under normal diving conditions. I have certainly done dives where I expected to be bent post dive.

Follow my lead at your own perrel

ATB

Mark

 

[Cameron_R]

Is it fair to ask someone with so much authority / respect in the diving world what deco profiles they use?

Is it not our job as qualified decompression divers to assess the information available and make our own decisions?

Being just another diver with no expertise in decompression past the usual diver qualifications and a bit of internet surfing, I can tell you I progressed in increments from 35/95 when this whole debate kicked off to 70/80 which is where I am today. I did 4-5 dives at each progressive stage most were 60-90m with a total run time of 180ish mins regardless of depth.

I am pleased to report I have felt pretty good and I am 51, moderate drinker, light smoker and 1.78m tall 105kg

That said I have been diving 26 years and only suffered niggles so I may be unbendable under normal diving conditions. I have certainly done dives where I expected to be bent post dive.

Follow my lead at your own perrel

ATB

Mark

Fair? Well I don't think it's unfair. Cheeky maybe. I'm not going to be worried if Simon doesn't answer. He's a busy guy and I realise that my question is the least of his worries.

Thanks for the info on your journey though. Don't worry I won't be following (not because I think you are wrong in anyway). I just find it interesting. Like I said - I read the entire debate in 2 days, so it all unfolded in sharp focus for me.

 

[Mark Chase]

Fair? Well I don't think it's unfair. Cheeky maybe. I'm not going to be worried if Simon doesn't answer. He's a busy guy and I realise that my question is the least of his worries.

Thanks for the info on your journey though. Don't worry I won't be following (not because I think you are wrong in anyway). I just find it interesting. Like I said - I read the entire debate in 2 days, so it all unfolded in sharp focus for me.

Not having a go just pointing out that some one in Simons position would be lothed to offer up something people may blindly follow bassed on his reputation.

Every one knows i am a tosser so its no problem for me

 

[Sven Becker]

Not having a go just pointing out that some one in Simons position would be lothed to offer up something people may blindly follow bassed on his reputation.

Every one knows i am a tosser so its no problem for me

I blindly follow your reliable comments!
The only whimsicall decision you have taken, was to sell your first KISS C ; >

 

[Simon Mitchell]

Not having a go just pointing out that some one in Simons position would be lothed to offer up something people may blindly follow bassed on his reputation.

Mark, thank you for the sentiment.

Cameron, Mark is right to a large extent, though I am always happy to provide commentary that reflects the best evidence available. In that regard I can say a couple of things about how to approach deep stops.

First, the best evidence currently available suggests that over-emphasising deep stops (especially relative to the amount of shallow stop time you do) does not produce a safer or more efficient decompression as originally hypothesised. Obviously this begs the question, what does "over-emphasising" deep stops mean? This question cannot be answered with precision but (again, with reference to the best evidence available) it almost certainly encompasses the approach prescribed by currently used bubble models.

Second, notwithstanding the first comment, "the truth in the universe" about optimal decompression is not defined / known. Thus, while we perceive that the bubble model approach is not optimal, I cannot tell you exactly what optimal is.

My personal response to this evaluation of where we currently sit has been to stop using bubble models (which I did for a long time with minimal problems [2 cases of niggles]) and back away from deep stops by using gradient factors. The problem is knowing how far to back away. Personally, I have evolved a little since that last post... I am currently at a GF low of 50. But because a lot of my diving these days is done on "expedition style" trips to remote locations (and I am often the only physician), I do make my decompressions quite conservative by using a GF high of 70 or 75. Especially in tropical locations I really don't care about extra time in the water.

In the spirit of what Mark was saying I do want to emphasise that this is a personal response to my interpretation of the current evidence and it is not intended as a recommendation for general use. I certainly have no idea of whether it is optimal, but I do think it is sensible. The problem with recommendations is that whereas DCS can occur no matter what (realistic) decompression profile is adopted, if someone starts adjusting their decompression and then suffers DCS they will inevitably blame it on the adjustments.

Simon M

 

[Cameron_R]

Not having a go just pointing out that some one in Simons position would be lothed to offer up something people may blindly follow bassed on his reputation.

Every one knows i am a tosser so its no problem for me

Mark, thank you for the sentiment.
...

My personal response to this evaluation of where we currently sit has been to

...

Thanks Simon. What I find interesting is not the specific GF's or model you use now, but the shift over the years.


Funnily enough the first person in the 'debate' that I saw predict the shift was Mark back in 2012

....

If we accept Bhulmans model then shouldn't we be running 80/80 rather than 20/80?

...

 

[Michael Thornton]

First, the best evidence currently available suggests that over-emphasising deep stops (especially relative to the amount of shallow stop time you do) does not produce a safer or more efficient decompression as originally hypothesised.

I am glad you said this. "(especially relative to the amount of shallow stop time you do)"

I have heard several sub par GF discussions among divers or even presentations where they are so focused on ascending as fast as possible even on dives where everything is going fine. I switched from 15/85 to 30/70 and had significant less bubbles. But I think it more had to do with the upper number changing because I wasn't actually diving a 30/70. I was in Cayman and coming up slow because there was still a lot to look at.

I agree that if I had to bail out and had a limited supply of gas I would be much better off getting shallow fast but my bubble count went down when I switched my computer to 30/70 but I was still coming up so slow it was probably more like 15/70. People are getting really caught up in the lower number and not realizing that some of the change can be from the high number.


Sent from my iPhone using Tapatalk

 

[rossh]

I am glad you said this. "(especially relative to the amount of shallow stop time you do)"

I have heard several sub par GF discussions among divers or even presentations where they are so focused on ascending as fast as possible even on dives where everything is going fine. I switched from 15/85 to 30/70 and had significant less bubbles. But I think it more had to do with the upper number changing because I wasn't actually diving a 30/70. I was in Cayman and coming up slow because there was still a lot to look at.


Sent from my iPhone using Tapatalk

EXACTLY ! This in not about deep stops. It is about the addition of extra shallow time that is working the magic here. Stay longer, and reduce the stress This is a basic principle we all learned in scuba class.


Going from 25/85 to 40/70 has this typical basic change to decompression time:

A 1% reduction in deep stop time; A 12% increase in shallow stop time.


The 12% extra is where the changed outcome occurs, not the 1% reduction. But some people want to pretend its the deep stops.


By making the changes to 40/70 type of profiles, its adopting a longer and slower profile and reducing stress, at the cost of extra and unnecessary decompression stop time. One could do the same additions to a deep stop model plan too, and get similar changes.

It all depends on what ones goals are, and the desired condition of "optimal" is.

 

[Mark Chase]

EXACTLY ! This in not about deep stops. It is about the addition of extra shallow time that is working the magic here. Stay longer, and reduce the stress This is a basic principle we all learned in scuba class.


Going from 25/85 to 40/70 has this typical basic change to decompression time:

A 1% reduction in deep stop time; A 12% increase in shallow stop time.


The 12% extra is where the changed outcome occurs, not the 1% reduction. But some people want to pretend its the deep stops.


By making the changes to 40/70 type of profiles, its adopting a longer and slower profile and reducing stress, at the cost of extra and unnecessary decompression stop time. One could do the same additions to a deep stop model plan too, and get similar changes.

It all depends on what ones goals are, and the desired condition of "optimal" is.

The point is I / we were adding significantly deepr stops and REDUCING the shalow stop time in the beleife that I / we were doing better deco.

Way back in the 00s when I started GF deco we ran 35/80 which seemed to be a default seting a bit like VPMB2.

As we became aware of the trend taward deep stops and reduced shalow stops, this became 20/95 and as I have said in the past some friends were running 5/125GF profiles

The big move for me today is to run longer shalow stops with less deep stops and my inwater time has remained prety much the same.

And again I do this (bassed on the best available information) in order not to do less deco but to make the deco I do more eficient.

Sure I could run 15/70 and still remain very much in the comfort zone but that would have an impact on my "usable dive time to deco" ratio

I am expecting to do a 45min bottom time at 70m and 2hours 15mins deco. I have done this dive many times and not been bent but I am always looking to improve my decompresion, especialy as i get older.

The only question then is, do I run deep stops and short shalow stops or no deep stops and extended shalow stops within that 2 hour 15min window.


ATB

Mark