Question Why GF high not lower that GF low?

[bardass]

Computers like the ScubaPro G2 were running like GF 90/90 at the "L0 conservatism" level.

how do you know that ?

 

[bubblemonkey2]

how do you know that ?

Good question, fortunately there is an older thread on here where this came up, and someone actually compared profiles to arrive at this:

Difference between MB levels and Gradient Factors

Hi all! I was wondering if anyone could explain the difference between MB Levels (found on Scubapro computers) and gradient factors? I have a Scubapro G2 running Buhlman ZHL-16 ADT MB and I'm unsure what the distinction is between the different programmable MB levels (for added conservatism) vs...

scubaboard.com

I had been told it was just plain old 100/100, but evidently someone went on a 'deeper dive' to find out!

https://www.divetable.info/skripte/G2_GF.pdf

 

[rjack321]

It doesn't matter because Erik Baker's GFs never cross the M-value line. If you consider 100/100 "safe enough" then 100/90 should be "safer than enough"; it is unlikely that you'd have more people bent on 100/90 that on 100/100.

The testing will only tell you something new if you push a(*) GF past 100.

*) either of them

Negative ghostrider...

First I'm pretty sure that 100/100 hasn't been considered statistically "safe enough" for quite some time. Second with enough monkeys developing probability curves of DCS isn't that hard. Which is exactly what the latest navy models are doing. And it wouldn't take a super deep dive for 100/100 to have a 100% probability of DCS, mostly likely somewhere in the 80-90m range. Would fewer than 100% of people be bent at 100/90? Probably, but go 10m deeper and you're right back to 100% DCS again because Buhlmann isn't isorisk, which makes calling it "safer" than 100/100 profile specific.

 

[dmaziuk]

Negative ghostrider...

First I'm pretty sure that 100/100 hasn't been considered statistically "safe enough" for quite some time.

You're missing the point. What I was replying to is

You could do 90/70. It's like extra anti deep stop. I doubt there is any human testing of this kind of approach so best of luck with it

and the point is: by definition, 90/70 is no less "safe" than 90/90, which is no less "safe" than 100/100(*). There is little point in testing that on humans: it's gonna cost too much and the chances that you'll see more clinical DCS at 90/70 than you will at 90/90 are slim.

Probability monkeys are crunching the same empirical datasets that everyone else derived their M-values from (obviously we get more data points as time goes on). As long as our criterion is incidence of clinical DCS, to get new data points you have to bend humans. There are issues with that, and without that you can't expect any groundbreaking revelations.

*) Because GFs don't change how the model calculates on- and off-gassing: spend more time at deep stop by setting GF Low to 10, and it will keep track of your on-gassing there and pad your shallow stops accordingly. Reduce GF High and it will pad the intermediate stops so you'll off-gas more before you hit the surface. Theoretically no GF setting X/Y can get more people bent than 100/100, only fewer.

 

[inquis]

Theoretically no GF setting X/Y can get more people bent than 100/100, only fewer.

This is only true for X=100. If we depart from that and compare 10/100 to 100/100, I believe the probability of DCS is higher in the 10/100 case. They both have all tissues below their respective M values. However, the slower tissues are closer to their limits -- and remain that way for longer -- in the 10/100 case.

 

[dmaziuk]

This is only true for X=100. If we depart from that and compare 10/100 to 100/100, I believe the probability of DCS is higher in the 10/100 case. They both have all tissues below their respective M values. However, the slower tissues are closer to their limits -- and remain that way for longer -- in the 10/100 case.

Nope. Read Erik's paper.

 

[bubblemonkey2]

If we depart from that and compare 10/100 to 100/100, I believe the probability of DCS is higher in the 10/100 case. They both have all tissues below their respective M values. However, the slower tissues are closer to their limits -- and remain that way for longer -- in the 10/100 case.

I had to read this twice to make sure I was agreeing with it--here is more words about the idea

For the same bottom time, a 'more conservative GF,' whether it is GF 10/100, GF 30/70, or whatever, would push deco and runtimes to be so long, that there could be more net overall gas going into and out of the diver, because of all of the extra time at deeper stops, and because of, in theory, 'multiple compartments' that saturate at different rates.

So to keep low GF below ~50 for your 'controlling tissue compartment'--usually a faster one--you stay deep, continuing to on-gas all of the other 'compartments.' The net effect is that for the same bottom time, a 'conservative' GF is going to keep you in the water for a much longer time, deeper, exchanging much more gases into the body and back out. This could lead to unanticipated problems. There is also exposure, drifting, fatigue, operational schedules etc.

-OR- you drastically shorten your bottom time, spending a lot more of your fixed runtime doing extra deco stops and deeper depths.

For fixed runtime, 10/100 would be much 'safer', mostly because your plan or computer would seriously limit your depth and time, and instruct you to abort and decompress almost immediately--perhaps even while you are still trying to descend

The real questions:
You have a fixed 100 minutes of total runtime available.

1. What is the ideal apportionment of that fixed runtime to diving and decompressing?
2. At which depths do you believe your decompression will be optimal?
3. What GFs are going to optimally balance your bottom time, stop depths, and deco time?
4. What would be the relative effect on your dive plan, deco, body and health using a 'flipped' GF (e.g. 90/70)?

 

[lermontov]

Negative ghostrider...

First I'm pretty sure that 100/100 hasn't been considered statistically "safe enough" for quite some time. Second with enough monkeys developing probability curves of DCS isn't that hard. Which is exactly what the latest navy models are doing. And it wouldn't take a super deep dive for 100/100 to have a 100% probability of DCS, mostly likely somewhere in the 80-90m range. Would fewer than 100% of people be bent at 100/90? Probably, but go 10m deeper and you're right back to 100% DCS again because Buhlmann isn't isorisk, which makes calling it "safer" than 100/100 profile specific.

are you referring to this

Dial In Your DCS Risk with the Thalmann Algorithm

As some tech divers have painfully discovered, there’s an escalating risk of decompression sickness (DCS) with Buhlmann algorithms as dives get progressively deeper and longer. The risk can easily get into double digits in tech ranges—30 minutes at 120 m/394 feet with conservative 50/85 gradient...

indepthmag.com

 

[inquis]

Nope. Read Erik's paper.

I have, but I don't recall anything that contradicts what I said. It is quite focused on the leading compartment, and I am primarily talking about the rest of them. Would you mind being more specific about which statement(s) you disagree with? Are you saying that 10/100 does not yield higher tensions in the slower tissues than 100/100? That there is not more total inert gas in the body (all tissues combined) after surfacing on 10/100? Or that the total loading does not remain higher for longer on 10/100? Or that those aspects do not increase the probability of DCS?

 

[bardass]

Good question, fortunately there is an older thread on here where this came up, and someone actually compared profiles to arrive at this:

Difference between MB levels and Gradient Factors

Hi all! I was wondering if anyone could explain the difference between MB Levels (found on Scubapro computers) and gradient factors? I have a Scubapro G2 running Buhlman ZHL-16 ADT MB and I'm unsure what the distinction is between the different programmable MB levels (for added conservatism) vs...

scubaboard.com

I had been told it was just plain old 100/100, but evidently someone went on a 'deeper dive' to find out!

https://www.divetable.info/skripte/G2_GF.pdf

i know this
but it's absolutly wrong
I show this in this document (in french, sorry)