Effect of slow compartments size in relation to NDL and DECO

[tbone1004]

@Dan_P your logic escapes me.
Dive tables that are printed for specific dive plans say "go up now"
Dive computers, at least the good ones, track your theoretical gas loading in real time and remove a huge amount of risk placed on a diver by following ratio deco. I.e. ratio deco is horrific for cave dives where you have huge multi-level profiles and you have one of two options. Run the risk of getting bent by making a math error in your head, or spend an inordinate amount of time on deco.

All of this "strategy" talk can't be incorporated into dive tables because they are printed plans. There is nothing that says you can't apply those strategies to computers that have a preprogrammed safety buffer into them *your gradient factor*

Relying on your brain to accurately calculate decompression is downright stupid when there are reliable and flexible computers on the market. If you want to employ Ratio Deco, fine, you have yet to make an argument on why you shouldn't at least run a computer to follow behind you to make sure you don't seriously mess up.
There is a damned good reason that no one uses Ratio Deco as their primary ascent strategy in cave country, doing deep hypoxic cave dives on rebreathers, or shallow easy dives on nitrox OC.
It's one thing to advocate doing it in open water with square profiles where the risk is much lower, but it's downright dangerous to advocate use as a primary decompression profile. Add to that that it takes longer than reasonable GF settings to get out of the water and was based on "how AG felt" when he got out of the water, and you have a recipe for someone to get bent. Oh wait, they already have, many times

 

[RainPilot]

Thanks for the correction @tbone1004 and @victorzamora

I need to relook at surfacing GF and how it affects the curve. My understanding seems to have been based on a faulty assumption.

 

[tbone1004]

Thanks for the correction @tbone1004 and @victorzamora

I need to relook at surfacing GF and how it affects the curve. My understanding seems to have been based on a faulty assumption.

@victorzamora can correct me if I missed anything but super TLDR

The Buhlmann algorithm is linear and we use 10ft intervals for stop depths to make ascents easier. Slightly less efficient, but it's basically impossible to follow the GF-curve linearly to the surface because no one has the ability to control their ascent rates that accurately

GF-low determines how deep your first stop is as a function of your maximum depth-this is why with a dive profile you can pick your GF-low and then change your gf-high to whatever you want and the first stop depth won't change.
The GF-low does have an impact on your total decompression time because your slow tissues are still ongassing at the deeper depths. The lower the GF-low the longer your total decompression obligation to clear those slow tissues

GF-indirectly high determines how long you have to stay at each stop, starting with the first stop determined by the GF, but is the maximum theoretical loading you will allow your tissues to see when you surface.

The ratio between the GF's determines the ratio between the m-value line and the GF line at different points. If your gf's are 1:1, then you will have the same delta between the gf line and the M-value line throughout your ascent. If you have a value >1 between high:low, then the gap will be larger at depth, and narrow as you get closer to the surface. If you have a value <1 then the gap will be small at depth and grow as you get closer to the surface.

Ratio Deco has a VERY low GF-low equivalent *somewhere between 5-15 depending on the dive profile* which means that those slow compartments are ongassing for most of your ascent and that's why they have much longer shallow stops and a longer total decompression obligation for the same bottom profile as "normal" Buhlmann profile as shown with Victors plot. The problem comes with the fact that they have a very high GF-high which is why people get bent when diving it because those slow tissues haven't had time to fully offgas.
NAUI RGBM has the exact same problems and is a huge reason why many NAUI technical divers will not teach NAUI Tech because it is required to teach RGBM in the courses.
Ratio Deco is the sole reason I will not consider teaching for UTD, and RGBM is the sole reason I will not consider teaching tech for NAUI despite being quite proud to be a NAUI voting member for close to 10 years

 

[dmaziuk]

Thanks for the correction @tbone1004 and @victorzamora

I need to relook at surfacing GF and how it affects the curve. My understanding seems to have been based on a faulty assumption.

GF Low determines the depth of your first stop. Before you calculate that, you have to decide if a stop is needed at all. That's where GF High comes in: if you can reach the surface without violating it, then you don't need a stop and GF Low does not apply. So in principle there may be "borderline deco" profiles in the grey area that makes algorithms "conservative" or "liberal", where GF High is the deciding factor. Once you're out in deco land, that should no longer happen and I expect Eric Baker's graph was drawn for "well in the deco land" case.

 

[tbone1004]

@dmaziuk correct

 

[rossh]

This might help understand how GF is applied.

Its just a simple math exaggeration of the base line. The Lo is set first, so it creates a deeper start. The Hi result is dependent on the state, the outcome of the Lo addition, and the resulting jumping of cell time boundaries. Meaning the Hi is not truly proportional to the baseline, and instead quickly compounds excessively and becomes an over-inflated value.

.

 

[Dan_P]

@tbone1004 hold on, bent divers on RD 2.0? I just want to be certain what you're saying here, exactly.

Regardless, I don't think my point is clear.
What I'm saying is practicability versus accuracy.

To illustrate, that if one diver on ccr using RD bails out, it makes no difference in terms of the deco compared to his/her buddy's deco profile.

It's simple and cohesive with RD, standard gases and a BOV, that's what I'm saying. It's only an example, but my point is practicability versus accuracy on a general note.

The point of RD is not to be 100% accurate in mimicking what happens in the body in a world where no algorithm is.
That says something about why I personally don't feel a need for computers - it's like having a hard-ish time seing through the windshield of my car but advocating that I should use cruise control for safety.
Blow the whistle, ref.

It's not that I think it would be dangerous for me to carry around a computer with RD as a software in my pocket, but rather that I doubt it would tell me anything useful that my buddies can't and I should be extremely comfortable with as a baseline in my diving.

I don't think it'd help me if I'm so stressed out I can't think.

On the note of having a computer because a human brain is no good - is that not why we have buddies?
To function as backup brains?
If I decide north is south, they interfere.
I don't need an electric compass that tells me to go to 359,21 degrees and blare alarms if I deviate a bit.
I need a functioning brain, an arrow with an "N" and my team.

Honestly, if I'm doing a dive with RD that I'm trained for, and I can't even think out the decompression, then I've already put myself way over the edge.
To me, the issue here is staying within my comfort zone.
Can automation blur that line? Be honest, tell me that's not a very real risk - especially if we scale up to a whole organization of divers.

In lieu with the example of ccr, the equivalent would be to employ automation solely because I can't trust myself to look at the reading and adjust every once in a while like I do with my drysuit or wing.
That's exactly what the recipe for disaster looks like - offboarding control to an automated system in a human/machine interface.

Meanwhile, speaking of deviations
I'd be happy to continue our discussion and answer questions as well as hear your thoughts, but maybe pm would be suitable?

 

[tbone1004]

@Dan_P 2.0 hasn't been out long enough for the insignificant amount of divers that use it to have racked up enough deco dives to bend themselves yet, however I can almost guarantee that someone has bubbled from it already.
I know of at least a dozen that have been bent on the original RD platform.

Your reliance on your buddy as your redundancy for decompression makes me more nervous than the reliance on your brain over a computer to conduct mathematical calculations in real time.

I am not arguing that divers shouldn't understand decompression and conduct proper planning prior to getting in the water.
This is the best analogy I can think of. You are building a race car and have to figure out how you want to control the engine.
Ratio Deco *in any form, not specific to UTD, just the principal of only calculating decompression in your head* is like having to control fuel mixture and timing in real time, and if you let go of the levers that control them, the engine dies.

Cutting tables is like using a traditional carb and distributor. You choose what you're going to "tune" it for, and it's optimized for that instance, but may not be great outside of that parameter. It has the advantage of running without other input from the driver. Where this could be bad is if you have a drag car and then an emergency happens and you have to drive it home through rush hour. It's not going to run well at all and is going to be horribly inefficient.

Running a modern computer is like having EFI with electronic ignition where the sensors are constantly adjusting the fuel mixture and timing in real time.

In either circumstance, in order to tune the carburetor or the EFI, you have to understand how and why it works in order to make it perform the best.
Understanding that you can program the computer to make those calculations faster and more reliably than you could ever do is why you run a computer. In the event a computer fails, you still understand how to get yourself out, but you use that as a contingency

 

[Dan_P]

hasn't been out long enough for the insignificant amount of divers that use it to have racked up enough deco dives to bend themselves yet, however I can almost guarantee that someone has bubbled from it already.

What is the number of divers that use it?

And how long would it need to be in application to satisfy your measure of significant statistical data sample?

I only ask because I think it's fair to be clear on statements so noone ends up like that guy who went out on forums saying the MX is a euthanasia device the better part of a decade ago only to find years and years later still noone died on it.
What was his name again...

I know of at least a dozen that have been bent on the original RD platform.

So these dozen or more guys that you know who got bent, they are all UTD divers?

Your reliance on your buddy as your redundancy for decompression makes me more nervous than the reliance on your brain over a computer to conduct mathematical calculations in real time.

Conversely, the notion that I shouldn't expect neither myself or my team to do simple thinking while in the water, leaves me paradoxally flabbergasted.

Ratio Deco *in any form, not specific to UTD, just the principal of only calculating decompression in your head* is like having to control fuel mixture and timing in real time, and if you let go of the levers that control them, the engine dies.

Exept for the fuel mixture part. And timing in real time. And the engine dying.

It's more like knowing that your full tank of petrol will get you 300km and figuring out approximately how far half a tank will get you. And then you check with your passengers to see if they agree.

I don't mean for this to be condescending, but I have to ask if you've actually been trained in using RD?

Cutting tables is like using a traditional carb and distributor. You choose what you're going to "tune" it for, and it's optimized for that instance, but may not be great outside of that parameter. It has the advantage of running without other input from the driver. Where this could be bad is if you have a drag car and then an emergency happens and you have to drive it home through rush hour. It's not going to run well at all and is going to be horribly inefficient.

Okay...

Running a modern computer is like having EFI with electronic ignition where the sensors are constantly adjusting the fuel mixture and timing in real time.

https://www.ntsb.gov/investigations/AccidentReports/Reports/HAR1702.pdf

 

[RayfromTX]

I rarely see an analogy used in a helpful way. When they are used to try to build a case or prove a point they often backfire as they are off topic enough to allow an off topic response to now be legitimate if only in the context of competing analogies. My two favorites are car analogies and sports analogies. It might be more effective to put the creative energy into crafting a more convincing response. It's like when the football coach is driving the team bus and the players are trying to figure out what the funny noise is coming from under the hood and it turns out it's a corvette mechanic in the cargo hold playing baseball with his wrenches and sockets and......oh never mind.