Breathing rate, air integrated computers and DCI correlation

[redacted]

If you want an AI wireless computer, the Galileo is still one of the best out there. In this class there is not a lot of competition. By the way I have mine set the MB 1 as I found that setting it to 2 was putting me into level stops out of step with everyone else diving in Cozumel. And I have my workload set to heart rate only. Setting workload to breathing was causing too many high workload alarms.

Why is your computer giving you stops in Cozumel?

 

[Hatul]

Why is your computer giving you stops in Cozumel?

In the Galileo you increase the conservatism by setting an MB level above 0. When set to zero you get the standard Uwatec ZH 8 tissue algorithm that allows non deco and deco dives. Setting MB above 0 increases conservatism, so before going into deco the computer gives you optional deco stops called "level stops". In Cozumel I was diving with Aldora with 120 cu ft tanks allowing long dives. When set to MB 2 and on air I was getting into these level stops while others were surfacing and getting back into the boat. I found setting MB to 1 made the Sol behave more like other divers' computers and still provided some extra safety. Nitrox also helped.

 

[RonR]

<snip>......
With whiz-bang gizmo computers, what we find are all sorts of model interpretations by programmers under the direction of product liability lawyers. You can search for a fascinating conversation I had with RonR here on ScubaBoard about the Atomic Cobalt. The Cobalt runs "Proprietary RGBM" (whatever that means) and I observed it did weird and highly unpredictable things beyond 150ft. Turns out, the manufacturer indeed included a "Folded Model" (whatever that means) and it does modify the model (how we'll never know) in real time at greater depths. Okay. No Problem. How do we plan a technical dive with your "multi-gas" computer accurately? Anyone who's ever spent time around a Suunto computer knows about the "Suunto Minute", also a recreational interpretation of "Proprietary RGBM".

Now I know by now people are going to jump on me for picking on bubble model, but don't waste your time. I'm not picking on bubble model, what I'm doing is pointing out that when the manufacturers or their lawyers manipulate the model we can't understand where we started or what we need to adapt our processes to best utilize it.

It is my belief these features exist to convolute the dive planning process and limit legal recovery. I can imagine that a diver using one of these devices with a half dozen settings has a reduced opportunity to sue/recover damages upon being injured. I can visualize the manufacturer's attorney saying, "You failed to set the computer to morbidly obese as shown on page 134 paragraph 2, therefore you didn't do it correctly, sue your instructor who failed to teach you how to use this properly and oh btw – learn to read the manual."

I've no desire to get into the fray here, except to say that I'm skeptical about the marketing of features when there isn't really a clear basis for knowing what to do with the information. But when this thread jumped over to the Computers forum I saw I had been mentioned. I will try to clarify points above.

The Cobalt has two algorithms inside. As the Cobalt's dad, present from birth to graduation, I can guarantee that lawyers had nothing at all to do with this or any other functional decision concerning the development of the Cobalt. They did review the manual.

At normal recreational depths, everything shallower than 150', Cobalt uses folded RGBM. "Folded" RGBM is a term that gets applied to Haldanian algorithms that take the information from a fully iterative bubble model, and "fold" that into the Haldanian method of calculation (i.e. using m-values, tissue half times, etc.), adding in factors designed to minimize bubble formation. It's what essentially all recreational "RGBM" computers are, including Suunto, and very much analogous to adding bubble factors to the computers that use Buhlmann algorithms. That is to say, marketing and buzzwords aside, it's about what almost all current dive computers do. The biggest reasons for doing this rather than just using the fully iterative algorithm are that 1) the fully iterative bubble models are very computationally intensive, and doing them in real time in a dive computer just was not feasible until low power processors got faster, and 2) the folded methods tend to yield good results at shallow depths, with an excellent track record. You still need to generate no-stop times for recreational divers, so fully iterative calculations (which essentially only do deco schedules) at recreational depths just represent more overhead for no gain.

The "proprietary" part really just refers to the fact that each manufacturer can dial in their own preferences for overall conservatism, or add features they think will help them sell computers- yep, marketing. Atomic wanted to go very middle of the road (not the most conservative or the most liberal) for overall baseline conservatism, add adjustments for expected workload (diver determined), age, and offer three levels of user desired conservatism adjustments. That's really what makes it "proprietary".

We did not tie the algorithm to gender or gas consumption or water temperature, because those factors are much harder to reliably correlate with real world risks in any useful way. I can be freezing cold in warm water, and overheated in cold water. I could be a big person, very relaxed, using lots of air, or a tiny person, working hard, but with low consumption. I use more air in cold water even with a low workload. And air consumption is not a direct measure of actual muscle effort, which is the concern. We can't tell what you will be doing when you get out of the water- another risk factor. Basically we opted to keep things simple rather than offer what, in my opinion, is a false sense of precision- and complicate the user interface. All dive computers are imperfect guides. I'm with HIGHwing in that understanding the limitations is preferable to a false sense of precision. We do provide onboard calculation of SCFM (or l/min) consumption in the log because it is a useful metric for the diver to evaluate their performance.

When the diver goes below 150' the Cobalt switches to a fully iterative RGBM algorithm. This depth is arbitrary, based on not wanting to confuse recreational divers by the schedule shift. The full schedules will start deeper, generally, and look a lot like a very slow ascent. They are based on the Cobalt modeling, in real time, the known physics of bubble formation and trying to minimize that bubble formation through a controlled ascent. On deeper dives, deco is a given, so the limitations above don't apply and we calculate schedules using the full iterative model. While I don't in any sense think of the Cobalt as a Tech computer, this crosses over a bit onto the border lands. So far as I know, the Cobalt is the only recreational computer to incorporate a fully iterative bubble model, though both RGBM and VPM are now available for some tech computers.

Both these algorithms are included in the Cobalt's onboard planner, and are available in third party planning software as well. I'm not happy about the fact that RGBM itself is closed and proprietary as opposed to open and available, but having seen both sides I can say that generally algorithms are more alike in their results than they are different. My own feeling is that algorithms get too much attention, they have all worked pretty well, and the things that probably matter more to the diver, like the user interface, get too little focus.

Ron

 

[gcarter]

I work in managerial accounting. Whenever I look at a bit of data, I ask myself if it is going to move any needles - will it cause someone to change behaviour, will it influence decisions, will it change any results in a statistically significant way. If not, I don't bother wth it.

I have not yet seen anything to convnce me that computer monitoring of my heart or breathing rate falls on the right side of that ledger. Interesting data, yes. But I am not interested in a contest to move my heart rate from X to X-1. I am interested in DIVING. I am interested in gaining experience and the comfort and relaxation that comes from that. Heart rate, breathing rate, air consumption are all results that follow. Measureing them does not change the factors that result in them, so why do I need to waste attention on looking at their numbers during a dive? The right breathing rate is what is comfortable. The right heart rate will follow from breathing rate and general fitness.

At the end of the day, will knowing the numbers for any of those things change anything?

Not at all. It isn't a contest.

This is not armchair baseball where anything that isn't nailed down gets a statistic attached to it just because someone can.

And I will repeat myself from a previous post: precision DOES NOT equal accuracy. Being able to measure something, even to multiple decimal places, does not mean that your measurement is an ACCURATE representation of anything. It could just be more precisely wrong - or irrelevant.

YMMV

 

[lowviz]

I've no desire to get into the fray here, except to say that I'm skeptical about the marketing of features when there isn't really a clear basis for knowing what to do with the information. ...//...

...one reason that I always read what this guy writes.

...//... The "proprietary" part really just refers to the fact that each manufacturer can dial in their own preferences for overall conservatism, or add features they think will help them sell computers- yep, marketing. Atomic wanted to go very middle of the road (not the most conservative or the most liberal) for overall baseline conservatism, add adjustments for expected workload (diver determined), age, and offer three levels of user desired conservatism adjustments. That's really what makes it "proprietary". ...//...

Would it be in bad taste to mention that Atomic began with disgruntled ScubaPro regulator designers?

...//... I'm not happy about the fact that RGBM itself is closed and proprietary as opposed to open and available, ...//...

Massively complex calculations. I'm guessing "untidy" assumptions or shortcuts are being left undisturbed.

...//... They are based on the Cobalt modeling, in real time, the known physics of bubble formation and trying to minimize that bubble formation through a controlled ascent. ...//...

Any comments on leaving the bottom and rate of ascent to the point where you begin off-gassing?

...//... My own feeling is that algorithms get too much attention, they have all worked pretty well, and the things that probably matter more to the diver, like the user interface, get too little focus.

Ron

Wow. If I was a manufacturer of DC's I'd have Apollo Robbins design the UI. Now there is a person that truly understands what people actually see.

Thanks for posting!

---------- Post added September 20th, 2013 at 02:29 AM ----------

...//... The right heart rate will follow from breathing rate and general fitness.

At the end of the day, will knowing the numbers for any of those things change anything?

Not at all. It isn't a contest. ...//...

To add to your point, my barometer (at home) goes from about 790mm Hg to 710mm Hg. That is a range of 80 mm Hg.

Let's say I want to get one last dive in before a bit of bad weather (or a "one and run"). Let's also suppose that the barometer drops 20mm as the front moves in while I'm under. This can be a point of real concern.

My calcs give me a 2+ letter group advance on this not so far-fetched scenario for a scant 20mm Hg drop. My DC, any DC, is incapable of correcting for this. How could it know that I switched to altitude diving while underwater? I don't care how freaking precise it is, it is not tracking what really counts.

Bottom line: If I'm using an "aggressive" DC and dancing on the line in this particular situation, I deserve to get bent. If I'm using a Swiss watch, it will be wrong too.

---------- Post added September 20th, 2013 at 04:19 AM ----------

-for those actually doing the calcs, the 20 mm Hg differential takes into account my best guess of the time it takes a DC to re-equilibrate itself during barometric change. Informed input would be most welcome...

 

[redacted]

In the Galileo you increase the conservatism by setting an MB level above 0. When set to zero you get the standard Uwatec ZH 8 tissue algorithm that allows non deco and deco dives. Setting MB above 0 increases conservatism, so before going into deco the computer gives you optional deco stops called "level stops". In Cozumel I was diving with Aldora with 120 cu ft tanks allowing long dives. When set to MB 2 and on air I was getting into these level stops while others were surfacing and getting back into the boat. I found setting MB to 1 made the Sol behave more like other divers' computers and still provided some extra safety. Nitrox also helped.

It seems to me that this could be teaching some divers bad (dangerous) habits. Why would you select some level of added conservatism and then reduce that level for no more reason than other divers in your group are able to end their dives (and begin their surface interval) while you are still stuck on a "level stop" (decompression stop)? Why have this feature that both the diver and the computer can blow off if a stop turns out to be inconvenient or missed. "The computer is giving me stops but I don't have to."

Why not just go with a less expensive, more liberal computer; and follow it?

 

[nimoh]

It seems to me that this could be teaching some divers bad (dangerous) habits. Why would you select some level of added conservatism and then reduce that level for no more reason than other divers in your group are able to end their dives (and begin their surface interval) while you are still stuck on a "level stop" (decompression stop)? Why have this feature that both the diver and the computer can blow off if a stop turns out to be inconvenient or missed. "The computer is giving me stops but I don't have to."

Why not just go with a less expensive, more liberal computer; and follow it?

or go with a bottom timer, plan your stops before the dive and ensure you have enough gas to make those stops with appropriate reserves.

Then, everyone is making the same stops.

 

[JohnN]

It seems to me that this could be teaching some divers bad (dangerous) habits. . . .

I lifted the following from the Scubapro Galileo Luna web page. . . I'd be very interested if anyone has consumed this much Kool-Aid and actually would use this:

Predictive Multi-Gas upgrade: Purchasing this PMG feature at a later time, or whenever you need it, allows LUNA to use up to 3 Nitrox mixes (21-100% O2) during the dive, calculating the ascent schedule for all possible combinations of the programmed mixes. The outcome of the different decompression calculations can be shown on the screen in a dedicated deco summary screen.

This "feature" strikes me as 100% dangerous, doing a multi-stage deco dive, while flying your computer ??

 

[Dr. Lecter]

I lifted the following from the Scubapro Galileo Luna web page. . . I'd be very interested if anyone has consumed this much Kool-Aid and actually would use this:



This "feature" strikes me as 100% dangerous, doing a multi-stage deco dive, while flying your computer ??[/FONT][/COLOR]

One hopes anyone using it that way would have cut tables for a square profile and based gas planning on that ahead of time, and then stuck to that profile's depth and time limits during the dive. So long as the computer keeps functioning, they may get some deco credit for the multilevel aspect of the dive conducted; if it fails, they can fall back to the tables and backup timer/depth gage. Hopefully nobody is just taking it out and saying 'gee, I'll just stay down until the ascent time seems like what I have enough gas for!'

 

[flots am]

This "feature" strikes me as 100% dangerous, doing a multi-stage deco dive, while flying your computer ??[/FONT][/COLOR]

It's actually worse than that. You need to switch the compter between tanks, using a method first seen on digital watches in the (80s?). You have to press a button to cycle between tanks, so not only do you have to make sure you're breathing the right tank, you need to make sure the computer knows which tank you're breathing on.

I have no idea why an AI computer needs a manual tank switch, but it does.

Hopefully nobody is just taking it out and saying 'gee, I'll just stay down until the ascent time seems like what I have enough gas for!'

I'm pretty sure it actually calculates that for you, but I've never given it the chance.