Suggestions for getting my first dive computer

[KenGordon]

Anyway, just because you may have taken a long time to get to where you could stay down long enough to be limited by NDL does not mean that anything like "most" people are like that. I was consistently limited by my NDL (using Nitrox and a liberal computer) within 6 months of getting OW certification. And my SAC is nothing special. Lots of "new" divers have a SAC as good or better than mine. I completely disagree with your statement that [new divers] "will probably find that other factors limit the dive before the computer." But, my disagreement is based on my feeling that anyone with less than 50 dives could legitimately be regarded as a "new" diver. If you're talking about people with less than 10 dives since OW, okay, maybe. But, why would you recommend someone buy a computer just based on the needs of their first 10 dives?

Ah, a personal attack. Basically you are saying my gas consumption is so poor I cannot manage a dive that is limited by NDL.

I was trained from the start to do deco dives, so this perverse obsession with NDLs has passed me by. All my dives are gas limited, temperature limited, bladder limited or limited by commitment to stops, I only care about NDL in the context of a buddy who is not prepared or qualified to do stops,

 

[scubadada]

I was trained from the start to do deco dives, so this perverse obsession with NDLs has passed me by.

Why have you spent so much time and energy advising newer divers on SB to purchase conservative computers?

 

[KenGordon]

What do you mean there is no planner for a Shearwater? You can download Subsurface for free and plan away. Or are you suggesting that maybe Shearwater doesn't use a strict Buhlmann w/GF implementation and, instead, maybe has something proprietary in their algorithm implementation that makes it not behave just like any number of plannings applications (e.g. Subsurface, Mult Deco, etc.) work?
.

Certainly the OSTC does not implement GF as per the writings of Eric Baker. I have posted their code credited in the Subsurface sources which has a hack to use the deepest ceiling for the GF low start depth. The deepest ceiling is not the first stop depth, time is spent ascending.
I guess that the Shearwater has to do something similar otherwise all sorts of amusements arise from slow ascent rates.

Subsurface, Multideco etc do not emulate a Shearwater. They do what they do, as far as I can tell MultiDeco is pretty faithful to Eric Baker's code. Given that we do not have the code that drives the Shearwater or a planner to play 'what if' regarding repeat dives etc we cannot say how it will behave for the well known (and taught) bad practices such as saw tooth profiles. As such a new diver may be better off with a tool which does those things automatically.

 

[stuartv]

A GF computer which allowes a 95/95 setting is more aggressive than a Puck i think you will agree. Part of that means that it has a longer NDL time for any given depth. Do you deny that a longer time at a given depth is riskier than a shorter time?

Nope. Do you have any evidence that the difference in time at depth between the NDLs of these computers makes any statistically meaningful difference in incidence of DCS?

We're back to the question of distance from the precipice. Is the Puck 2m away or 2km away? The most aggressive SW setting in Rec mode is GF45/95. Is that 1m away? 1.5m? 1km? 1.5km?

There are differences between very liberal and very conservative computers, after 2 or 3 dives, where the NDL can be different by 40 or 50 minutes. Is the liberal computer, with its really long NDL, in that case, likely to get the diver bent? If not, then doesn't that suggest to you that the conservative computer is several kilometers from the precipice (in that case)? I mean, there are lots of people diving DSAT computers on liveaboards, diving multiple times per day, for multiple days in a row. And we certainly don't seem to be getting bombarded with stories of those people getting bent and ruining their trips for all the Cressi and Suunto divers aboard.

Look at this test:

http://ads.bonniercorp.com/scuba/PDF/ScubaLab-Computer-Test-September-2014-data.pdf

Look at some the differences in NDLs on dives 3 and 4. There are some big differences between some of the computers. And if the liberal ones of those computers don't even have anecdotal evidence to support the idea that they get people bent more often than the conservative computers, then it seems to me that those conservative computers must be REALLY far from the precipice. And that the differences of just a few minutes, for more normal 2-dive-a-day charter diving, is really negligible in terms of the chances of getting bent. BUT, those few extra minutes a liberal computer gives you are very NOT negligible (to some of us, anyway) when it's the difference between spending 5 minutes on a deep wreck versus 10 minutes while on vacation somewhere.

 

[KenGordon]

Why have you spent so much time and energy advising newer divers on SB to purchase conservative computers?

Because I genuinely think those thelling them to buy aggressively computers are hurting people. Given a significant population diving to longer NDLs some will get bent. Something like one in five of these new divers will have a PFO and so eventually they might get bitten by diving to longer NDLs.

Also the obsession with NDLs is bogus, the proper thing is to just do (short) deco dives. The training that means that people cannot do that is inadequate. All these inadequate divers needing hand holding leads to a poorer experience for those that are not trained to the lowest common denominator when they visit holiday destinations.

 

[stuartv]

Certainly the OSTC does not implement GF as per the writings of Eric Baker. I have posted their code credited in the Subsurface sources which has a hack to use the deepest ceiling for the GF low start depth. The deepest ceiling is not the first stop depth, time is spent ascending.
I guess that the Shearwater has to do something similar otherwise all sorts of amusements arise from slow ascent rates.

Subsurface, Multideco etc do not emulate a Shearwater. They do what they do, as far as I can tell MultiDeco is pretty faithful to Eric Baker's code. Given that we do not have the code that drives the Shearwater or a planner to play 'what if' regarding repeat dives etc we cannot say how it will behave for the well known (and taught) bad practices such as saw tooth profiles. As such a new diver may be better off with a tool which does those things automatically.

Are you saying that there are computers which do things automatically that the Shearwater does not (regarding NDL calcs)? What are those things?

And are you saying that you know for a fact that the planner tool for a Puck (or pick your rec computer) is implemented in a way that exactly matches how the Puck's deco algorithm (not it's built-in planner) actually works when you dive it?

 

[KenGordon]

We're back to the question of distance from the precipice. Is the Puck 2m away or 2km away? The most aggressive SW setting in Rec mode is GF45/95. Is that 1m away? 1.5m? 1km? 1.5km?
.

There is no precipice, just a long and gentle slope. You know the cliche about there being no bright line through the shades of grey, There are just probabilities. No black and white. I have seen dumps of profiles where divers surfaces with excessive GF, like 200, and were fine. I have also met people who were bent on OW lessons. The further up that slope the less likely we are to need a chamber visit.

 

[KenGordon]

Are you saying that there are computers which do things automatically that the Shearwater does not (regarding NDL calcs)? What are those things?

And are you saying that you know for a fact that the planner tool for a Puck (or pick your rec computer) is implemented in a way that exactly matches how the Puck's deco algorithm (not it's built-in planner) actually works when you dive it?

Not for a fact, I just bet my life on that when I plan dives. I have not read the code for either my computer or it's planner. However I do notice that the plans change when I select different models and settings for the range of supported computers.

Do you know for a fact that Multideco and your computer generate the same plan?

 

[KenGordon]

Are you saying that there are computers which do things automatically that the Shearwater does not (regarding NDL calcs)? What are those things?

All those things that are used to complain about the unpredictable behaviour of the likes of Suunto and Marss computers. Excessive ascent rates, repetitive diving, etc. Those have no place in a GF implementation, Eric Baker does not mention them, the original Buhlmann implementations do not do anything other than consider the disolved gas tensions, how you get those is just time and depth, not the shape of the dive.

Mares explicitly say they change the M values to account for higher risk behaviours.

 

[stuartv]

Because I genuinely think those thelling them to buy aggressively computers are hurting people.

Can please post some citations of your evidence for this?

Also, for the record, I have never told anyone to buy an aggressive computer. I have said (repeatedly) that it does people a disservice to tell them that any aspect of their computer is something they should not worry about. Telling people "don't worry about that. It won't matter to you" or "you aren't smart enough to understand that bit" is just wrong. Give people the info and let them decide for themselves what is important to them.

The algorithm may not matter - to YOU. AI may not be useful - to YOU. AI may not be worth the price premium - to YOU. But, I think is simply wrong to make those judgments and then foist them onto other people. Especially new divers who don't have the experience to appreciate that some "expert's" statement that "AI is not worth it" simply means that it is not worth the price, based on that person's needs and budget. It's too easy for them to take a simple statement like that as "there is no value to anyone in having it" - which is patently untrue.

You say "the algorithm doesn't matter" and they don't appreciate that it doesn't matter to you because you are trained for deco, and/or it took you 5 years to get to the point of being able to stay down to your NDL, or whatever your various reasons. You don't know the OP. You don't know that your reasons fit them. If you have facts to support your statement that the algorithm doesn't matter, then share them with us all. If it's just that you have some "feeling" that it's significantly more likely to get them bent, then you could at least do them and the rest of us the courtesy of making it clear that saying the algorithm doesn't matter is based on no facts and just a feeling that you have. A hunch. Something you "think". Whatever. If Simon Mitchell "thinks" something about deco. Or David Doolette. Or a list of others, I would take that seriously. If your credentials are on par, that what you "think" should be taken as legitimate evidence by a new diver, with no facts cited to support it, then please share them. Otherwise, I will continue to try and share what facts and hard data I can, so that new divers can make their own informed choices, instead of just taking my or anyone else's word for what matters (or should matter) to them.