Tables & computers

[lowviz]

I can't see the math:

…//…I don't think we know ANYTHING about whether the models are simulations of what's going on in the human body. …//…

Big clue here…

…//…What we know is that the models create limits that result in an acceptably low incidence of symptomatic DCS.

Math models can also be created to just give us the results that we know are "correct" (by statistics, "exploded goats", etc).

…//… I can't see the math …//….

In school, we were all taught that the mathematics is a formal and precise way of describing something. So if you understand the math, you also get a better insight into what it is describing. Often true, certainly not always. Math is a tool that can be used in different ways.

Often,especially in modeling, you just look for a function that gives you what you want. I’ll offer an example. I wanted a function that will give me the end-of-dive group letter given the depth and time. So I searched for a function that would generate a number corresponding to a letter in the English language based on dive time and depth. Obviously this is a highly contrived situation, but no problem for "the math". "Understanding" this function is kind of pointless, it is just a function that gives me what I want, nothing more. Just for grins, here it is from one of my past posts:

Found a well-behaved function(well behaved for the intent of extrapolation) that predicts “NOAA No-Deco Air”end of dive Group Letter. (NOAA Diving Manual pg III-1) Quickest way to checkits utility is via Excel, shared below:

=(EXP((3.0343259 + (-15.355991/LN(DepthInFt))))*TimeInMinutes^0.72366692) InExcel, make a depth and time cell first, highlight “DepthInFt” and replace withthe depth cell, similar treatment for “TimeInMinutes“…



Nothing to understand other than an exponential function best fits the requirement. OK, OK, a bit of insight there...

Computers and tables find their base in an often identical "complex mathematical algorithm." The difference is only in the display, think of a computer as an animated cartoon of the "complex mathematical algorithm" and tables as a series of snapshots of the"complex mathematical algorithm." The math is exactly the same, or at least very close.

I too see dive computers and tables as VERY, VERY similar things. The DC is just able to chop your dive into smaller increments and give you a dive-time advantage of not having to be overly cautions. However, there are times when tables will better serve you and not just because they are more conservative than a dive computer. Picking the algorithm that best describes YOU is advantageous. This is not an elementary task.

Mark Powell’s book really is that good, I’ve re-read it several times.

 

[Pullmyfinger]

Agreed, Mark Powell's book IS that good. Easy to digest.

-Mitch

 

[bullshark]

right

 

[Paladin]

I use the tables exclusively. I prefer them. I have a computer, but have rarely used it and even then it was in gauge mode.

The tables were developed over more than a century of real world diving and not by some guy sitting at a computer. The Navy adopted the first dive tables in 1915 and continually updated them by experimentation over the decades.

I have been using the US Navy tables since 1965 and am completely comfortable with them. I am not comfortable relying on a computer, however.

 

[redacted]

A noob with Less than 20 dives and more than 1200 posts is goin' all bubbuh on a guy that has more 2300 dives.

A noob who thinks running his finger across a table full of inscrutable numbers of unknown origin (to you) is arguing with a professional software engineer by trade that writes decompression programs for over 10 years as a hobby. The clarity of the writing together with the exacting choice of words should be evidence enough for anyone with a will to see it, that someone has a pretty good idea what they are talking about.

No, there is no respect intended in your disagreement...



My assertion was 100% correct, and it's a shame you don't understand it. It is quite humorous that you immediately prove me right in the same sentence though.

While I agree that you are correct (nothing to do with credibility or dive count - just facts), I see nothing to be gained by your attack based on the poster's newness. I have also seen fairly experienced experts learn something from inexperienced noobs.

 

[Thalassamania]

But for a misunderstanding by a non diving engineer who did not get the differences between a diffusivity coefficient and a diffusion half time, we would all have been diving "computers" since the mid 1950s.

 

[drrich2]

Paladin:

The tables were developed over more than a century of real world diving and not by some guy sitting at a computer.

We drive cars designed by people sitting at computers, amongst many other things in our world than benefit from computer aided design. The people sitting at computers use real world data in their designs.

I have been using the US Navy tables since 1965 and am completely comfortable with them. I am not comfortable relying on a computer, however.

Despite the fact untold legions of recreational divers do fine diving computers. For that matter, consider how the dive hobby has grown over time, I wonder how many dives with tables vs. how many dives with computers have been done as of today, if we're talking about having a good track record. I suppose a person can be 'uncomfortable' about almost anything, but is there a logical basis for that? (I know some tec. divers prefer to us their computers in gauge mode; I'm not spinning computers as the only way to dive, just question why relying on this piece of gear is that different from relying on your SPG, regulator 1'rst stage, BCD, etc...).

Richard.

 

[Paladin]

Paladin:



We drive cars designed by people sitting at computers, amongst many other things in our world than benefit from computer aided design. The people sitting at computers use real world data in their designs.



Despite the fact untold legions of recreational divers do fine diving computers. For that matter, consider how the dive hobby has grown over time, I wonder how many dives with tables vs. how many dives with computers have been done as of today, if we're talking about having a good track record. I suppose a person can be 'uncomfortable' about almost anything, but is there a logical basis for that? (I know some tec. divers prefer to us their computers in gauge mode; I'm not spinning computers as the only way to dive, just question why relying on this piece of gear is that different from relying on your SPG, regulator 1'rst stage, BCD, etc...).

Richard.

I have no doubt that many divers use computers and that countless dives have been conducted with them. That is immaterial. I prefer using the tables. For the type of diving I do, there is nothing a computer can do for me that the tables can't. It's a personal choice and I'm perfectly happy with it.

 

[bullshark]

I would respectfully differ about everything in the above paragraph. First off, models can vary a LOT -- we have personal experience with a set of three dives where, out of five divers, four had no deco at all, and the fifth ended up with TWENTY MINUTES of mandatory deco. (We had four different methods of determining decompression on that dive; three tracked fairly well, and the fourth was a wild outlier.)

Why is it that wherever you go on the internet there is always some wag that will chime in with an unverifiable, unproveable story about some non-repeatable occurrence that is necessarily spewed on a thread to try and disprove something that is obviously true?

First, you just verified me by 4 out 5 instances. None of you got the bent, thus 5 out of 5 of you experienced safe dives based on well known model limits. Thousands of dives were conducted today in perfect safety relying on those well known model limits. Read Weinkes books if you don't believe me, and then go argue with him about the close correlation and the figures he presents there.

Second, the outlier was obviously operator error or involved a computer no longer in production. The first error that is most commonly observed is that the diver in question had a hockey-puck/ Pelagic/other Nitrox reset and failed to set his nitrox mix, and then wouldn't admit that to you. If it wasn't that, then the operator set his computer to absurd levels of conservatism which is possible in some (even recent) computers. Uwatec Smarts have Micro bubble level settings that will create level-stop demands that are indistinguishable from deco stops during the running dive, yet are in no way mandatory in as much as the computers will not lock up when skipped. Other examples of operator error are too numerous to detail here. RTFM.

Third, you could not reproduce that outcome using any recent computers and you know it. It doesn't happen without operator error on comparably set computers. So what was your point. I know exactly what it was. Wish granted.

Fourth, there was no point to the whole concoction unless to say that in your opinion all computers are radically different, there is no correlation between models and that you daily expect any and all models to run flagrantly out of control. Is that your point? Better tell Dr. Wienke, correlation among the models is the underpinning of the design of his Thermodynamic EOS. He's wrong. You're right. Got it. Check.

Second, I don't think we know ANYTHING about whether the models are simulations of what's going on in the human body. What we know is that the models create limits that result in an acceptably low incidence of symptomatic DCS.

I said no such thing. Why is it that the guy who jumps in to correct everyone (Mister Corrector guy who never originates anything) all the time continually misreads and misunderstands what he corrects. Read what I said again.

"reliable simulations of human physiological no-decompression limits in a recreational setting and well beyond."

Please respond with your explanation of where I indicated or used the word "human body". I even began the sentence in question with the warning, IN BOLD, declaring the theoretical posture of the models, yet in your zeal to spank someone (a decidedly non-moderator activity) you failed to comprehend the words.

How is it that you can ignore all this, stuff words in my mouth that I didn't say, reinterpret what I did say, and then correct it, without adding any plausible value to the proposition:
The models work. The models are reliable. Thousands of hours today, thousand of times today, yesterday, tomorrow and for years and eons yet to come.

You did not read the words. Shame on you.

 

[DiveNav]