Shearwater for a newbie ?

[dmaziuk]

@dmaziuk

Shearwater uses open versions of VPM and Buhlmann ZHL. Unadulterated, open source, you can model exactly what that computer will do with any software out there.

Models, yes. Software: no.

A fortran program (Erik Baker's) may be considered software if it runs in the computer as is. If you have to reimplement it in a programming language your computer understands, it's not software anymore. At best it's "an algorithm". And a textual description of the model with some formulae thrown in is not even that. There is an unknown distance between what you're talking about and what is actually running inside any given dive computer.

From an instructor standpoint, that is a safety thing and I refuse to dive any computer that can lock me out when I'm teaching.

I get it, but there are certain pesky laws of physics that don't change when you're teaching as an instructor. As a simple contrived example, if your CPU has 8-bit registers, it can only count to 255. It can't tell 256 from 1001, so any gas loading over 255 it's calculations are totally screwed. As a computer programmer I'm telling you, there is a point where it's gotta go titsup. Part of the definition of "proprietary" is we don't know what/where it is.

 

[Barefootdog]

A less expensive DC you get today gets you in the water diving/gaining experience. That's what's important. Some of the less expensive units are very capable for the recreational/advanced recreational diver.

Possibly it then becomes a backup unit when/if you decide to get the more capable/technical/expensive DC.

More important are the base features of what you buy today...flexibility of being loaded with multiple algorithms...mixed gas/air/gauge modes..Lots of capabilities are built into lower cost units today.

FWIW my first dive computer(s) were a set of Orca/Delphi/Phoenix units from the early 90's(?).
Very simple units compared to what's available for $200 today but they both still function perfectly for simple recreational dives.

Quit grinding on the decision and get yourself in the water. $200 for a dive computer is chump change...Capiche?

 

[tbone1004]

@dmaziuk so we know the limits of the algorithm if it is unadulterated. With the Shearwater, yeah, you probably don't know the full limit of how far it can go, but you're going to be dead long before the computer runs out of the ability to track you.
He probably mistakenly used software instead of algorithm. All of their software is going to be proprietary as it isn't open source, but the algorithms being proprietary are scary as hell to me.
The Shearwater runs a known algorithm that is not proprietary, that is important.

 

[rongoodman]

I don't think "proprietary" when applied to software means what you think it means.

I suppose if you want to pick nits I should have said algorithm instead of software. My point stands, however. Buhmann and VPM are widely understood and implemented. Suunto "RGBM", not so much. There was an aftermarket version of RGBM for the Liquivision Xeo which claimed to be a full implementation, but Liquivison is no longer selling computers.

 

[JMBL]

I'm glad there are some French people too! I am mostly diving in Marseille and Port-Cros, and quarries around Lyon. What about you?

Been diving the Med mostly : Port Cros and Porquerolles are my favorite of course, but I dive off the coast near Menton also. Love the Atlantic of the coast of Arcachon, but it's complicated : deep ocean, cold, dark, trimix needed and top weather conditions (forfeited Marseille and Cassis some years because of the damned Mistral btw). Also cave dive in le Lot.

Cheers.

 

[dmaziuk]

I suppose if you want to pick nits I should have said algorithm instead of software. My point stands, however. Buhmann and VPM are widely understood and implemented.

If I wanted to pick nits, I'd say "model". Where I work an algorithm is a sequence of steps to be performed to produce an output. Over here "algorithm" is normally used to mean "model" and I can use it in that sense too. Regardless, you're probably much smarter than I am because when I look at various fudge constants and coefficients in there, "understood" is not exactly what I would call it.

@tbone1004: I'm sure you know a few decades ago there was a whole lot of programmers who were certain their software will be long dead before year 2000 comes along and their date calculations start counting current year 00 as being a century before the previous year: 99. They thought wrong. In practice it turned out to be not much of problem, by the time it actually happened, just like a petrel running out of numbers won't be a problem to its dead user. But personally I prefer knowing that when it "doesn't compute", my computer will just stop and ask for a cold reset. That's simple stupid and IME simple stupid works much more often than any alternative.

Bottom line is just because you know what the model is, doesn't mean you know what/where the corner cases for any given computer are, when you might run into one, and what's gonna happen when you do. And if the rumour about OSTC not being OS any longer is true, that is the case of every dive computer on the market, not just low-end "RGBM" ones.

 

[wetb4igetinthewater]

I don't have the urge to replace it.

You should just try a Perdix for a bit. You'll see the difference and might actually change your mind and upgrade. It is that good.

 

[KenGordon]

I
If you stupidly miss many stops yes I expect it will give a years deco, trying to help, if you don't have the gas, either way you are bent or dead
).

If you skip stops on ZHL16 your deco time will be decreased by the algorithm. This is because the shallower you are the faster the off gassing.

It will not make up for the missed stops.

 

[RainPilot]

If you skip stops on ZHL16 your deco time will be decreased by the algorithm. This is because the shallower you are the faster the off gassing.

It will not make up for the missed stops.

I disagree. Let's say you are diving 50/80. Your first stop is where you get to GF50. Now you skip the stop, ascending to e.g. GF 70. The computer SHOULD keep you there until you are below the initial calculated GF line and then resume the ascent.
The other option of recalculating a GF line from 50-80 when you are at 70 would be impossible? I believe that it will increase your later stops if you skip early stops in order to get back below the line.

Whether that works out in practice due to suboptimal ascent rates etc is anyone's guess.

 

[KentB]

I was in the same boat as you as a new diver when my geo 2 crapped out. I bit the bullet and picked up a perdix AI- I have only used it in rec mode but the screen alone sold me on it. The navigation of it is and simplicity to use is just phenomenal.