Mares M1 RGBM

[Genesis]

BRW strongly implied that the new M1 RGBM was a "true" RGBM computer. It pretty clearly is not from what I've had described to me as its behavior, but I have no way to verify that.

The fuzzy nonsense needs to stop. Its marketing-speak and it is, IMHO, harmful instead of helpful.

A PRECISE definition of what's in what make and model of computer is what's needed here, along with whatever algorythm "tweaks" have been made. Right now one of the issues that comes up repeatedly on another list regarding deco software (for computers that are dry, not dive computers) is that several versions of commercial RGBM implementations exist and all of them give different results for the same input! Attempts to find out what is "correct", per the model, have been unsuccessful.

At least with VPlanner I know that if I run a profile on VPM-B at my house and then again on a friend's computer on a dive boat, I'll get the same schedule assuming that I select the defaults (or the options screen matches what I have at home.) If I grab the various "pure RGBM" software implementations out there I have absolutely no reason to believe this will be the case.

This makes the claim of "pure" RGBM more than a bit suspicious at best. It also means that I cannot build up a trust in RGBM over time, but rather only of one implementation of RGBM at a time.

The latter effectively destroys RGBM as a "selling feature" for me in a decompression program/computer; the "branding" of RGBM is, in the present state of things, worthless to me and appears to be nothing more than pure marketing-speak.

If there is no "default" setting for this algorythm that is "approved" by the author, then what is it, really? If everyone is tweaking it in different ways but nobody (including the author!) knows what those tweaks are, or how to "align" all these implementations to a common "best" set of data, then what am I buying? A guess?

Well sure, we know all deco software (including recreational computers) are indeed a guess. But, to blatently steal one of Mr. Spock's lines, should we not make the best guess that we can?

In other to do that, do we not need to know exactly what defines the "best guess", what deviations from a base model's parameters have been made - and why?

I argue that we do, and that until we know this claims that "RGBM" is in some meter or piece of software are, to me, mean only one thing - I'm paying more money to someone, but have no actual validation of anything, since I have no way to know WHAT is in the meter or software that I just bought or how it has been altered.

 

[Jai Bar]

I think that you misunderstood something about computer softwares (PC's)...
Ther might be a slight difference between results if you compare two (or more) dive-planner softwares. There might be some differences, some parameters that in one software can be modified and in the other not. However, if you run the same software in 2, 3 or 100 different PC's you'll get the same results- there is no random or monte-carlo guessing in decompression softwares.. Same input- same output It is not something unique to V-planner you can test it with your friend's computer with any software.

 

[Genesis]

Its not a slight difference between the several commercial packages that all claim to be "RGBM".

Look, if I load four Vplanner versions on four computers, I expect reasonably close results. Sure, there may be differences as the algorythm is tweaked, but I can go look at the release notes and figure out why.

If I load 4 different pieces of software all of which claim to be "full up" RGBM, I get 4 different results, and not "slightly" different either. If the algorythm is the same, the output should be the same. But its not, which means that there are significant differences.

Why? That's the question, and the answers are not forthcoming when the question is asked.

 

[Jai Bar]

Which softwares you've been comparing? Can you show the outputs? ("Print Screen" is enough). BTW, which softwares are "full RGBM" today?

BTW, Vplanner is not based on RGBM, it uses a varying permeability model. Not the same thing so don't expect it to yield same results as, say, Abyss running on RGBM, or GAPrgbm.

 

[Genesis]

There have been a number of comparisons posted on TDS and also on Ross's mailing list..... including one guy who did a REAL extensive evaluation of several RGBM packages along with VPM, then reduced all the data to graphical form.

The RGBM stuff was literally all over the board.

Yes, I know that VPM-B is not RGBM...

 

[leadweight]

On my recenttrip to Ambergris Caye in Belize, I got to put the Mares M1 RGBM through its paces.

After 10 dives the left button quit working. Fortunately the computer remained useful underwater, although the logbook and other features were not available. It has since been returned and I expect a refund soon.

As compared to my Aladin Pro Nitrox no-stop times on the first dive of the day were the same. On the second dive of the day, the M! would show about 3 minutes less do stop time when getting near the end of the dive if the surface interval was 90 minutes. When the surface interval was about 70 minutes the difference was more like 5 minutes less no stop time. However, the Aladin would respond more to reductions in depth. In fact, as the M1 no stop time got down to 4 minutes, it would take an even larger reduction in depth to extend no stop time.

These dives were about 80 foot max for about 10 minutes on the first dive with another 40 minutes at around 50 to 55 feet. Second dive was a 60 foot max, but mostly around 50 to 55 feet. Total duration of each with safety stops was around 55 minutes.

Two other features of the M1 botherd me. The logbook will record a "deco" entry for dives that passed the no stop limits. I really would not want to have to show this to certain PITA boat captains. Also, the pin that holds the buckle on the wrist strap can work its way out and cause a loss of the instrument. I wonder what they were thinking when they designed that.

None of the dives that week were deep enough to invoke the deep stop feature. Max depth was 88 feet for the week.

As compared to the Aladin, total desat times were much shorter. Because the Aladin has far fewer features, it is easier to access certain features that are used frequently such as changing FO2 or surface inerval time in the logbook. That is a design trade-off.

Any questions?

 

[Genesis]

Thanks - nice mini-review.... Bummer on the button though!

 

[DeepScuba]

Don't mean to be a GUE drum beater here, as I am on record as being "pro" computer for recreational diving.........

but since there's so much ambiguity(sp) with all these models why on earth would anyone waste lots of $$$ on these dive computers in the first place?

I have a many year old XTC100, no gauge mode, no gas switching (Does do Nitrox) capabilities. She's a bare-bones nitrox computer.

The point is, since it's evident no-one knows WHAT exactly they're diving, and since we all seem to be using deep stops whether credited or not..........I beg to ask. Why bother with the hub-bub, new fandangled $$$$ 'puters?

The funny thing that I find, for example my last dive. 180ft for 11mins (It was damn cold, sorry about the wacko B/T). I set my puter on "air" (That's what I was diving). I ran deco planner, set the deep stops in, printed out the schedule and dove it. I did the dive "as air" as far as my puter knew, and it cleared me within 2 mins that deco planner did with 80% on the deco stops. (Obviously air for the deeper ones ) It did put me a far ways into deco, but it never shut me out, and it cleared!

The above paragraph isn't really a point in all this, just some info I thought was funny in a wierd kind of way. I think I have Deco planner set on "Wennie" mode.

I guess what I'm saying is, don't waste the money on high $$$ puters, dive your plan, dive YOUR puter, and stick in deep stops..........what's the problem?

P.S. Used a Mares M1 2 months ago. (Actually my wife did, as a requirement for some dives) We thought it sucked.

sorry for the short and undetailed review.

 

[leadweight]

Hey DeepScuba

So far as cost is concerned, the M1 at $240 is pretty close to the very lowest in a nitrox computer. I think you can get a Tusa IQ600 for $200, and that would be rock bottom.

It is somewhat frustrating that there is such a wide range of no stop dive time allowed by different brands of computers. None the less, the accident rate for recreational divers is very low. This would suggest that there is a lot of headroom built in. There are also significant differences in required stop times once the no stop limits are exceeded. Unfortunately, the accident rates for technical divers are much higher. Some attribute this to a lack of enough data points. In the last DAN report technical divers were (and I can't remember the exact number) something like 15 or 30 times more likely to get bent. While a diver may not know exactly what is going on inside the meter, certain characteristics are well known, such as Suunto with its very short no stop times when surface intervals are short and those made by Pelagic where no stop times are long but very long deco stops are required once the no stop limit is passed.

Then we can move on to the other issues like lock outs or the appearance that meters are being designed by lawyers and marketing guys rather than by divers. Even major magazine computer tests look more like they being edited by the legal department. Diver's latest test (at least the web version) is devoid of any hard numerical data. Rodale's has never published test results for performance beyond the no stop limits. If you look at Rodale's tests over the years you might notice that their *****ing about batteries that have to be replaced by the manufacturer and wet finger contacts has vanished. they must have needed the advertising revenue from Scubapro.

 

[supervisor]

....... Also, the pin that holds the buckle on the wrist strap can work its way out and cause a loss of the instrument. I wonder what they were thinking when they designed that....
Any questions?

That sucks big time I know it. My old tutor had the same problem. I guess Mares is not learning from design mistakes in the past.