Dive Computer Brawl

More information on VPM and other decompression models:

VPM Decompression Site:
http://www.decompression.org/maiken/...m_Site_Map.htm

RBGM:
ftp://decompression.org/pub/Wienke/

Decompression information:
http://www.deepocean.net/

Decompression model comparisons:
ftp://decompression.org/pub/Maiken/
ftp://decompression.org/pub/

So that's VPM and NOT VPM-B?!

I got a response from Kevin about the future VR3s. When you buy a new VR3 you have a choice of the current decompression model (derivative of the Buhlmann ZH-L16 algorithm (includes microbubble stops), the exact same derivative used in the Proplanner decompression software) or VPM. If you upgrade, you can swap to VPM. So it appears that each VR3 will only have one onboard decompression model.

[Genesis]

But WHICH VPM?

Sounds like I'm out of the market on this one..... barf. (no simulator aka Proplanner for the VPM version either?!)

Somehow I need to be able to cut backup tables that MATCH the computer's profile! These guys have to understand that... no?

But WHICH VPM?

Sounds like I'm out of the market on this one..... barf. (no simulator aka Proplanner for the VPM version either?!)

Somehow I need to be able to cut backup tables that MATCH the computer's profile! These guys have to understand that... no?

I specifically asked about VPM, VPM-B or RBGM. The answer I got was VPM.

I agree - you need software to cut bailout tables which MATCH! The VR3 has the built in ability to cut future tables, biased upon your past dive, but I don't know if they are going to offer a VPM version of ProPlanner.

You can download VPM dive planning software, but I don't know if it will be an exact match for the VR3 (it should match, unless they modify VPM).

Trimix Computer Deco Models & Dive Planning Software:

VR3 - derivative of the Buhlmann ZH-L16 algorithm (includes microbubble stops) (free user’s manual at http://www.vr3.co.uk/vr3/main.php?content=manual )

HS Explorer – 10 decompression models (free user’s manual at http://www.hs-eng.com/manuals.html )
CF 0 = RBGM, F=100, -light work, warm water
CF 1 = RBGM, F=97, moderate work, cold water
CF 2 = RBGM, F=94, -heavy work, cold water
CF 3 = ZH-L16C Computer, Dr. Buhlmann’s Computer Model
CF 4 = ZH-L16C Computer, Asymmetric 118, RBGM light work, warm water
CF 5 = ZH-L16C Computer, Asymmetric 118, RBGM moderate work, warm water
CF 6 = ZH-L16C Computer, Asymmetric 118, RBGM heavy work, warm water
CF 7 = ZH-L16C Computer, Asymmetric 135, RBGM light work, cool water
CF 8 = ZH-L16C Computer, Asymmetric 135, RBGM moderate work, cold water
CF 9 = ZH-L16C Computer, Asymmetric 135, RBGM heavy work, cold water

NiTek He – modified Buhlmann ZH-L16 algorithm (free user’s manual at http://diverite.com/products/nitek/nitek_he.htm )
Conservatism and the NiTek He http://diverite.com/tec_talk/conservatism/index.htm

Dive planning software links:
http://www.therebreathersite.nl/soft...ebreathers.htm
http://www.hs-eng.com/simulators.html
http://www.vr3.co.uk/vr3/main.php?content=simulator

[Genesis]

I have VPlanner, which sports VPM capability as well.

For most of the "low-level" deco dives I do, VPM (orginal) is MORE conservative, in that it starts your stops deeper and takes longer to get you out.

However, this only holds for 'light' deco dives on Nitrox. The plans tend to converge around the 200' mark, and from there I understand that for really deep stuff VPM tends to underestimate the required deco.

Without knowing exactly what figures they've got in the algorythm, there's no way to know exactly what it produces, or if it can be reproduced with a table generator. If I can cut tables with VPlanner, and the VPM model they're using offers some form of 'tweaking' to get added conservatism if I'm not happy with the original, then it might be ok.

I still haven't gotten any response back whatsoever to my query to Delta-P on all of this however.... and buying a "serious" diving computer is not quite the same thing as buying a recreational computer for no-stop diving! I live with the "no real answers", or even what I consider misleading answers, from the rec computer folks - that simply won't do though when we're talking about a computer for this sort of use.....

Without knowing exactly what figures they've got in the algorythm, there's no way to know exactly what it produces, or if it can be reproduced with a table generator. If I can cut tables with VPlanner, and the VPM model they're using offers some form of 'tweaking' to get added conservatism if I'm not happy with the original, then it might be ok.

I still haven't gotten any response back whatsoever to my query to Delta-P on all of this however....

The beauty of VPM-B is that it shouldn't require tweaks (at least in theory).

I will ask Kevin some more questions. Hopefully, I will be able to get the answers for you.

[Genesis]

Thanks....

In theory VPM shouldn't require any "tweaks" either. In practice it bent a bunch of people, which is why there's a "B" model, and even the "B" model is known to be a bit squirelly for truly extreme exposures... 8)

The thing is, Vplanner allows you to set the algorythm parameters, provided you know what to set them to. Toying with those constants is unwise in the extreme unless you do know what you're doing, of course, but I find it difficult to believe that Delta-P doesn't know that VPM tends to underestimate deco for deep dives, which would mean they did something to get to an acceptable end result, given their intended customer for the unit.....

This is the unfortunate reality of 'black box" computers and deco software - nobody wants to talk about what's really behind the curtain, and lack of that knowledge can hurt you. Hopefully this is changing - even if just a bit.

The open nature of Vplanner and how it "exposes" these constants where you can see them, and leaves the actual code out in the open where you can look at it, is one of the attractions from my perspective. None of the other algos or computers out there - none - do this. Not even the RGBM stuff.

I still hold a glimmer of hope that the new VR3 model is a computer that I can actually have some confidence in purchasing for this sort of diving - but Delta-P has some work to do in terms of available support for those of us in the US, and also in terms of getting out there exactly what's available and when. For instance, their web page still is mute about the upcoming VPM algo availability - yet you said that it was available now if one ordered a new unit.

It is a bummer that the original algo is not still available with the VPM one though - that would be very nice, as it would permit simple comparisons.....

I've researched the math behind both VPM and RGBM beofre deciding which to trust. The mathematical modeling in VPM is much cleaner and progressive while Bruce relies on older analysis models, which is indeed typical for most physicists. I have found small glitches in RGBM, true it would take a lot of deco time for them to ever make their pressence known, but they are there. Have yet to be able to make the VPM model fail that way. My background btw is in theorectical mathematics, noncommutative algebras and quantum groups.

Wienke, B.R., Decompression Theory (ftp://decompression.org/pub/Wienke/ file named deco.pdf, pages 76 - 77):

“The RBGM has its roots in the earlier work of the Tiny Bubble Group at the University of Hawaii, drawing upon the so-called Varying Permeability Model (VPM) to multidiving, altitude, and mixed gas applications. While certainly not radical, the RBGM is both different and new on the diving scene. And not unexpectedly, the RBGM recovers the Haldane approach to decompression modeling in the limit of relatively safe (tolerably little) separated phase, with tolerably little a qualitative statement here.



Here, our intent is to (just) look at the underpinnings of both meter and diveware implementations of the RBGM algorithm, one with extended range of applicability based on simple dual phase principles. Haldane approaches have dominated decompression algorithms for a very long time, and the RBGM has been long in coming on the commercial scene. With recent technical diving interest in deep stop modeling, and concerns with repetitive diving in the recreational community, phase modeling is timely and pertinent. And, of course, since the RBGM extends the VPM, much of the following applies to the VPM directly.”

For instance, their web page still is mute about the upcoming VPM algo availability - yet you said that it was available now if one ordered a new unit

VPM isn't available for the VR3 yet, my words said the "future VR3s". The VR3 VPM release is “probably” 3 months away!

It is a bummer that the original algo is not still available with the VPM one though - that would be very nice, as it would permit simple comparisons.....

I agree, it would be very nice to have both (the original and VPM), but to do that you will need 2 VR3s.