Perdix CEIL, does it take into account the estimated off-gassing on ascent (with an assumed ascent rate)?

[NX32]

Shearwater came back to me, so we have a definite answer at last.

‘To summarize and address your current questions, if I understand correctly the answers would be B and 2. All decompression predictions, including calculation of the ceiling are based on an ascent rate of 10m / min. The only exception I believe is surface GF which is a tool that assumes an instantaneous ascent.
I hope this info helps and please have a great day.
Kind Regards’

That's amazing, thanks for asking them directly and reporting that.
While writing my own implementation (and a Perdix emulator based on that too) I had doubts as well. I mean I was 99.9% sure that's the case, because I've always observed my surfGF going a few % above my GFHi while still having a bit of NDL, but their answer makes it official and now I can sleep at peace!

 

[TooCold]

Interesting thread. Thanks @Rogerdd for "refreshing" it, and posting Shearwater's response. Until this thread I don't think it had occurred to me to wonder if the CEIL value depended on ascent rate, but now I'm interested and I don't find the Shearwater response sounding very confident or authoritative. Is it just me? It doesn't read like someone who really knows the implementation.

Ok, so all SW deco predictions are based on an assumed constant ascent rate. So if my DC says CEIL is Y metres it is telling me "if you immediately start ascending at a constant 10m/min you may ascend to Y metres, but no shallower" If I do ascend at that rate the ceiling shouldn't change during the ascent (or else the prediction was wrong). Similar for NDL: say I dive to 40m and hang out until NDL hits zero, then immediately start a 10m/min ascent direct to the surface, my NDL should remain at zero all the way to the surface. That doesn't jive with my (subjective) experience. What's wrong with my logic?

Beyond the basic questions of what Shearwater implements is the more interesting one (to me) of what should be done if you claim to implement ZH-L16c + GF. Several posts by several folks (@inquis, , @dmaziuk , @EFX, others...) Some have strongly stated that CEIL and/or NDL don't, can't or shouldn't incorporate offgassing during an assumed fixed ascent rate.

So, given the Shearwater response, are they not doing it right? Have any of you more knowledgable folks changed your mind? For myself, as someone only somewhat aquainted with the theory, I'm very interested to know.

 

[inquis]

It's an implementation choice that makes little practical difference -- there is no fine line between being safe by staying down X mins and bent at X+1 sec. Or safe ascending at 9.9 mpm vs bent at 10.1 mpm.

Similar for NDL: say I dive to 40m and hang out until NDL hits zero, then immediately start a 10m/min ascent direct to the surface, my NDL should remain at zero all the way to the surface.

Good luck instantly achieving and maintaining a 10.000 mpm ascent rate. But yes, there will be a point where it will say you have a mandatory stop (NDL is no longer shown), and it might not clear during the ascent. Computers HAVE to draw a fine line. Incidentally, if you do stop and it then clears, it will likely show NDL=99 because it is evaluating at that new depth.

 

[dmaziuk]

Beyond the basic questions of what Shearwater implements is the more interesting one (to me) of what should be done if you claim to implement ZH-L16c + GF. Several posts by several folks (@inquis, , @dmaziuk , @EFX, others...) have strongly stated that CEIL and/or NDL don't, can't or shouldn't incorporate offgassing during an assumed fixed ascent rate.

Buhlmann's formula does not "incorporate off-gassing". The implementation may or may not. In practice it probably only makes appreciable difference on deeper "bounce" dives and once you get used to the idea of "transient ceiling" and that -- gasp -- "light deco" won't kill you, you'd stop caring anyway.

 

[inquis]

So, given the Shearwater response, are they not doing it right?

Right/wrong is drawing a sharp line, but there may be other factors that shift the interpretation. For example, this observation by @dmaziuk early on seems to have nailed it:

If I were marketing a DC to no-stop divers who are not expected to understand how the computer works, I would want to implement that to avoid the "YOU'RE NDLS TO SHORT" screaming and wailing from the likes of ScubaLab "computer testers" and such.

 

[TooCold]

So thanks for the replies to my post @inquis, @dmaziuk. I have sheepishly edited my comments - look again to see what I've crossed out. After first reading this thread I had the impression that several had been arguing over the correctness of including/not including off-gassing during ascent in CEIL and NDL calculations. A quick review just now and I don't see that except for @EFX. The rest is mostly working towards a reasonable inference about what Shearwater could and/or is doing. Asking whether Shearwater is doing it right is down to just @EFX

Keep in mind that the NDL calculation for time remaining at depth knows nothing about ascent rates. It is, by definition, a calculation of time remaining at the current depth.

and maybe just curiosity if @VikingDives is unhappy:

I don't think that there's anything in a Perdix that accounts for ascent, as there's no way for the computer to determine what you're going to do, or how fast or slow your ascent will be.

More to the point, I wouldn't want a computer that accounted for my ascent - at least I don't think I would.

 

[TooCold]

Buhlmann's formula does not "incorporate off-gassing".

Please explain. Seems to me on-gassing and off-gassing as function of time, pressure and gas composition is precisely what Buhlmann and most (all?) other deco models model.

 

[dmaziuk]

Please explain. Seems to me on-gassing and off-gassing as function of time, pressure and gas composition is precisely what Buhlmann and most (all?) other deco models model.

CEIL = (P - a) * b

where P is the current tissue loading and a and b are the Buhlmann's coefficients: basically the reciprocals to the M-value line. Notice the conspicuous lack of "off-gassing" anywhere in there.

PS. The point of the discussion is whether the CEIL is "transient", i.e. it will clear by the time you get there due to off-gassing on ascent, or "hard": a mandatory deco stop. And whether the numbers shown to the user in the "NDL dive" mode should account for that off-gassing on ascent or not.

 

[TooCold]

Good luck instantly achieving and maintaining a 10.000 mpm ascent rate.

Ha ha (said he before the invention of emoticons). Instantly achieving (i.e. infinite acceleration)? - kinda difficult I agree; maintaining? Do it every dive - more or less. (By "mpm" you do mean millimetre per minute, right?

Notice the conspicuous lack of "off-gassing" anywhere in there.

Ah. I see. We haven't really been talking to each other. I was talking about the Buhlmann deco model; you were talking about that one formula.

 

[dmaziuk]

Ah. I see. We haven't really been talking to each other. I was talking about the Buhlmann deco model; you were talking about that one formula.

That formula is the Buhlmann deco model.

That's one of the reasons we like it for computer implementation: the KISS principle.