Identical gases

[Dirk Hohndel]

"Robert C. Helling" <helling at lmu.de> writes:

> Hi,
>
> let me open another can of worms: Currently, we identify cylinders by 
> their gas content. This is problematic if one has more than one cylinder 
> with a particular gas mix (like for bottom stages or in a sidemount 
> configuration). Will we ever support this or shall we tell users that such 
> cylinders should be combined to one big virtual cylinder?

One of the things where the designs of different dive computers (and
lack of telepathic skills on my part) have lead to bugs and confusion in
the past is the difference between

a) what gas am I breathing
b) what tank is that gas from
c) what tank sensor is active

To the diver (and the deco calculation) (a) is really what matters.
Libdivecomputer with its gas change events only gives you that
information. Our Uemis support could give you all three but because of
the fact that everywhere else we just use what libdc gives us, we only
get (a) and (c).

So if you have two tanks with identical gas, libdivecomputer doesn't
even give you the ability to distinguish which one you switched to
(that's because the current design of libdc is based on what Suunto
computers give you - which is not a "switched to gas number n" but
instead "switched to the gas with o2=xx" (or in the HelO2 "switched to
the gas with o2=xx and he=yy").

The easy way to work around this is to cheat. If you have to tanks with
air have one of them be a EAN22... not perfect, but does the trick.

/D