Should Shearwater add Air Integration to its computers?

[stuartv]

As I said before: I don't feel the need for AI. In my case: I don't care how I can check my air supply, to me it's important I can. SPG is cheap and does what it's supposed to do. My experience with Suunto AI is not that good that I want to spend a lot of money on AI. Therefore: I don't want AI right now. If it becomes as cheap and reliable as an SPG, I would probably consider it. Not now.

Now THAT makes perfect sense. No debate from me, thought I do question whether it might not already be the case that at least some manufacturers are making AI that is reliable as an SPG. Maybe Suunto isn't, but my Oceanic has been, so far, and it seems like a lot of people say their Oceanic has been as well.

 

[muzikbiz22]

Backup camera would be an even better analogy.

Ha, now there's an interesting one. I'm for AI, I like it, but I don't need it. There's that. Now, this whole backup camera thing on a car being mandatory ,that's where I'll draw the line. My hat goes off to the amazing salesperson that marched up to capitol hill, made a sales presentation to the nations lawmakers and said "Yeah, this camera thing that goes on the back of a car to help drive backwards ? I think it's necessary for all drivers, and you should make it the law, that way I can sell a $**t-load" ! Congrats, who ever you are !!

 

[stuartv]

I asked Andy this over in another thread, but it seems relevant to pose the question here:

- What did you use for a depth gauge 30 or 40 years ago?

- Which is more important to you? Knowing your depth or your tank pressure? I mean, many of you are so experienced and plan your dives so well that you don't really need to look at your pressure gauge at all, right?

- Do you carry a mechanical depth gauge now? If you are an instructor, do you require your students to have a mechanical depth gauge for backup or is it okay to have 2 electronic depth gauges and no mechanical?

And, lastly, if you do not carry a mechanical depth gauge, what is your rationale for trusting the depth readout on your computer, but not trusting a pressure readout on your computer? (obviously, this is only for the folks opposed to AI - and if your only thing against AI is the cost, no need to reply)

 

[Dhboner]

Most serious divers will have at least two independent ways to measure time and depth...computer/bottom timer/depth gauge/wrist watch. These devices, though not infallible, do not rely on a transmitter to work properly and they are not known to fail regularly, if at all. The issue is not now, nor has it ever been not trusting the pressure readout, it is simply that the transmitter introduces a vulnerable failure point without providing any measurable advantage over a mechanical SPG. I don' think anyone is opposed to AI so much as it attempts to solve a problem that doesn't exist while introducing a new failure point in the process. If you want to use an AI equipped computer, knock yourself out. I am not going to try to convince you that you are wrong.

 

[stuartv]

Most serious divers will have at least two independent ways to measure time and depth...computer/bottom timer/depth gauge/wrist watch. These devices, though not infallible, do not rely on a transmitter to work properly and they are not known to fail regularly, if at all. The issue is not now, nor has it ever been not trusting the pressure readout, it is simply that the transmitter introduces a vulnerable failure point without providing any measurable advantage over a mechanical SPG. I don' think anyone is opposed to AI so much as it attempts to solve a problem that doesn't exist while introducing a new failure point in the process. If you want to use an AI equipped computer, knock yourself out. I am not going to try to convince you that you are wrong.

If you replace a mechanical SPG with an electronic one, haven't you reduced your failure points by quite a few?

I have asked this before and gotten no answer. Is anyone aware of a case where an AI transmitter failed during a dive and it resulted in a loss of gas?

If so, was the failure something other than getting unscrewed or blowing out the O-ring?

If it was getting unscrewed or blowing out the O-ring, aren't mechanical SPGs equally vulnerable to that?

It seems to be commonly accepted that it has happened upon occasion that a mechanical SPG (including the HP hose in this statement) has failed during a dive before, resulting in the diver losing some amount of gas. If mechanical SPGs are known for that (rarely) and electronic SPGs are not known for that ever, then doesn't that suggest that mechanical SPGs are more failure prone in the only way that is a serious safety issue?

Isn't a mechanical SPG, even one that has good O-rings and a good HP hose, more likely to "stick" and give a falsely high reading? Isn't it know that they sometimes do that? Does anyone know of a case where an electronic gauge has given a falsely high reading?

I suppose a false high reading would actually be the only other serious safety issue with SPGs, right?

Reading high or losing gas. Reading low or not reading at all is not going to kill anyone, right? Either of those just means you end your dive early.

I'm asking sincerely. I have not been diving nearly as long as most of y'all, so just because I haven't heard of it doesn't mean it's not a real issue, so I'd like to know.

If you do know of any electronic SPG failures of the type I'm asking about (losing gas or reading high), can you relate how old the unit was? I'm curious if it was a problem with anything remotely current, or if it was a failure of a unit that came out in the early days of AI where it would be expected to be less reliable - just like computers were not so reliable when they first came out.

 

[pedro_rj]

Guys, no offense but don't you realize this is all just going in circles?

The exact same arguments and rebuffs have been posted over and over. I think everyone made their respective points pretty clear by now:

1) part of tech community don't want shearwater messing with technology perceived as "recreational"/irrelevant to them. Failure points, masking poor gas management, overstretching resources etc
2) part of (?) community want the best computer with AI for a number of reasons. Easier readout, less task load for certain types of diving, doubling as "rec"/"tech" computer etc

That's it, why don't we go discuss split fins for a change?

 

[dmaziuk]

If you replace a mechanical SPG with an electronic one, haven't you reduced your failure points by quite a few?

No. In both cases you need an analog pressure sensor. Mechanical SPG has it connected directly to the needle. The digital one has it connected to a Digital-Analog Converter and then a bunch of other electronics all the way to the display. Plus a power source. I.e. there is more "stuff" in the digital version.

Isn't a mechanical SPG, even one that has good O-rings and a good HP hose, more likely to "stick" and give a falsely high reading?

Potentially, yes. Moreover, non-linear response is potentially more likely with the mechanical SPG. Plus, as you said, more hose and pressurised couplings mean more places to potentially spew bubbles. In reality the pressure sensor in a WAI transmitter may be just as bad as the analog one. There is no needle that can get physically stuck, but there is the DAC that can be bad instead.

Overall, I'd say at the current level of technology, the "mechanical SPG more reliable" argument can be safely ignored.

 

[stuartv]

Overall, I'd say at the current level of technology, the "mechanical SPG more reliable" argument can be safely ignored.

That's how it seems to me, too. But, what about the "mechanical SPG is safer" argument? It seems to me that the wireless AI is actually safer (as in, less likely to spew bubbles and less likely to give a reading that is way high but still believable).

 

[stuartv]

No. In both cases you need an analog pressure sensor. Mechanical SPG has it connected directly to the needle. The digital one has it connected to a Digital-Analog Converter and then a bunch of other electronics all the way to the display. Plus a power source. I.e. there is more "stuff" in the digital version.

ps. I don't really buy into looking at it this way. Is a computer a million transistors and thus a million failure points? Or is it one failure point? I view it as one. Similarly, I view the AI transmitter as 1 electronic failure point (or 2, if you count the computer that is receiving it, which I wouldn't for purposes of this discussion. If the computer fails, dive over, AI or SPG), plus the one threaded connection and O-ring.

And I don't know how they work, but I would (warning: total speculation alert) guess that AI transmitters have some kind of solid state pressure transducer, versus the Bourdon tube/mechanical setup that an SPG has. Comparing those, I would think that the pressure transducer is inherently many less failure points than a Bourdon tube gauge. Thus why most everyone uses digital depth gauges now, instead of mechanical depth gauges.

Having a depth gauge integrated into your computer is advantageous over a discrete depth gauge because it allows the computer to do deco calculations. But, even if one were using a discrete depth gauge, wouldn't most people choose an electronic one over a mechanical one? Don't most people diving with just bottom timers use ones that have an electronic depth gauge?

I conclude that if a mechanical pressure gauge the reads out your depth were safer or more reliable than an electronic one, then we would/should all be using mechanical depth gauges. Most don't and are considered to be diving safely. So, why would anyone think that a mechanical pressure gauge that reads cylinder pressure by inherently safer or more reliable than an electronic one? The only difference is that the tank pressure gauge (that we're talking about) uses wireless to communicate the pressure. That does make it somewhat less reliable. But, it makes it NO less safe. The wireless aspect of it won't cause you to lose gas and it won't cause you to get a false reading. And whereas you NEED reliable depth readings, all the experts seem to feel like they barely need an SPG at all, so having it be somewhat less reliable than a mechanical gauge seems like no reason at all to shun it. Particularly since a properly maintained transmitter (meaning, it has a good battery) seems to only suffer in reliability in that it may take a second or two to give you a readout sometimes.

 

[dmaziuk]

It seems to me that the wireless AI is actually safer (as in, less likely to spew bubbles and less likely to give a reading that is way high but still believable).

It comes with different failure modes but I don't think it's "safer" or "less safe" for any practical meaning of the word. As in I don't consider 1/10,000 vs 1/10,001 a practical difference -- someone else's MMV.

Edit: reject rate on microchips depends mostly on the chip and its intended purpose. I imagine the ASIC inside a WAI transmitter shouldn't be operating anywhere where it's likely to cook itself, so if it didn't fail on the first power-on/power-off, it should be good. Unlike, say computer RAM chips that can croak under load. Or CPUs that are so complex any number of defects may slip past testing.