Will Air Integration in dive computers replace the SPG?

[CuzzA]

Lorenzoid, thanks for posting the links.

SPG's are definitely not 100% reliable. We really need to know how today's AI transmitters stack up against SPGs in terms of failure rate. Perhaps we'll never know. It sounds like many divers, whether they know it or not, are beta testers for new technology. But, if someone could statistically predict AI is less likely to fail compared to SPG's I imagine the industry would eventually adopt them as the standard. And if transmitters were standard, the price would likely be substantially less.

 

[doctormike]

CO contamination is still very rare, I don't know anyone who routinely tests for that

I guess you don't know me, and this post isn't going to change that but..

Hey, now I know a person who routinely tests for CO!

 

[tbone1004]

I'm no car expert, but as of 2013 there was one production car with steer by wire. Have they all switched in the past two years?

electric steering isn't fly by wire. Note that new cars don't mean all the current sold, just the new designs. Most new designs use electric assist steering that works similarly to the hydraulic steering that everyone is used to. They have to do this to get around EPA fuel economy rules because the hydraulic assist is constantly sucking power and a significant amount of it vs. the electric that only uses any power when it's being used and only as much as it needs

 

[CuzzA]

Kelemvor,

You bring up smartphones and I look at these devices and see how incredibly reliable they are with great capabilities and amazing processors that can work with a lot of different sensors. Heck some companies have even turned the iPhone into a dive computer. We're also seeing companies take the backwards compatible approach to dive computers and offering operating system updates rather than having to replace the entire device to upgrade.

So given the tremendous track record of smart phone reliability you have to believe that AI can achieve equally as good reliability and dive computers are only going to get better. Perhaps there's a lot to learn from smartphone manufacturers for dive computers.

Also, you bringing up smartphones made me think about some of the battery charging capabilities we're seeing where you can simply lay the device on a charging pad without actually connecting it. I don't know how that works as it's not my field of expertise, but if AI transmitters could simply be laid down on a charging pad rather than have to be opened up for battery replacement I would think this could eliminate a lot of the reports of flooded transmitters.

 

[tbone1004]

the issue with rechargeable pods is that you have to bring your whole reg set into an area where there is power to recharge them, whether that is thru contacts like the H3, the straight up USB port like the Nitek, or thru QI inductive charging *which is painfully slow*. That means a dedicated power outlet for each regulator as well as having to bring wet regulators inside to charge, etc etc. Definitely not optimal. You can get reliable sealing power with a AA battery with no issues, but they don't have enough voltage to kick out the transmitters properly, the coin cells are at 3v so that's part of why they're used. Yes there are higher voltage options like the Saft, but they're expensive and not easily available.

The phone standardization is based on the WiFi standards, Bluetooth Standards, and NFC standards to make everything place nice. Unfortunately none of those technologies work particularly well underwater. Seabear may have the answer with whatever it is they are cooking up, but we won't know until we see the transmitters if and when Scubapro releases them, then it is up to the rest of the mfg's to get off of their high horse and standardize. IF they do standardize on one protocol for wireless AI then we MIGHT see Shearwater add that feature into their computers because they won't have to spend the umpteen thousands of dollars developing a new transmitter, but it is very very unlikely that the industry will standardize. Hell, we just saw the rebreather world MOSTLY standardize on the Shearwater DiveCAN, but even that isn't 100% compatible because each company had to be just a little bit different than the others.

 

[Nuribromanski]

SPG has a couple advantages that integrated digital AI never will: it's an independent source of information, and being analog it has fewer things to go wrong/more obvious failure modes.

If my computer dies, for whatever reason, I can still use an SPG and watch to determine depth (at least close enough to approximate my stops). And while the SPG certainly can fail, it's a direct reading of tank pressure moving a tube/needle - rather than a battery-powered sensor/transmitter followed by another battery-powered receiver/converter/display. If it stops working, there will either be bubbles (failed hose, o-ring, or tube) or the needle will stop moving - with the computer, there are software bugs that could cause failures you cannot easily detect. Kinda like the failure mode differences between OC (obvious) and CCR (will kill you before you ever suspect anything is wrong).

 

[decompression]

Maybe.....if the transmitter were air powered......


Sent from my iPhone using Tapatalk

 

[CuzzA]

Maybe.....if the transmitter were air powered......


Sent from my iPhone using Tapatalk

Now we're talking. I had to Google it, but pressure electricity is a real thing.

https://en.m.wikipedia.org/wiki/Piezoelectricity

 

[CarcharodonCarcharias]

This isn't a discussion about what features "I" want out of an AI computer. It's a general discussion about AI technology replacing SPG's in the future. That is one of the reasons I asked that we keep brands out of the discussion. Audible alarms can be turned off on some computers so the point is moot.

Right. I got that. And I've mention in my post that in my opinion, the answer is no and I don't think is a good idea. AI is cool, but not a replacement for SPG.

Audible alarms can be turned off on some computers so the point is moot.

I brought up this point to merely chime in with another earlier post that was talking about the added benefit of AI technology regarding safety features and why people think rely on AI is a great idea.

 

[kafkaland]

I have a question for DiveNav and others... Lets move the bar a little bit.

How difficult would it be to incorporate CO and O2 analyzing into the AI? Would that not provide some real value to AI other than just tank pressure and consumption data?

I have been wondering about this as well - at least the O2 part. It would eliminate user errors due to setting their computers to the wrong nitrox mix. But I think something needs to happen on the sensor side. The current fuel cell technology dates back to the 50s, and is rather unreliable. The sensors need to be replaced often. I can't believe that a better solid-state or optical sensing technology can't be developed. But it might be too expensive for a rather small, price-sensitive market.