CuzzA
Wetwork for Hire
They would have the exact same amount of failure points. So a transmitter does not equal higher risk of failure.
Should Shearwater add Air Integration to its computers?
- Thread starter boulderjohn
- Start date
Please register or login
Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.
Benefits of registering include
- Ability to post and comment on topics and discussions.
- A Free photo gallery to share your dive photos with the world.
- You can make this box go away
So Tech divers are buying Shearwater because it doesn't have AI??????
AI makes it complicated but the decompression algorithm, compass, etc. don't? I'd say drop the "dive computer" from the dive computer and go back to table slates with pencils and forget the complicated poopoo that would take you down when you least expected underwater.
It is just the "macho" image and ego of some people that want to appear "rough" that makes them make these statements.
AI makes it complicated but the decompression algorithm, compass, etc. don't? I'd say drop the "dive computer" from the dive computer and go back to table slates with pencils and forget the complicated poopoo that would take you down when you least expected underwater.
It is just the "macho" image and ego of some people that want to appear "rough" that makes them make these statements.
SillyQuestions
Contributor
- Messages
- 182
- Reaction score
- 60
- # of dives
- 200 - 499
So you are suggesting an AI transmitter isn't a large device with more complication than a blanking plug with more chance of getting hit?
Especially if you put it on a hose!
Sent from my iPhone using Tapatalk
Especially if you put it on a hose!
Sent from my iPhone using Tapatalk
I have written about this several times. The notion that including AI software or circuits in a computer reduces the reliability of a computer is so thoroughly debunked as to be ridiculous at this point. There are no and I repeat no documented instances of a computer failing to operate as a computer (ie tracking depth, time, and NDL/deco) due to the presence of AI circuits or software. This is with the hundreds of thousands (maybe over a million by now?) of AI computers out there, over many years and millions of dives.
As one post in another thread suggested, installing a bluetooth chip and software poses similar "risk" (non-existent) and no one has criticized Shearwater for that choice--rightly so. Shearwater also has complex connectors and software for rebreather monitoring and control, certainly a more life-endangering function than AI, and no one criticizes them for this--rightly so. Do OC tech divers feel that the extra software and physical connections for rebreathers, which they will never use, make for a more dangerous or less reliable computer? I have never seen anyone say this, but this is no different at all from the presence of AI (which does not even require a physical connection to the computer).
Now, whether a tech diver wants to deal with multiple transmitters and setting up the computer accordingly--that is purely their decision and I have no fault with that. Similarly, if they feel transmitters have a higher likelihood than button gauges of being damaged in collisions with cave walls or roofs, or inside wrecks, or just bumping against other stage tanks, that seems legit. Or, they just don't want to use it, period--that's fine also. But, these are practical considerations to the diver and none of these are reasons to not offer the feature. Any notion that "AI inside" makes a computer less reliable if the AI is not in use is just not so based on objective evidence and experience.
Also, you can resolve the issue of transmitter impact damage by putting it on a short 6" hose, so it is not rigid and flexes away from any impact.
The main "failure" of AI is loss of transmitter signal, something which is actually not too common anymore, and is usually temporary and has no affect on computer functioning at all. More rare is a broken transmitter due to mis-handling, but, again, that only impairs the AI function not the computer function.
So, putting AI in a Perdix or Petrel would not be a "downgrade" from the view of reliability and the tech community would be just as well served even if they never used it.
However, if Shearwater is concerned about tech diver image issues of having this "rec" feature in their "tech" computer, then a separate AI model would be fine.
As for the other thread regarding a "recreational" computer, I don't get that. The Shearwater is just fine as it is. AI is the only thing I would want.
As one post in another thread suggested, installing a bluetooth chip and software poses similar "risk" (non-existent) and no one has criticized Shearwater for that choice--rightly so. Shearwater also has complex connectors and software for rebreather monitoring and control, certainly a more life-endangering function than AI, and no one criticizes them for this--rightly so. Do OC tech divers feel that the extra software and physical connections for rebreathers, which they will never use, make for a more dangerous or less reliable computer? I have never seen anyone say this, but this is no different at all from the presence of AI (which does not even require a physical connection to the computer).
Now, whether a tech diver wants to deal with multiple transmitters and setting up the computer accordingly--that is purely their decision and I have no fault with that. Similarly, if they feel transmitters have a higher likelihood than button gauges of being damaged in collisions with cave walls or roofs, or inside wrecks, or just bumping against other stage tanks, that seems legit. Or, they just don't want to use it, period--that's fine also. But, these are practical considerations to the diver and none of these are reasons to not offer the feature. Any notion that "AI inside" makes a computer less reliable if the AI is not in use is just not so based on objective evidence and experience.
Also, you can resolve the issue of transmitter impact damage by putting it on a short 6" hose, so it is not rigid and flexes away from any impact.
The main "failure" of AI is loss of transmitter signal, something which is actually not too common anymore, and is usually temporary and has no affect on computer functioning at all. More rare is a broken transmitter due to mis-handling, but, again, that only impairs the AI function not the computer function.
So, putting AI in a Perdix or Petrel would not be a "downgrade" from the view of reliability and the tech community would be just as well served even if they never used it.
However, if Shearwater is concerned about tech diver image issues of having this "rec" feature in their "tech" computer, then a separate AI model would be fine.
As for the other thread regarding a "recreational" computer, I don't get that. The Shearwater is just fine as it is. AI is the only thing I would want.
The HP hose with an analog gauge is more of an explosion threat than a AI transmitter for sure.
AJ
Contributor
Floating around exploring, no real task. But during a task I would not bother to check my SPG or my DC for air remaining if that's what you're steering at. I know how much air is left at any time by quite accurate guestimate. Still see no need for AI because checking air on my SPG or on my DC remains basically the same activity, i.e. interpret a measuring device. Unclipping an clipping my SPG is second nature to me, it does not mean any significant taskload and costs just seconds.
I have to disgree with that. A blind plug is very unlikely to blow. A hose or a measuring device that is under high pressure has a higher risk it will blow.
CuzzA
Wetwork for Hire
Putting it on a hose is silly IMO and defeats the purpose of "hoseless" AI. I would agree there may be a slightly higher risk of failure from hitting it on something. But that risk is pretty small and if you actually hit it on something hard enough to break the seal on the first stage you probably have greater problems due to the impact your head/neck probably took as well.
---------- Post added January 11th, 2016 at 11:04 AM ----------
I understand. I've never said "need" in this thread. I think I've called it convenience a few times.
A higher risk, probably. How much? Who knows. It's still a failure point anyway you look at it.
Has anyone ever experienced or heard a story of a transmitter's physical connection ever failing? I tried to do a search and came up with nothing. I've heard stories of SPG hoses and face plates blowing out, but never a transmitter.
---------- Post added January 11th, 2016 at 11:04 AM ----------
I understand. I've never said "need" in this thread. I think I've called it convenience a few times.
A higher risk, probably. How much? Who knows. It's still a failure point anyway you look at it.
Has anyone ever experienced or heard a story of a transmitter's physical connection ever failing? I tried to do a search and came up with nothing. I've heard stories of SPG hoses and face plates blowing out, but never a transmitter.
Entirely right, but my first stage (Poseidon) has a recessed port which did not allow the transmitter to seat, so the hose was necessary. It is entirely unnoticeable, and I do have some comfort now that a DM won't try to lift my rig by grabbing the transmitter (although that has never happened). My wife directly attaches hers with no problem.
Putting a hose on an reg with an AI transmitter is pointless, but as a tech diver you have to assume things are going to break so you have to have redundancy so if the transmitter failed or for what ever reason you still have to know what the pressure in that tank is.
As for making it more complex any circuitry you add increase errors and complexity of the system. I fly helicopters for a living and I have a computer onboard that is in the millions far more than any dive computer. I have personally be in the situation hey we added a new feature in your software then go fly the computer encounters problems, i dim it down because no it is a paper weight, land and I am told by the engineers it shouldn't have been able to do that. Well it did. Dive computers are the same way give me what I need to do my dive with out things I don't.
As for trusting the computer we do expect it to fail which is why I usually have 2 there is always redundancy. To top that when I plan my dive there is a plan and alt plans for staying to long, to deep, or losing gas written in pencil on my slate right next to the wrist watch so if both computers fail I can safely return to the surface.
AI is a convince which as you dive more and learn your sac rate where you know what is in your tank or in our case tanks before you look at the spg becomes much less important.
As for blue tooth we don't use it during the dive it is to update software and down load dives. They could have put in a port to plug in USB but that would increase the chance of a computer flooding. Rec diving and tech diving have very different mentalities, very different views on gear and compromise in the tech world can often leave you in a tough spot when things go wrong. No one is saying all computers should ditch AI every computer should be like the petrel, it is just being said that this is a company that focuses on tech diving computers and makes damn good ones. If you are wanting different features use a computer that is designed for that kind of diving. The right tool for the right job, that is why most tech divers have more than one diving kit each one is set up for a certain kind of diving. I know many that even have a SW for tech diving and use oceanics and other recreational computers when they are doing recreational diving as the features or "tools" required for the different kinds of diving are different.
As for making it more complex any circuitry you add increase errors and complexity of the system. I fly helicopters for a living and I have a computer onboard that is in the millions far more than any dive computer. I have personally be in the situation hey we added a new feature in your software then go fly the computer encounters problems, i dim it down because no it is a paper weight, land and I am told by the engineers it shouldn't have been able to do that. Well it did. Dive computers are the same way give me what I need to do my dive with out things I don't.
As for trusting the computer we do expect it to fail which is why I usually have 2 there is always redundancy. To top that when I plan my dive there is a plan and alt plans for staying to long, to deep, or losing gas written in pencil on my slate right next to the wrist watch so if both computers fail I can safely return to the surface.
AI is a convince which as you dive more and learn your sac rate where you know what is in your tank or in our case tanks before you look at the spg becomes much less important.
As for blue tooth we don't use it during the dive it is to update software and down load dives. They could have put in a port to plug in USB but that would increase the chance of a computer flooding. Rec diving and tech diving have very different mentalities, very different views on gear and compromise in the tech world can often leave you in a tough spot when things go wrong. No one is saying all computers should ditch AI every computer should be like the petrel, it is just being said that this is a company that focuses on tech diving computers and makes damn good ones. If you are wanting different features use a computer that is designed for that kind of diving. The right tool for the right job, that is why most tech divers have more than one diving kit each one is set up for a certain kind of diving. I know many that even have a SW for tech diving and use oceanics and other recreational computers when they are doing recreational diving as the features or "tools" required for the different kinds of diving are different.
So the absence of documented instances means there is an essentially zero likelihood that it happens?
How many people report instances of strange behavior of their computers? How many computers are AI? How many of the kind of divers who like AI are also the kind who would even notice transient strange behavior of their computers?
Omitting a subsystem like AI is a way to ensure with 100 percent certainty that it will not cause any problems.
I'm not saying I disagree with the point you're trying to make, but I am disagreeing with your premise. I do believe there is some, perhaps very very small, likelihood of some unintended effect. (And yes, even the addition of Bluetooth and the compass bother me a little in that respect.) I would want Shearwater's guarantee that the AI system is completely walled off from the deco computation system. Could they do that? In their manual, Shearwater candidly admits what all engineers know: that the computer has as-yet undiscovered bugs and that the computer may do things that they didn't intend it to do. Do they really need to compound that by adding more systems? If it were me doing the engineering, I would see what I could remove from it and still have it function as intended before I would see what I could add to it. Less is better.
Similar threads
