Computer with manual AI?

[jgttrey]

As I understand it: From a macro perspective, computers without tank pressure input are not significantly safer than gauges and a watch. Computers with tank pressure input, give the diver actual remaining dive time and instructions with considerations to depth, time at depth, stops, etc.

I believe you are conflating two issues here. The key benefit of dive computer versus tables and a watch is that a dive computer tracks your nitrogen loading status in real time, based on your real dive profile, not max depth like a table. It will continuously update your remaining "No Deco" time based on what you've actually done so far. Safer than dive tables, a watch and depth gauge? No, not if you follow those things. Both will keep you out of deco. Is a computer better? Yes. By far. On multilevel dives you avoid the unnecessary conservatism built into dive tables.

None of that has anything to do with gas consumption or "gas time remaining." We're just talking about NDLs.

The second issue is gas consumption, which is totally different stuff. That's where AI - or not - comes in.

Some computers are air integrated. They give you a snapshot of how fast you are going through your gas and extrapolate, based on that snapshot and your depth, how much time you have left at that depth assuming some sort of normal ascent profile and a safety margin of remaining gas at the surface.

I do not find that feature to be terribly useful beyond helping you, in the very beginning, learn to make estimates and seeing in real time how different work loads impact your rate of consumption.

Why is it not that useful as a "safety" feature? If your gas consumption rate changes - as it would by orders of magnitude if you changed what you were doing, your GTR will change almost instantly. Gas consumption will go down on your safety stop when you are not moving, and will go up (very) significantly if you were swimming against a current or dealing with some emergency. It be a mistake to treat GTR as anything other than a rough estimate based on current consumption.

If you want to understand your gas consumption better, you don't need AI. Note your starting pressure and ending pressure and average dive depth (which any computer will show you). Then do the calculations to figure out your RMV.

AI's benefits are more about convenient monitoring without looking at gauges and automatic logging of tank pressures making RMV calculations automatic when you load the dive log into a computer. GTR is a long way down the list and is not something you should ever rely on beyond developing a better understanding of your gas consumption rate in general and then planning your NEXT dives appropriately. Obviously, if GTR drops very low you should shallow up or end the dive, but beyond that, it is not very helpful on that dive, but the converse is not true -- just because you might have what seems to be enough GTR remaining, it doesn't mean you are good to go, not if your workload changes. That's part of pre-dive planning and understanding, and planning, when you should start your ascent. If you treat GTR, by itself, as "permission to stay down longer" without understanding its limitations, you will get into trouble.

 

[Joneill]

@jgttrey..... good point, not considered. One possible solution to this would be to enter start and ending dive pressures. After maybe 10 dives the computer could have calculated a reliable consumption rate, but that would be diver specific of coarse.

That would be VERY dangerous as one's air consumption rate can be affected by a variety of factors on any given dive (current, are you cold, how do you feel that day, etc.,) - your computer has no way of knowing any of those factors for a dive!

 

[BoltSnap]

One possible solution to this would be to enter start and ending dive pressures. After maybe 10 dives the computer could have calculated a reliable consumption rate, but that would be diver specific of coarse. Seems like there could be some possible work-arounds, but healthy safety factors could/should be integrated.

This is impractical, has no benefit, lots of room for gross errors, irrelevant and most likely won't make the cost of this computer any less than the real AI computer.

 

[eelnoraa]

@jgttrey..... good point, not considered. One possible solution to this would be to enter start and ending dive pressures. After maybe 10 dives the computer could have calculated a reliable consumption rate, but that would be diver specific of coarse. Seems like there could be some possible work-arounds, but healthy safety factors could/should be integrated.
.

Except it doesn't really work that way for any particular die. SAC is a very situational thing. You can do 100 uneventual dives in warm water and get an average SAC. the next dive, you maybe a more stressful because of your buddy, because water is a bit colder than you feel comfortable, stronger current, darker water .... you SAC can easily double because of these. You don't want to assume your average SAC in such a dive.

 

[Doctor Rig]

All my previous comments were based on the research I’ve recently done, as I’m getting back into the sport where computers were non-existent. In that research, I found that the DIR (doing it right) dive philosophy does not promote the use of computers at all, except maybe as a back up for basic dive data. This approach led me to believe a computer with AI would be overkill for my relative meager sport activities and needs.

 

[AJ]

DIR does deny the use of computers, they do not want to rely on them. Is AI overkill? Well it's convenient but not neccesary to dive. You can do the math yourself. It's not that hard. To decide if it's overkill is up to you.

 

[victorzamora]

My biggest issue with AI computers is that people tend to rely blindly on the "Gas Time Remaining" which is never calculated with what I'd consider sufficient conservatism. The computer will give you a GTR of zero only when you would end with a pre-determined "final pressure" (~500psi) after ascending from your current depth with your normal SAC and no buddy. What it does not take into account is an emergency that would elevate your SAC or put another diver (with elevated SAC) on your tank.

I think the biggest benefit of AI is having your pressure available to you on the same device as your deco status.

If you want safety, dive a computer and dive "Rock Bottom" or "MinGas" calculations as your pressures for turning the dive.

 

[dmaziuk]

As I understand it: From a macro perspective, computers without tank pressure input are not significantly safer than gauges and a watch.

No. They are not significantly safer then a gauge and a watch and a pre-computed runtime on a square profile. The moment you deviate from your pre-computed plan, all bets on your gauge and watch are off.