we're not talking about turn pressure (which you can set to alert at any pressure you want) or max depth (determined not by AI but other factors like mix and NDL). We are talking about what AI can offer with regard to a reserve for a gas sharing ascent--and I in particular am a rec diver with that perspective.
I may need to clarify how the ATR works. The "reserve" is not what you begin your ascent with, it is what you will have on the surface. Take the 500psi example. If I am at 100', the computer will not send me up when I reach 500 psi. The computer might send me up (depending on my real-time consumption) when I reach 800 psi, because it is predicting, based on real-time dive consumption, that it will take me 300 psi to reach the surface. Double that means 600 psi for two of us to reach the surface. So, the computer gets us both up with a 200 psi cushion. This might seem slim, but the computer is predicting a normal ascent rate and a safety stop. In reality the cushion is pretty substantial if you omit the stop and ascend faster.
This is self-adjusting for depth and effort. The deeper you are and/or more you are working,, the more you are consuming, and the computer will send you up with more in your tank.
This could be adjusted pretty easily for a "rock bottom" calculation tied to real-time dive conditions. For example, a calculation based on double your current consumption, getting two of you to the surface with, say 200psi or whatever reserve you set, perhaps even with adjustable parameters for the ascent. That, seems to me, would be better than a pre-dive estimate based on assumptions that might or might not apply on a given dive.
I may need to clarify how the ATR works. The "reserve" is not what you begin your ascent with, it is what you will have on the surface. Take the 500psi example. If I am at 100', the computer will not send me up when I reach 500 psi. The computer might send me up (depending on my real-time consumption) when I reach 800 psi, because it is predicting, based on real-time dive consumption, that it will take me 300 psi to reach the surface. Double that means 600 psi for two of us to reach the surface. So, the computer gets us both up with a 200 psi cushion. This might seem slim, but the computer is predicting a normal ascent rate and a safety stop. In reality the cushion is pretty substantial if you omit the stop and ascend faster.
This is self-adjusting for depth and effort. The deeper you are and/or more you are working,, the more you are consuming, and the computer will send you up with more in your tank.
This could be adjusted pretty easily for a "rock bottom" calculation tied to real-time dive conditions. For example, a calculation based on double your current consumption, getting two of you to the surface with, say 200psi or whatever reserve you set, perhaps even with adjustable parameters for the ascent. That, seems to me, would be better than a pre-dive estimate based on assumptions that might or might not apply on a given dive.