Hmmm... That's odd as I did 24 dives in 13 days so far this year (March and August dive trips) most at 60-75 minute duration. I also had accidentally left the Perdix AI on high screen brightness (versus auto) on 1/2 of them and I bluetoothed every dive to my iPhone after I got back on the boat/back to shore. My Perdix AI 3.6 V Saft is still reading 3.44 V. My non-AI Perdix backup worn on all dives (also left on high brightness) is still showing 3.49 V on it's 3.6V Saft batery. So, to me, the battery life seems good. Changing a battery is easy too - pinching 2 O-rings is major user error (even pinching 1 would be), and not even sure how one could do that unless they were very careless.I just dove (OC 47 dives) my Shearwater PERDIX AI and it's an awesome, fun dive computer but the battery is an issue. My Saft LS 14500 3.6 V lithium-thionyl chloride battery was brand-new, it's reading one bar at 1.5V after 16 days of diving. I dove the NEW Cochran Lifeguard side by side and I started with 3.1 volts on my Wrist Unit, PBAT & SBAT of my CPU and end with 3.0 on all units after 16 days of diving.
On a personal note I liked a lot of the features on both dive computers but I believe the Cochran algorithm is better. Also being able to upload my dives instantaneously to my iPhone was pretty cool.
WRT algorithms, I dove Nitrox on about 1/2 of the dives and was only NDL limited on 1 dive where we had spent significant time below 100 feet on the earlier dive - so the algorithm worked great and I never found it limiting (buddies Zoops were beeping). Why do you feel the Cochran algorthm is better?