Don't get me wrong, if you just look at the curves/math.. it looks like the rechargeable should perform really closely, but basically what I see with my wall clock is unstable operation. It seems to gain and lose time with the rechargeable, but with primary cells I feel like I had a good feeling of how much it would drift when it was getting low to the point where I could ignore the fact that batteries were low because I had trust in its behavior (my brain was capable of compensating the error for quite a while so long I Had an idea of what it should say that week), but with rechargeable I basically gotta deal with it as soon as I notice it because the error gets funky. Actual measurement on something like that is too annoying/trivial to deal with. Like I almost swear I started to notice 'monotonacity' problems that disappear on the next day when I switched it to NiMH, but usually I only deal with this clock in the morning and forget about it later. I don't want to try to study it because its too simple to just revert back to the alkaline battery.
But for the headlamp I stick to NiMH batteries because I burn through them too quick and it gets expensive so I just deal with the reduced capacity also.. And in general I just keep alkaline around because doing the whole NiMH upkeep thing all the time sometimes feel imposing so its like a 'cheat day' where I pay for the luxury of not having to deal with the chargers, so I get to it a month later and then have charged NiMH on hand. Maintaining operations is nice if you are tired.