There are two possible factory calibrations - one lens matched to one body (if you send both in AND they are both from the same manufacturer) or both adjusted independently to be within the factory tolerance. So no dimension or specification is exact - everything has a tolerance (I mention this as the quality manager for a high precision aerospace manufacturing company where we calibrate EVERYTHING). The net-net is that both the lens and the camera can be within factory tolerance, but when mated, the tolerance stack up creates an unacceptable error if max focus accuracy and sharpness is the goal. On the other hand, if you factory match every lens you own to one body, then it won’t be matched to another body, and let’s not forget the inconvenience, potential damage, cost and potential shock which can change the calibration during shipping. There is also no reason to believe that their “calibration” is anything different than applying a “blanket” change to the AF. Ideally, there would be a curve of corrections applied such as that created using some aftermarket lenses “docks”, but we have no way of knowing if the factory has this ability on all (or any) of their lenses.
Having said that, I will gladly grant that on most cameras with MFA, there are a max of two adjustments for zooms and one for primes, so not perfect as the “correction curve” that would be ideal, BUT it is very likely, if properly applied, to be preferable to no adjustment. The claim is routinely made that it can cause misfocusing (but if you don’t like the results of your calibration, just turn it off or reset to zero). It is also claimed that it is appropriate only at one distance and at one FL for zooms or can prevent focusing at infinity. Obviously, I can’t test every every possible lens, but I have tested a lens at multiple distances and infinity, and found that while the adjustment does vary some with distance, the KEY is that at every distance, it was an improvement at EVERY distance and focused fine at infinity. I published the results on UHH (
https://www.uglyhedgehog.com/t-508300-1.html), and if anyone disagrees with the conclusions, I invite them to repeat the test on their lens(es) and publish the results.
For all those that claim to have never needed to adjust a lens, I would answer that without trying, one never knows if their lens can perform better or not. Needless to say, I, and many pros, do calibrate their lenses. I am in no way as good a photographer as some that do not calibrate, but I can represent that by careful calibration and measurement, I know that my lenses are delivering the max performance that they're capable of. This is one of those “religious” issues like Canon vs Nikon, PC vs Mac, etc., so those that don’t calibrate are unlikely to try it, and those that do, believe in the results they get. And of course, one has to wonder why manufacturers include this feature if it has no value when properly applied.
There are two possible factory calibrations - one ... (