[QUOTE="Paul"]\nThe background is overwhelming, let me try to summarize my current\nunderstanding.\n\nThe beep test uses a focus rail and if you stop to re-check rather than\nholding a half-press from the initial position, then you see a consistent\nfocus band between a 3mm front focus and perfectly in-focus.[/QUOTE]\n\nIt doesn't matter whether you stop and recheck or hold a half-press all the\nway through, you see the same thing.\n[QUOTE]\nThe AF test, letting the camera turn it's own focus ring shows the same\nbeep band, front focus or perfect focus but there is a crossover\nphenomenon depending where initial focus starts, which is rather\ncounterintuitive. \nYes.\n\nbut still shows an overall front-focus.[/QUOTE]\n\nI find that a very artificial concept, like saying that on average a man who\nis one half of a heterosexual couple is half female.\n[QUOTE]\nPerhaps the crossover can be explained by the different character of the\nbokeh from front & back, possibly it has to do with magnification changes?\nTurning the focus ring changes the enlargement or focal length if you want\nto say it that way... I don't know that this would produce such a result\nbut it's another idea, it is something that changes.[/QUOTE]\n\nNew and interesting, but a long shot I think. :- )\n[QUOTE]\nAnyways I'm going to say this doesn't matter, the beep band is more\nrelevant for analysis although yes I know the AF motor results are what's\nreal in the end, we don't have an explanation and the results are the\nsame. \nRight.\n\nThe front-focus could be explained by mis-calibration but I recall you\nsaid it works perfect at f/2.8? and by f/4 or 5.6 it's probably hard to\neven see any discrepancy.[/QUOTE]\n\nI haven't gone very far with stopping down (maybe up to f/5.6 or so), but I\ndon't see a drift. We know the lens applies a looked-up correction for that\nin AF.\n[QUOTE]\nMy current explanation is aperture related focus shift where the AF\nsensor sees it at f/2.8 but the final shot is taken at f/1.8.[/QUOTE]\n\nSo we come back to where we were three weeks ago. If the remote Live View\nimage (at 100% on the PC monitor) is in optimal focus, either by AF or beep,\nwhere's the shift? There is no shift that I can see. The misfocus happens\nunder identical conditions, except we started with the POF on the other side\nof the subject. I still cain't see no shift.\n[QUOTE]\nThis explanation should show a little back focus when stopped down\nto f/4 or more but perhaps that's been calibrated out with a pinch of\nfront focus added to the correction tables??? Hell, I don't know :-)[/QUOTE]\n\nWell, we can eliminate the influence of the correction tables by doing the\nbeep test, and we get the same results (same two focus points) as we do with\nAF, so where's the shift?