gschlact wrote in post #17418197
Don't your stats also include a dose of subjectivity? For example, you include Yellow as a keeper with green. Can you remind us your definition of Yellow? Also, I think in general there needs to be a more clear indication when some claims OOF. (not referring to HUNTERSDad or PP). To explain, there is a difference between OOAF (motion blur, and/or other mechanical caused softness) and what most people should be saying is Out Of AUTO Focus. The difference is that OOF implies that there is not resulting plane achieving the camera-lens' maximal sharpness; whereas OOAF implies that the plane of maximal sharpness did not land where intended. There are different causes for each. Ppmax's original skier AF was oof caused by IS settling inducing mechanical mini blur. HUNTERSDAD has been having OOAF.
Traditionally, for phase detect Af mfa has been used as a lens offset to correct OOAF. As on moved from static to ai servo the tracking algorithms determine and accomplish the ability to maintain AF. This tracking is accomplished by software and the ability of the lense to respond to the cameras commands. Ai servo that works for static objects should always work for moving objects assuming Af point place on subject well And adequate Case settings and some other camera settings are proper. If not, it would point to the user's choice of settings, or some other decision impacting tracking. If not user, it would have to do with the specific lens, despite whether the lens is perfect on another body or body type. I would not expect multiple different lenses to work on a given body producing sharp ai servo static focus, but not for moving objects without there being a user software error as 99% cause, unless with exact same condition those same lenses and settings worked on another 7dii which would then print to an electrical/interface issue with lenses in first body. . This strongly supports the rising theory that HUNTERSDad issue is related to the specific lens (or combo) in use, and leaves open potential that it is a lens model. After all, what has been consistent is that he has had at least 4 bodies giving same behavior OOAF with same testing and common lens in each.
The only subjective part of the stats are if something is in focus or not...and I don't believe there is any absolute method that could be used to eliminate this. Having said this, although the process of grading focus is subjective, one can typically answer a simple question like "was focus achieved on the desired target?" My possible answers were Yes, No, Kinda (Green, Red, Yellow). I defined the "Kinda" category as a being a shot that was soft at 100% but "acceptable" at 50% or 25% zoom. Anyone can argue with this...but at least my metric was consistent.
I totally agree with the points you make about using the appropriate Case (and other settings) for the action: I posted about this many pages ago. In summary, due to my familiarity with the behavior of the 5D3 I assumed that I could use similar settings and technique across cameras with the 7D2. While this is generally true, it has become clear to me (through this process of collecting and analyzing my photos) that the 7D2 is it's own beast and needs to be used in a manner consistent with it's own features, functions, and strengths and weaknesses. Frankly, despite the similarity of the UI presented in menus to the user, the AF performance of the 7D2 is quite different vs. the 5D3 and the user needs to recognize this and adapt to the tool's behavior. I think an appropriate analogy is the Copy/Paste functionality provided in Windows vs. that provided in OS X: Generally, they do the same things...but there are significant differences between how each vendor has implemented these functions, and the user would do well to learn how to use both to maximum benefit.
Ppmax's original skier AF was oof caused by IS settling inducing mechanical mini blur. HUNTERSDAD has been having OOAF.
IS was indeed causing some lateral blurring--this was a lesson learned, but it was not the cause of all of my OOAF shots. I can post plenty of shots taken with IS off to prove this