After taking numerous shots of many things in my house, the 135L is consistently focusing right on and then the next shot of the same thing, focused in the same spot is front focused by alot and then the next shot will be sharp where it's suppose to be again for about two in a row!
In the upper two pictures, I think we
We're a Canon/Profoto family: five cameras, sixteen lenses, fifteen Profoto lights, too many modifiers.
Originally Posted by peety3
This is something that many people do not know, but it's absolutely correct. Another relevant bit of information is that the actual AF sensor is a bit larger than the box representing it in the viewfinder. Finally, you're right that the center AF point on the 7D is actually a combination of a "+" and an "x" - but different parts of that AF point have different sensitivities. When using the 135L (or any lens of f/2.8 or faster), the "x" part of the center AF point is used, and with slower lenses, or if the "x" subsensor can't achieve a lock, the "+" is used.
Denise, I took a pair of your examples and superimposed an approximation of the 7D's center AF point onto them. The green lines are the f/2.8-sensitive high-precision sensor used when focusing with the 135L. As you can see, on the left it looks like the lower bars of the "x" were used, and on the right it looks like the upper bars of the "x" were used. Technically, based on the limitations of the AF system, both of the shots are in focus.
[img]/resized-image.ashx/__size/550x0/__key/CommunityServer-Components-UserFiles/00-00-00-35-15/DT-AF-example.jpg[/img]
This sort of thing is why AF microsdjustment tools (LensAlign, SpyderLensCal) have aflattarget for the AF system, and only the depth of field scale is at an angle to the camera. It's also why even with a precisely adjusted AF system, there are still misfocused shots - sometimes it's just randomness, but usually it's because unlike AFMA tools, the real world isn't composed of flat, perfectly aligned, high contrast targets. But all an AF system can see is a contrasting feature crossing a pair of sensor lines of the selected AF point. If there is more than one of those, and the subject is at an angle, and the DoF is thin enough for it to matter, the camera has no way to know which one you're actually trying to hit.
As a side note, you mentioned that it is, "...consistently focusing right on and then the next shot of the same thing, focused in the same spot is front focused by alot." I don't know if I'm giving too much credit to the AF system, but it may actually be designed to do just that - i.e., with a couple of shots in succession using the same AF point, where different parts of the AF sensor give different planes of focus, the system alternates which subsensors are given preference from shot to shot, to increase the likelihood that you get a shot where the AF system locked onto the feature you intended.
It never fails ...every day I either learn something new from members on this site or I gain a better understanding of things I thought I knew! Thank you so much John for going to the trouble of putting your response together for me! I extremely appreciate it!!
I wish I had more time to accurately test this lens and my others during the week. I still have some time left to still return it so I want to make sure I am overjoyed with it before that time runs out. If I
Posting Permissions
Reply With Quote