Regarding possible backfocus - while it is possible, your other shots you posted show no signs of it. From what I've read of Canon's focusing system, the actual focus area is a bit bigger than the square focus point. If you cover the autofocus point with something very low in contrast but there is an adjacent area with higher contrast/detail the camera may not focus on your intended point. Other autofocus issues to account for are subject movement between the time of focus lock and taking the shot, minimum focus distance of the lens, and your own movement of the frame after obtaining focus lock.
Things like back/front focus should be fairly consistent. If you're seeing lots of sharp, in-focus shots with the occasional focus-error, it's likely related to some other issue (see above). Focus errors are best evaluated in a very controlled environment so as to eliminate as many confounding variables as possible.
I'd be interested in seeing some repeat shots with the 30D and no filter as well.
For testing purposes, I would suggest using "One Shot" AF drive mode, so that you don't have the camera second guessing you. It also seems that several shots have the camera averaging across multiple focus point, thus making the focal pt alot bigger. You may also make sure there is no focal pt change between using your AF-ON button and the shutter button, perhaps changing the camera to let shutter button do focus as well, just to eliminate any possibilities.
Its really hard to tell, though it does look like backfocus to me as well.
While you are at it, you may as well test this for sure. go download this chart (at the top of the page), read the instructions and check out how it works. This will absolutely demonstrate back/front focus with ease.
same problem Its unreal that you bring these problems up with your Canon lens. I am fixing to send my lens to Canon for a check up unless someone could tell me of a better place to send it. My lens is the 70-200 2.8 Non IS. I have had the same problems with this lens shooting with the 20D, 30D, and 40D. Thanks
Although I'm new to this thread, the above tests seem to indicate that the lens is focusing quite well.
At first I thought I saw BF, but then realized that the 2mm lettering is a bit larger than the words "be perfectly in focus", so the chart can be misread.
Then I compared the front and rear 2mm letterings, and now feel that the first test front-focused a bit, while the second two back-focused a tad. But these are very small variations - considering that a lens will seldom focus perfectly the same every time, these deviations are small indeed.
What I find interesting in the first group of images is the Harbour Air floatplane; I live about a ten minute walk from their Victoria base.
"There is nothing that some man cannot make a little worse and sell a little cheaper, and he who considers price only is that man’s lawful prey". John Ruskin 1819 - 1900
I did test the Canon 70-200mm, f2.8L (non-IS) against the f4L, IS (IS turned off) and they are almost indestiguishable. If I didn't know which was which, I couldn't tell. Initial tests of the IS turned on are that it works very well at 200mm down to 1/100th. (Consistantly good)
I am still testing the f4L IS for accurate closer focus as time permits. First results here:
Ziggy, is the 70mm shot wide open? That's the sort of relative softness I'm used to seeing on mine at f/4. I have learned to use the lens better and to love it more, but I find near-universal claims of "sharp wide open" to be a bit exaggerated.
Ziggy, is the 70mm shot wide open? That's the sort of relative softness I'm used to seeing on mine at f/4. I have learned to use the lens better and to love it more, but I find near-universal claims of "sharp wide open" to be a bit exaggerated.
It is wide open and I believe it is front focused and my fault. The front edge of the paper plate is probably too close to the region that the single point autofocus could detect.
Other tests with more usual subjects show a sweet spot at f5.6, the same as my f2.8 version of the lens.
I do believe that f4 on both lenses is about the same and that f5.6 is also about the same.
It is wide open and I believe it is front focused and my fault. The front edge of the paper plate is probably too close to the region that the single point autofocus could detect.
Other tests with more usual subjects show a sweet spot at f5.6, the same as my f2.8 version of the lens.
I do believe that f4 on both lenses is about the same and that f5.6 is also about the same.
Interesting. I guess what I mean by "unsharp" is the kind of hazy, uneven look at f/4, which I'm used to seeing when I zoom in—they look fine when viewing at 25% or so. I think I figured out that I need to start printing more photographs because pixel peeping is starting to ruin me a little. It entirely possible that too much exposure to 100% crops is hazardous to your enjoyment.
Now that I've proved to myself that my lens is OK (the operator... not so sure) it was helpful to see examples of other specimens.
Success!!
Finally gaining the experience with this lens that it deserved. A recent zoo trip delivered, if not great photos, at least a really good in-focus rate, basically 95%. Thanks to all you who helped me crack this nut. She's a keeper!
Comments
I'm beginning to think this had a lot to do with it, as my old 30D had no problems with my other lenses.
Things like back/front focus should be fairly consistent. If you're seeing lots of sharp, in-focus shots with the occasional focus-error, it's likely related to some other issue (see above). Focus errors are best evaluated in a very controlled environment so as to eliminate as many confounding variables as possible.
I'd be interested in seeing some repeat shots with the 30D and no filter as well.
Its really hard to tell, though it does look like backfocus to me as well.
While you are at it, you may as well test this for sure. go download this chart (at the top of the page), read the instructions and check out how it works. This will absolutely demonstrate back/front focus with ease.
Downloaded. Will perform soon. Thanks.
Its unreal that you bring these problems up with your Canon lens. I am fixing to send my lens to Canon for a check up unless someone could tell me of a better place to send it. My lens is the 70-200 2.8 Non IS. I have had the same problems with this lens shooting with the 20D, 30D, and 40D. Thanks
Okay,
Used that focus chart and here are the results from three different angles. Did all the things cmason said.
45-degrees, as specified. Approx 4.5-5 ft distance, middle AF point:
full size original—45-degrees
100%:
Approx 37-degrees. Approx 4.5 ft distance, middle AF point:
full size original—37-degrees
Approx 30-degrees, Approx 4.5 ft distance, middle AF point:
full size original—30-degrees
At first I thought I saw BF, but then realized that the 2mm lettering is a bit larger than the words "be perfectly in focus", so the chart can be misread.
Then I compared the front and rear 2mm letterings, and now feel that the first test front-focused a bit, while the second two back-focused a tad. But these are very small variations - considering that a lens will seldom focus perfectly the same every time, these deviations are small indeed.
What I find interesting in the first group of images is the Harbour Air floatplane; I live about a ten minute walk from their Victoria base.
Thanks for looking. I think I agree that you cannot and will not get absolute consistency.
Incidentally, I took that shot from that water park in Stanley Park. Visited Victoria that same trip.
I did test the Canon 70-200mm, f2.8L (non-IS) against the f4L, IS (IS turned off) and they are almost indestiguishable. If I didn't know which was which, I couldn't tell. Initial tests of the IS turned on are that it works very well at 200mm down to 1/100th. (Consistantly good)
I am still testing the f4L IS for accurate closer focus as time permits. First results here:
http://dgrin.com/showpost.php?p=839048&postcount=8
Moderator of the Cameras and Accessories forums
- Ansel Adams.
Ziggy, is the 70mm shot wide open? That's the sort of relative softness I'm used to seeing on mine at f/4. I have learned to use the lens better and to love it more, but I find near-universal claims of "sharp wide open" to be a bit exaggerated.
It is wide open and I believe it is front focused and my fault. The front edge of the paper plate is probably too close to the region that the single point autofocus could detect.
Other tests with more usual subjects show a sweet spot at f5.6, the same as my f2.8 version of the lens.
I do believe that f4 on both lenses is about the same and that f5.6 is also about the same.
Moderator of the Cameras and Accessories forums
Interesting. I guess what I mean by "unsharp" is the kind of hazy, uneven look at f/4, which I'm used to seeing when I zoom in—they look fine when viewing at 25% or so. I think I figured out that I need to start printing more photographs because pixel peeping is starting to ruin me a little. It entirely possible that too much exposure to 100% crops is hazardous to your enjoyment.
Now that I've proved to myself that my lens is OK (the operator... not so sure) it was helpful to see examples of other specimens.
Finally gaining the experience with this lens that it deserved. A recent zoo trip delivered, if not great photos, at least a really good in-focus rate, basically 95%. Thanks to all you who helped me crack this nut. She's a keeper!