Fovea35 wrote:
Two different and independant french magazines ("Chasseur d'Images" and "Réponse Photo") tested the 50 f/1.2. It appears that the wider aperture gives a lot of trouble to the autofocus system leading to very poor focusing accuracy (even on a 1DsMarkII)...
The results show that the f/1.8 has the better focusing accuracy, the f/1.4 second and the f/1.2 the worse.
This would be linked to the wider apertures used during focusing that lead to reduced microcontrast, which is not compensated by the increased luminosity : the autofocus has trouble locking.
Fovea35, if you'll read this thread you'll find that the new 50L f/1.2 seems to have the least difficulty when focusing wide open at f/1.2. It is after stopping down where the trouble becomes readily apparent. So I believe the magazine articles may have judged this in error.
Also, I have owned 3 -50L f/1.0s and also an 85L with my 1DsMarkII body and have never experienced consistent focusing issues from ANY of these lenses.
And yes, I do understand the concept of reduced microcontrast at wider apertureds that can lead to issues, but I don't believe this is exactly what is at work here.
Plus stopping down actually has nothing to to with focusing theoretically...because the Canon's system always AFs when the lens is wide open. The stop down does not occur till the shutter button is depressed.
Photon wrote:
Clearly you have a problem that I'm not having.
Did you also shoot at wider apertures? Are you finding consistent back focusing, or like Sam are you finding that focus is good at 1.2, 1.4, 1.6...?
You had the focus POINT on the left battery. The highest contrast by far is the right side of the right battery. That looks to be in focus. The focus sensor is considerably larger than the focus point.
ghuff wrote:
By the way all....the lens did the same on both my bodies...and I sent my 1Ds2 in on the last calibration trip to Canon...No problem found with the body.
Net result of all the efforts? Backfocusing fixed at short distances...but front focusing induced at large distances....
I gave up...
SAM......chin up!!!
Ed, thanks for chiming in!
I am sincerely not trying to be provacative.
What you are likely experiencing is the current state of the art of Canon autofocus. AF is an estimation. Af is a measuring tool that can be optimized at certain distances, but this will worsen perfomance elsewhere. Perhaps the new 50L has some changes for a future AF system that make it perform suboptimally with the current system. I don't know. There a small chance of a design mistake. Gary finding multiple lenses with the same performance may be an indicator of a design problem. But what is much more likely is that it indicates the lenses are in spec. You may just not like the spec.
ghuff wrote:
Also, I have owned 3 -50L f/1.0s and also an 85L with my 1DsMarkII body and have never experienced consistent focusing issues from ANY of these lenses.
Well, the 85L is probably OK because of shallower depth of field helping the autofocus mechanism. I don't know about the 50 f/1.0 L though.
ghuff wrote:
And yes, I do understand the concept of reduced microcontrast at wider apertureds that can lead to issues, but I don't believe this is exactly what is at work here.
Plus stopping down actually has nothing to to with focusing theoretically...because the Canon's system always AFs when the lens is wide open. The stop down does not occur till the shutter button is depressed.
Stopping down has nothing to do with the autofocus problem mentionned independently by the two magazines. It's just about accurate focusing (which always takes place wide open), irrelevant of the actual aperture used.
dcmiller wrote:
You had the focus POINT on the left battery. The highest contrast by far is the right side of the right battery. That looks to be in focus. The focus sensor is considerably larger than the focus point.
Dude, you're talking complete nonsense. You're completely ignoring that for the same framing/AF point the lens is focusing progressively worse as we stop down. I'm not an idiot, I've been working with this stuff a long time.
"And yes, I do understand the concept of reduced microcontrast at wider apertureds that can lead to issues, but I don't believe this is exactly what is at work here."
Really. Since the aperture is fully open at the point the AF measurement is taken, tell me about those issues, Dude
dcmiller wrote:
Where is this example of degrading performance actually posted. I don't see it in your files. And what AF point were you using?
It's been posted at several points in varying conditions, and it's something that's quite evident in practical photos. I general only use the center point during tests (on the MKII, my XT and the store's XTi) although I've used off-center points to establish it isn't an oddity with just center point focus.
I'm not going to post the full transition from f/1.2 to f/2.8 since frankly at this point, I don't really feel compelled to convince anyone - the evidence is perfectly clear to me and has been backed up by plenty of other testers, including William Castleman.
Here's f/1.4 (the difference in color temp here is because of the ambient light leaking in):
f/2.8, using the same, center point One Shot AF (and yes, I've tried with AI Servo too):
Edited by Sam Bennett on Feb 10, 2007 at 10:07 AM GMT
Edited by Sam Bennett on Feb 10, 2007 at 10:08 AM GMT
You guys can debate the issue till the cows come home (Texas term? ). Whether it's my body, or lack of experience, or lack of contrast, or lack of depth of field, or paranoia, or even if I don't like the spec of the lens.
Bottom line is that my 2 copies did not perform to my satisfaction, and apparently neither do others as well. When Canon makes this lens to perform as ALMOST ALL OF THE REST of the lenses in their line-up, then I will most certainly get another one. Till then, it AIN'T happenin.
And I have shot test charts, real world object tests like the one attached, battery tests, people eye tests, and random walking around tests. ALL of the tests appear to have mostly the same results on 2 bodies and through 2 lenses and 3 calibration trips to Canon Irvine. Canon even was not happy with THEIR results in calibration. But was at a loss to explain or change it at the time...
I don't debate there are lenses out there that work correctly, and I'm happy that there are. However, there are also lenses that DON'T work correctly. Some guys seem to have MISSED that point in several hundred pages of posts in this thread.
Here is a series of test shots (cropped) (CENTER point AF ONLY) EXACTLY ON THE TOP OF THE BOX with barcode....from my last lens. Tripod mounted. AF on every frame, defocus, Re AF, take the shot.
Sam Bennett wrote:
It's been posted at several points in varying conditions, and it's something that's quite evident in practical photos. I general only use the center point during tests (on the MKII, my XT and the store's XTi) although I've used off-center points to establish it isn't an oddity with just center point focus.
I'm not going to post the full transition from f/1.2 to f/2.8 since frankly at this point, I don't really feel compelled to convince anyone at this point - the evidence is perfectly clear to me.
Sam, really, what you you do for a living You've been trying to convince everyone from day one. Including giving me a hard time. You probably have the most posts in this thread and they way I see it , I really don't know what has been contributed. And when you start to get sarcastic with people and refer to them as "dude" all credabilty goes out the window.
Sam Bennett wrote:
It's been posted at several points in varying conditions, and it's something that's quite evident in practical photos. I general only use the center point during tests (on the MKII, my XT and the store's XTi) although I've used off-center points to establish it isn't an oddity with just center point focus.
I'm not going to post the full transition from f/1.2 to f/2.8 since frankly at this point, I don't really feel compelled to convince anyone - the evidence is perfectly clear to me and has been backed up by plenty of other testers, including William Castleman.
dcmiller wrote:
The AF selected the hard shadow to the right of the post card, dude. Good illustration of how AF works.
And so why did it select the shadow only in the 2.8 shot? And why did it focus perfectly for all three f/1.4 shots and misfocus for the f/2.8 shots, even though the position of the point never changed?
ghuff wrote:
You guys can debate the issue till the cows come home (Texas term? ). Whether it's my body, or lack of experience, or lack of contrast, or lack of depth of field, or paranoia, or even if I don't like the spec of the lens.
Bottom line is that my 2 copies did not perform to my satisfaction, and apparently neither do others as well. When Canon makes this lens to perform as ALMOST ALL OF THE REST of the lenses in their line-up, then I will most certainly get another one. Till then, it AIN'T happenin.
And I have shot test charts, real world object tests like the one attached, battery tests, people eye tests, and random walking around tests. ALL of the tests appear to have mostly the same results on 2 bodies and through 2 lenses and 3 calibration trips to Canon Irvine. Canon even was not happy with THEIR results in calibration. But was at a loss to explain or change it at the time...
I don't debate there are lenses out there that work correctly, and I'm happy that there are. However, there are also lenses that DON'T work correctly. Some guys seem to have MISSED that point in several hundred pages of posts in this thread.
Here is a series of test shots (cropped) (CENTER point AF ONLY) EXACTLY ON THE TOP OF THE BOX with barcode....from my last lens. Tripod mounted. AF on every frame, defocus, Re AF, take the shot.
I would be concerned about overall sharpness of that lens. But these shots look very close to the minimum focus distance of the lens. The progression of focus looks fine. The camera may be digital, but it's an analog world.
fotografur wrote:
Sam, really, what you you do for a living
I'm a web-application by trade, and make a pretty decent income as a semi-professional sports photographer.
fotografur wrote:
You've been trying to convince everyone from day one. Including giving me a hard time. You probably have the most posts in this thread and they way I see it , I really don't know what has been contributed. And when you start to get sarcastic with people and refer to them as "dude" all credabilty goes out the window.
When people give me a snide tone after I've gone through a lot of careful testing, which has been confirm by other careful testing after it seemed like we had things settled is a bit deflating. I'm pretty sick of it at this point, so I think I will be winding down my participation in this thread.
I have been reading this thread from day one (also any tests I could see). I do not have this lens but originally had great hope associated with it. Now I am just wondering if Canon was too ambitious or incompetent.
My 50 1.2 perform just like my other lenses. And I know how and when to use manual focus.
The point of close distance is that AF error increases, but you knew that.
There is no posted evidence of a problem with this lens. The non-shaky images posted to prove a problem with this lens show the opposite. There is a misunderstandings about AF and basic of photography.
Gary's target test demonstrates that his lens is fine. The box and screws seems to have been photographed during an earthquake.
...chin up!!!
You've been trying to convince everyone from day one. Including giving me a hard time. You probably have the most posts in this thread and they way I see it , I really don't know what has been contributed. And when you start to get sarcastic with people and refer to them as "dude" all credabilty goes out the window.