Luminous Landscape Forum
Equipment & Techniques => Digital Cameras & Shooting Techniques => Topic started by: Tim Gray on June 19, 2007, 02:45:37 pm
-
http://www.robgalbraith.com/bins/content_p...cid=7-8740-9006 (http://www.robgalbraith.com/bins/content_page.asp?cid=7-8740-9006)
An update on his original publication where he noted AF issues on a pre-production model.
-
http://www.robgalbraith.com/bins/content_p...cid=7-8740-9006 (http://www.robgalbraith.com/bins/content_page.asp?cid=7-8740-9006)
An update on his original publication where he noted AF issues on a pre-production model.
[a href=\"index.php?act=findpost&pid=123792\"][{POST_SNAPBACK}][/a]
Tim,
Thanks for the article. It'll be interesting to see how Canon solves this issue.
-
This is particularly interesting since RG indicated that Canon were fully aware there was an AF glitch in the pre-production model he was given, which apparently hasn't been rectified in prodouction. With any luck it will be a firmware upgrade.
-
I think this is the real issue.. Can it be fixed with firmware? I just wonder why they did not change the firmware before the release of the final production units or at least post the firmware before shipping.. At least to let people know there is a fix.
This is particularly interesting since RG indicated that Canon were fully aware there was an AF glitch in the pre-production model he was given, which apparently hasn't been rectified in prodouction. With any luck it will be a firmware upgrade.
[a href=\"index.php?act=findpost&pid=123832\"][{POST_SNAPBACK}][/a]
-
It would seem that there is no correction available today.
My guess is that Canon was faced with a dilema like all the other brands before:
A. further delay the release of the camera - with the risk of being overtaken by Nikon or some other competitor
B. release a faulty product and try to fix the issue asap after the release.
It would seem that they have selected option B.
Regards,
Bernard
-
What about C) it cannot be fixed
It would seem that there is no correction available today.
My guess is that Canon was faced with a dilema like all the other brands before:
A. further delay the release of the camera - with the risk of being overtaken by Nikon or some other competitor
B. release a faulty product and try to fix the issue asap after the release.
It would seem that they have selected option B.
Regards,
Bernard
[a href=\"index.php?act=findpost&pid=123846\"][{POST_SNAPBACK}][/a]
-
It looks that Canon might have selected Option C: Put head in sand, pretend problem doesn't exist.
I doubt further delay would have run the risk of a competitor, like Nikon, overtaking Canon. Canon has a too commanding lead to be overtaken that easily. But selling a defective camera to pros whose livelihood depends on the camera is a sure way to lose it.
It would seem that there is no correction available today.
My guess is that Canon was faced with a dilema like all the other brands before:
A. further delay the release of the camera - with the risk of being overtaken by Nikon or some other competitor
B. release a faulty product and try to fix the issue asap after the release.
It would seem that they have selected option B.
Regards,
Bernard
[a href=\"index.php?act=findpost&pid=123846\"][{POST_SNAPBACK}][/a]
-
And might further explain why the 1Ds Mk III isn't out yet ...
-
Let's see...
5D Dirty viewfinders
24-105 Missing baffle
1Diii poor auto focus
Do I see a trend?
-
But selling a defective camera to pros whose livelihood depends on the camera is a sure way to lose it.
If so, then the market is flooded with "defective" cameras. :-/
Calling the camera "defective" may be too strong a statement.
Certainly, the autofocus problems mentioned by Galbraith are appalling, but there are so many other attractive "corrections" of earlier "defects" that it's still a very, very attractive camera.
My take on this is that Canon's new autofocus sensor is breaking some new ground here, and that Canon is still trying to find a way to use it best. It's a shame that they couldn't settle for the older autofocus system in the 1D MkIII while they ironed out the creases.
Fortunately for me, though, my living does not depend on the 1D MkIII. (If so, I'd be out of luck anyway, because of the very limited availability!) And if my living depended on being the kind of photographer who need the kind of autofocus precision that the 1D MkIII doesn't deliver in those circumstances, I'd probably be using the 1D MkII N.
-
HEY! Let's show a little respect here! My camera isn't defective, it's quirky, OK?
-
HEY! Let's show a little respect here! My camera isn't defective, it's quirky, OK?
[a href=\"index.php?act=findpost&pid=124032\"][{POST_SNAPBACK}][/a]
I'd say it's "circle of confusion is confused", or perhaps that it is "sharpness challenged," "differently abled," or "that it's a 400D trapped in a 1 series body"
The last one is a bit much, but wanted to run with the PCisms
-
Rob has updated his article... TOday June 21
http://www.robgalbraith.com/bins/content_p...cid=7-8740-9006 (http://www.robgalbraith.com/bins/content_page.asp?cid=7-8740-9006)
An update on his original publication where he noted AF issues on a pre-production model.
[a href=\"index.php?act=findpost&pid=123792\"][{POST_SNAPBACK}][/a]
-
I get the impression here that the release of the 1D3 was to mark some special celebratory event (can't remember what it was, though).
In such circumstances, one could expect there would be even more pressure than usual for the people working on the project to deliver on time. Failure would mean loss of bonuses.
If this is true, then any admission that the 1D3 has faulty focussing could affect the bonus payments of a number of Canon staff.
Tricky situation really .
-
And might further explain why the 1Ds Mk III isn't out yet ...
[a href=\"index.php?act=findpost&pid=123990\"][{POST_SNAPBACK}][/a]
Hmm. . . got mine two weeks ago.
What I've noticed is not a problem with the focus. Rather, there s a different logic to the camera's focus selection points. Previously, if I raised the camera to someone's face 9 out of 10 times, the points would select eyes/mouth, etc. to focus on. With the Mark III, it seems the camera will still try to use the lower points on the body.
-
Hmm. . . got mine two weeks ago.
You got a 1Ds MkIII last week??!!
Can we see a picture of it, even if it's in front of a mirror? You're doing pre-production testing? Give us the lowdown ...
-
He obviously got his 1D MkIII, but his point about a different logic to the 1D3 focus points is interesting.
Rob Galbraith has a lot of credibility, but there may well be a question mark about the non-scientific evidence he's presented. He's appealed to the 'real world shoot' concept, which is fine, but perhaps paid insufficient attention to the different focus settings that the 1D3 might require, compared to the 1D2.
-
I get the impression here that the release of the 1D3 was to mark some special celebratory event (can't remember what it was, though).
The EOS system is 20 years this year, and the Canon name is 70 years this year, which at least has something to do with the expectations.
I can hardly wait until 2012 and 2037.
-
He obviously got his 1D MkIII, but his point about a different logic to the 1D3 focus points is interesting.
Rob Galbraith has a lot of credibility, but there may well be a question mark about the non-scientific evidence he's presented. He's appealed to the 'real world shoot' concept, which is fine, but perhaps paid insufficient attention to the different focus settings that the 1D3 might require, compared to the 1D2.
As I read the article, they tested this quite thorougly.
We shot most things using the centre AF point only, though at times we switched to a couple of the outer AF points or all 45 AF points. The result? The only setting that had any real impact on the basic character of the autofocus was C. FnIII-2, AI Servo Tracking Sensitivity. We'll talk about it next.
(An analysis of EOS-1D Mark III autofocus performance (http://www.robgalbraith.com/bins/content_page.asp?cid=7-8740-9006), Q. How do you know the autofocus settings dialed into the camera aren't the problem?)
-
I got my 1DIII last week and so far it appears to be everything I could want for 75% of my shooting. The other 25% will fall on my 1DsII and its replacement in the Fall. My first impressions of the 1DIII are very good and the LOW NOISE at higher ISO's is a huge benefit for my work .
This AF issue I've been hearing about has me concerned. I've not been able to determine if my sample has the problem yet. Need to do some testing soon on the AF. Hopefully Canon will get the issue addressed SOON.
-
I certainly cannot confirm or deny anything that Rob's documenting on his blog. At this point, with 2700-ish frames on it, my MkIII seems to do very well. Any files that were out of focus, I could immediately explain as user error (focus point slips off the subject during action shots, typically, which is definitely a "driver" problem). The camera focuses so much faster than my 30D, I'm not used to being as rigorous about keeping the area of desired focus under the AF Point.
In looking at Rob's examples, I wonder if this increased AF speed is what's getting him into trouble? The shot with the runner coming at him - her arms are periodically in front of her body and in the coverage of the AF point, and she is a low contrast target where the camera is trying to focus (the black shirt). It seems to me that this camera is fast enough to try to refocus on the moving arm as it crosses the body, potentially creating a front focus situation in relation to the face....
I've not shot much action with it, yet, so I definitely have not done an exhaustive test of the gear, but... mine seems to be pretty solid. 2500+ shots out of the first battery charge!
-
In looking at Rob's examples, I wonder if this increased AF speed is what's getting him into trouble? The shot with the runner coming at him - her arms are periodically in front of her body and in the coverage of the AF point, and she is a low contrast target where the camera is trying to focus (the black shirt).
Galbraith has thought about the shirt:
As with other tests, we repeated several sequences with the athlete wearing a different shirt, to see if the camera's autofocus sensor was somehow reacting badly to a particular colour or pattern. From these tests and real-world shooting, we can see that whether the clothing is black, grey, white, blue, red, yellow, logo, no logo, it doesn't seem to matter: the autofocus problems are the same.
It seems to me that this camera is fast enough to try to refocus on the moving arm as it crosses the body, potentially creating a front focus situation in relation to the face....
Apparently, front focusing is not the problem:
Generally speaking, the camera tends to backfocus when it's focusing incorrectly.
Also, you should note that the problem appears when the subject is stationary, see the example with the goalkeeper (yellow shirt, number 1). You can't blame contrast, you can't blame arms or whatever moving in front of the shirt, and you certainly can't explain why the 1D MkII N performs far, far better under the same conditions.
However, if I (like you) switched from my 20D (you: 30D) to the MkIII, I'd still probably see a general improvement in autofocus speed and accuracy. But I've also seen the impressive performance of the MkII over the 20D, first hand.
-
Also, you should note that the problem appears when the subject is stationary, see the example with the goalkeeper (yellow shirt, number 1). You can't blame contrast, you can't blame arms or whatever moving in front of the shirt, and you certainly can't explain why the 1D MkII N performs far, far better under the same conditions.
I've been disregarding that test case - I have yet to handle a Canon camera that focuses reliably on static subjects when set to "AI Servo" (at least, when shooting hand held). My MkIII does it better than my 30D. I have not personally handled a 1DMkII or 1DMkIIn, so I have no experience with those.
However, if I (like you) switched from my 20D (you: 30D) to the MkIII, I'd still probably see a general improvement in autofocus speed and accuracy. But I've also seen the impressive performance of the MkII over the 20D, first hand.
[a href=\"index.php?act=findpost&pid=125973\"][{POST_SNAPBACK}][/a]
I agree - this is not my first experience w/ Canon's pro-level autofocus systems, though. I used an EOS-3 for a long time. Same/similar issues there, as well. With no personal experience with what appears to be considered the "golden" platform (the 1DMkIIn), all I can say is that my MkIII exhibits roughly the same behavior as every Canon camera I've used (and, interestingly, the D2X I've used, as well, for the most part - not exhaustively tested, though) - the MkIII is just faster about it.
If Canon can make it better, heck, I'm all for it.
-
I've been disregarding that test case - I have yet to handle a Canon camera that focuses reliably on static subjects when set to "AI Servo" (at least, when shooting hand held). My MkIII does it better than my 30D. I have not personally handled a 1DMkII or 1DMkIIn, so I have no experience with those.
According to Galbraith, the problem is not limited to AI Servo:
Focus can shift slightly but constantly at times when the subject isn't moving. Under certain conditions, the subject may not actually come into focus through a sequence of frames, even though the point of focus can be seen to be shifting throughout the sequence. This is true whether the camera is set to AI Servo and focus is active throughout the sequence, or when it's set to One Shot and focus is activated between each frame.
I agree - this is not my first experience w/ Canon's pro-level autofocus systems, though. I used an EOS-3 for a long time. Same/similar issues there, as well. With no personal experience with what appears to be considered the "golden" platform (the 1DMkIIn), all I can say is that my MkIII exhibits roughly the same behavior as every Canon camera I've used (and, interestingly, the D2X I've used, as well, for the most part - not exhaustively tested, though) - the MkIII is just faster about it.
If Canon can make it better, heck, I'm all for it.
If we are to believe Rob Galbraith, Canon made it better in the previous model, 1D MkII (N).
Keep in mind the shooting conditions where the problems occur; the camera appears to handle autofocus superbly in other conditions.
The weirdness is that this happens under sunny and warm conditions, exactly the kind of conditions where one would expect autofocus to behave normally.
-
RG updated his article today July 2 and it is not pretty. Basically still many AF issues that are not solved by different CF parameter changes and with a 3rd production camera.
According to Galbraith, the problem is not limited to AI Servo:
If we are to believe Rob Galbraith, Canon made it better in the previous model, 1D MkII (N).
Keep in mind the shooting conditions where the problems occur; the camera appears to handle autofocus superbly in other conditions.
The weirdness is that this happens under sunny and warm conditions, exactly the kind of conditions where one would expect autofocus to behave normally.
[a href=\"index.php?act=findpost&pid=126053\"][{POST_SNAPBACK}][/a]