p.3 #6 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Steven Kelley wrote:
Andrew J,
I thought I read in RG's report that frame rate had no effect on % OOF shots. Am I correct on that or am I dreaming?
"In testing Canon's hardware and firmware changes in the EOS-1D Mark III, we had the camera set to the centre AF point exclusively, and varied only the frame rate (10fps, 8fps) and AF Microadjustment settings. As noted earlier in the article, the AF Microadjustment setting does make a small difference, but the frame rate doesn't."
p.3 #7 · ['NEW Fix' UPDATE!] - MkIII AF still broken
I wonder if shooting in Silent Mode would reduce this? That mode totally changes the way the mirror is slapped around. It should be pretty easy to compair staic shots in normal one shot and silent mode.
Z
p.3 #8 · ['NEW Fix' UPDATE!] - MkIII AF still broken
FretNoMore wrote:
In some cases like the logo on the t-shirt maybe there can sometimes be other explanations though, the edges of the stitching/printing is perhaps not all that sharp, maybe there's lens CA at the high contrast edge between blue and white, maybe there's sensor blooming because of local overexposure (though CA and blooming usually have a color to it). Just some ideas, I'm not saying this explains anything or everything of course.
That was the first thing I evaluated when I noticed the 'ghosting' effect, as I wasn't even sure I was seeing it, initially. There are quite a few in-focus images that show the nature of the logo on the shirt; in the closer-up images, you can even make out the clear 'edges' of the decal that was applied, and the edges of the white text is of uniform contrast/sharpness on all sides (top/bottom, left/right).
Note that there are a few images where the ghosting appears on a different axis, but I think this only was evidenced on the landscape-orientation shots.
------------------------------------------------------- Nathan Hobbs wrote:
if the lens had any focus movement during the process of an exposure it would cause the blurring, maybe the lens cant put the brakes on and stop the movement fast enough for ten frames a second. I would like to see Rob compare the two cameras dialed down to equal speeds 8 frames a second.
This makes some sense that there could be some sort of timing error somewhere in the system, but what I can't fathom is why it wouldn't affect the image in a 'non-directional' nature, i.e. the logo should be blurry on all sides, not just one side of one axis.
------------------------------------------------------- David Svensson wrote:
I think it is a combination of longitudinal spherochromatic aberration, and one very slightly skewed lens element. Possibly an issue with a reflex from the front filter (if present). Notice in the sharp picture, how there is a dark ghost text just where the bright ghost text is in the OOF picture. That is tell-tale of longitudinal spherochromatic aberration.
If this were the case, I would think the the MkIIn images would also exhibit the effect, unless of course it's a lens-mount issue on the MkIII. Mine feels pretty typical, for what that's worth.
------------------------------------------------------- maljo wrote:
Wait a second... It affects every single 1DM3 no matter when manufactured, because the 'fix' doesn't fix the problem. No 1DM3 is good.
And you just may (likely) be able to lump the 1Ds MkIII into that, unless of course Canon is able to (rather quickly) pull a rabbit out of its collective hat!
------------------------------------------------------- Andrew J wrote:
Is one workaround to set high speed at 8 shots per second?
Does 10 a second start a feedback loop with stabalization or something else?
I shot at lower frame rates way back in the beginning, but those inconsistent AF results very well could have been due to the submirror. I'd have to go back and check, as I feel that the submirror affects the chosen focal point in the frame, whereas this is a different beast altogether.
------------------------------------------------------- samd12 wrote:
I agree but you may want to be careful what you post. Quote from the bottom of the article: "These photos are for personal viewing and printing only. They may not be republished in any form without the permission of the copyright holder. This includes the posting of these photos onto another server."
I'll be happy to remove the images at Rob's request; I did intentionally leave his copyright intact for everything I posted, and I would not think that, given the nature of the thread, RG would consider this to be any sort of intentional misuse on my part. What good are the images if we can't 'use them' to evaluate his results?
p.3 #9 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Thank you very much for your consistent and lucid argumentation. Very interesting.
I was involved in some laboratory tests of MkIII's at a university workshop and we ended with the same questions:
1. What kind of malfunction could result in the unidirectional blur that seems to be typical for the camera?
2. Might there be a multifactorial disorder resulting from hypercomplexity in real-time processing?
After the recent discussion in these days it became clearer, that we are dealing with a systematic design error not with a production flaw (i.e. the submirror-fix theme).
The unwillingness of the manufacturer to communicate the apparent dysfunction is without precedent in a field of professionally used equipment. It is a pity that the above reflections and conversations are conducted completely separated from the company. It can be assumed that Canon was able to replicate the malfunction under laboratory conditions at least after Galbraith's last tests. Nevertheless since the introduction of the model they did never publish a product advisory statement and did not directly address the professionals, that depend on their products as business investments. Who of those could bona fide go on investing in EOS 1Ds Mark III bodies sharing the identical AF-technology in this situation of downright ignorance of the manufacturer against its clients?
p.3 #11 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Arianne Dubois wrote:
I was involved in some laboratory tests of MkIII's at a university workshop and we ended with the same questions:
....
After the recent discussion in these days it became clearer, that we are dealing with a systematic design error not with a production flaw (i.e. the submirror-fix theme).
The unwillingness of the manufacturer to communicate the apparent dysfunction is without precedent in a field of professionally used equipment.
Sounds very interesting. What was the university context, please? (applied optics, computer systems, marketing?)
I think it is a matter both of the design, and of production varability.
Sad to say, but I donīt think there is any historical lack at all of companies being unwilling to communicate openly about deficiencies in pro equipment.
p.3 #15 · ['NEW Fix' UPDATE!] - MkIII AF still broken
I would still say that those images are the result of the 1D-3 being ever-so-slightly (maybe 3 - 4") front-focused on the runner.
And I still think that if the 1D-3 had been set to a +1 or +2 (or some positve number), it would've resulted in a greater percentage of the shots appearing in focus.
But it's impossible to say for certain since that test wasn't done. Too bad they didn't try something like a +2 instead of a -5.
Jeff wrote:
And, the last thing I took a look at tonight was prompted by DavidP's thread, and relates to accuracy of comparison sequences. In the 'slow' series, I was able to find the runner in virtually identical positions, in nearly the exact same spot on the track. The bottom 'loupe' view shows two things: that the focal plane was in a nearly identical position, and the runner was no more than 1 to 2 inches fore/aft of the exact same position in both images (note the slight softness of even the specular highlights in the MkIII image).
More importantly, the upper loupe view shows a significant discrepancy in the sharpness of the runner's face, even giving some wiggle room for the runner being up to 2" off in distance to the camera. It's this kind of result that MkIII owners familiar with the MkIII's AF problems readily admit to being anything but 'normal' for any camera. There is simply no logical explanation for it, other than some technically complex hardware issue. ...Show more →
p.3 #16 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Arianne Dubois wrote:
"...After the recent discussion in these days it became clearer, that we are dealing with a systematic design error not with a production flaw (i.e. the submirror-fix theme).
The unwillingness of the manufacturer to communicate the apparent dysfunction is without precedent in a field of professionally used equipment. It is a pity that the above reflections and conversations are conducted completely separated from the company. It can be assumed that Canon was able to replicate the malfunction under laboratory conditions at least after Galbraith's last tests. Nevertheless since the introduction of the model they did never publish a product advisory statement and did not directly address the professionals, that depend on their products as business investments. ...Show more →
I don't think that we can assume that Canon knows of all these issues before the collective 'we' do, although it does seem fairly certain that they knew the camera had AF issues before its release. How they communicate what is going on on a day-to-day basis is a different matter, but I have to point out that they have indeed formally documented the submirror issue on their website, on the MkIII page.
I think it is reasonable to assume at this point that Canon thought they had the problem fixed, but only now that one issue is resolved does it become apparent that there was (all along) a distinctly separate problem that results in a different sort of image quality issue. I can only hope that they will take this issue as seriously as the initial reports of problems, because I have no intention of possessing a camera of this caliber that is half-baked in any way.
Interesting that your school has 'already' done tests on a specific AF flaw that really only could have been separated out from the submirror issue in the last 48 hours (or by anyone possessing a 'Blue Dot' body). To date, I had not heard anyone (RG included) comment on this 'secondary ghost image' issue. With the previous submirror issues, it would have been very difficult to even describe as a discrete problem until only very recently. I'd be very interested to see your school's test results and methodologies, if you'd care to share the results.
p.3 #17 · ['NEW Fix' UPDATE!] - MkIII AF still broken
My 1dIII does this........does anyone elses?
In single shot mode, it will lock onto the subject, and "most" of the pics will be in focus. There is that occassional one that is soft or just completely OOF for no reason, but 8 out of 10 are in focus.
In ai servo shooting a stationary subject from a tripod, the focus point is constantly shifting. The distance ring on the lens constantly moves back and forth slightly, and looking through the viewfinder you can see the focus shifting. If it can't lock onto a stationary object in ai servo, how's it going to consistantly track a moving object?
My camera is in the affected range and is scheduled for the mirror adjustment. I've tried fw v1.1.0 and v1.1.3, and the focus acts the same. I've adjusted the focus sensitivity just to see if that affected the instability, and it didn't.
And speaking of the "ghost" images seen in RG's photos. I've seen that as well. Today I was taking shots of a focus calibration chart. The numbers on the chart that were in front of the focus area showed ghost images, the ones behind the focus area were just fuzzy, no ghost images. I took dozens of shots and the ghost images were always present in the same way, on the front end of the focus area, but not the back.
p.3 #19 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Steven Kelley wrote:
My 1dIII does this........does anyone elses?
In single shot mode, it will lock onto the subject, and "most" of the pics will be in focus. There is that occassional one that is soft or just completely OOF for no reason, but 8 out of 10 are in focus.
My 1D2's did that. So did my 1D's. So did my D30. I'm pretty sure my 1D3's also do it. As did my EOS-3.
p.3 #20 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Steven Kelley wrote:
My 1dIII does this........does anyone elses?
In single shot mode, it will lock onto the subject, and "most" of the pics will be in focus. There is that occassional one that is soft or just completely OOF for no reason, but 8 out of 10 are in focus.
In ai servo shooting a stationary subject from a tripod, the focus point is constantly shifting. The distance ring on the lens constantly moves back and forth slightly, and looking through the viewfinder you can see the focus shifting. If it can't lock onto a stationary object in ai servo, how's it going to consistantly track a moving object?
My camera is in the affected range and is scheduled for the mirror adjustment. I've tried fw v1.1.0 and v1.1.3, and the focus acts the same. I've adjusted the focus sensitivity just to see if that affected the instability, and it didn't.
And speaking of the "ghost" images seen in RG's photos. I've seen that as well. Today I was taking shots of a focus calibration chart. The numbers on the chart that were in front of the focus area showed ghost images, the ones behind the focus area were just fuzzy, no ghost images. I took dozens of shots and the ghost images were always present in the same way, on the front end of the focus area, but not the back.
Has anyone else seen this with their camera?...Show more →
I'm on my 2nd 1D3 the latest 'blue dot"
I have experienced the same thing.
The oscillation is worse with the 300 than a 70-200.
Also approximately 1 out of 10 single shots are off.
