p.14 #2 · ['NEW Fix' UPDATE!] - MkIII AF still broken
I am confused why you'd evaluate focus at 1:2 since that it basically 200% zoom which adds distortion. It does appear to be a bit soft, but hard to say how that would look in print.
p.14 #3 · ['NEW Fix' UPDATE!] - MkIII AF still broken
jkurkjia wrote:
Jeff, what is the static AF accuracy of that lens? Perhaps you could comment on static AF accuracy pre and post AF fix (if you don't mind), thanks.
Regards,
Joe Kurkjian
It was hit and miss for no apparent reason. Looked pretty good today, with a few issues noted. Not as bad as the AI Servo stuff, though.
p.14 #4 · ['NEW Fix' UPDATE!] - MkIII AF still broken
I see that you added some 1:1 examples too.. So nevermind on my last message. That last one definitely looks bad. Is there any part of that image that is in focus?
p.14 #5 · ['NEW Fix' UPDATE!] - MkIII AF still broken
apdieb wrote:
I am confused why you'd evaluate focus at 1:2 since that it basically 200% zoom which adds distortion. It does appear to be a bit soft, but hard to say how that would look in print.
What shutter speed were these at?
Thanks for posting examples.
Looking at them at 50%, not 200%. It would look like crap printed larger than 4x6, trust me.
There wasn't much light, so it was ISO 800 at between 1/500th and 1/1000th. Not great, but irrelevant for focal plane, which is what I was evaluating.
I'm also on my wife's crappy little work PC tablet, so it's not very efficient to do this...
p.14 #6 · ['NEW Fix' UPDATE!] - MkIII AF still broken
apdieb wrote:
I see that you added some 1:1 examples too.. So nevermind on my last message. That last one definitely looks bad. Is there any part of that image that is in focus?
Only some snowflakes...
That's the great part about these conditions, you can indeed see that something somewhere was in focus. Canon will probably say those flakes were in the field of the AF sensor, and it's so good it thus it locked onto them...
p.14 #7 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Man, this is depressing. While I no longer own a 1D3, I own a 1Ds3 and wonder how "related" they are. I don't think I'll ever find out given my subject matter, but I guess I'll always wonder.
p.14 #9 · ['NEW Fix' UPDATE!] - MkIII AF still broken
stanj wrote:
Man, this is depressing. While I no longer own a 1D3, I own a 1Ds3 and wonder how "related" they are. I don't think I'll ever find out given my subject matter, but I guess I'll always wonder.
p.14 #10 · ['NEW Fix' UPDATE!] - MkIII AF still broken
you know it is funny I shot a soccer match last weekend with my two new BD M3's and my wife looked at the pictures and asked when I was going to get my cameras fixed. I wonder if it is me or the body? I never had her say that before.
p.14 #12 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Photon wrote:
I would have thought that for the small changes in "angle of aim" that result from our natural trembling (i.e., what we're trying to correct) the IS system would work best by offsetting a lens group from the axis but minimizing any resulting tilt of the focal plane. I can see, however, that it could be difficult to perfect the optical design, and even a perfect axial shift could result in a tipping of the focal plane just because the curves of the various lens elements are no longer aligned. I have neither the expertise nor the software to do a ray trace experiment to verify this.
It also seems to me that since IS works so well when things are right, it must not induce significant tilt of the plane (barring some problem) or we wouldn't be able to get such sharp hand held images of brick walls! ...Show more →
Photon, I believe that the IS does tilt the plane of focus just a bit but this is actually a good thing because the camera has also been tilted.
With regard to Jeff's photo #4 I previously said it was too soft overall. I had a look at it in a photo editor and applied lots of sharpening and it is then quite apparent that the left side sharpens up while the right side does not. Jeff would have seen this on clearly on his original.
I'd still like confirmation from you, Jeff, that the 70-200 lens is currently and consistently producing a uniformly sharp image on any other camera. If it is then the 1D3 must be defective. If it is not then the lens is defective but even so the 1D3 may also still be defective.
It's an interesting problem (for those of us who do not experience it first hand) because for the camera to be at fault it must either have a free-floating image sensor that produces an intermittent misalignment or else (say, if the lens body is warped) every image it takes with every lens should be showing the same misalignment effect. Maybe you can check to see if the camera rattles when you shake it - if it does then that'll be the sensor
p.14 #13 · ['NEW Fix' UPDATE!] - MkIII AF still broken
oh jeez. Jeff, sorry about your problem. i hope you do give them an ear-full. not that it'll do much (from my experience).
i'm back to using my 5d. i haven't changed my technique or my lenses (mostly 35L, 85L), and the shots i took yesterday are GORGEOUS. sure, i had some out of focus shots, but everything was easily explained. i.e. my 5d is repeatable, predictable and dependable. qualities that the schizo 1d3 may never (for some people) possess by the sounds of it. now that 5d is a great camera!
p.14 #14 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Alan321 wrote:
I'd still like confirmation from you, Jeff, that the 70-200 lens is currently and consistently producing a uniformly sharp image on any other camera. If it is then the 1D3 must be defective. If it is not then the lens is defective but even so the 1D3 may also still be defective.
It's an interesting problem (for those of us who do not experience it first hand) because for the camera to be at fault it must either have a free-floating image sensor that produces an intermittent misalignment or else (say, if the lens body is warped) every image it takes with every lens should be showing the same misalignment effect. Maybe you can check to see if the camera rattles when you shake it - if it does then that'll be the sensor ...Show more →
Alan, I still suspect there's a possibility that the lens could have been somehow compromised as soon as I started shooting with the MkIII. i.e. I can neither confirm nor deny it. These tests yesterday were for the primary purpose of evaluating the 'fix', to determine if the camera was able to focus on what it should have been trying to focus on. I didn't shoot any flat-field images, and I don't have a tripod with me, as we are effectively away from home with limited computer capability for the next week, unfortunately. But, on the brighter side, 'EM' Santa has something with 3 legs and a head under the tree for me!
p.14 #15 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Incidentally, there certainly seems to be significant anecdotal evidence to suggest that some MkIII's are being repaired to a more functional state, so perhaps my camera only getting one part replaced (and not two) is the issue. Hopefully the next trip to VA will do the trick, at least for this errant AF issue...
p.14 #17 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Wickedfn4u wrote:
you know it is funny I shot a soccer match last weekend with my two new BD M3's and my wife looked at the pictures and asked when I was going to get my cameras fixed. I wonder if it is me or the body? I never had her say that before.
p.14 #18 · ['NEW Fix' UPDATE!] - MkIII AF still broken
lidesun wrote:
This does mean sonetihng.....
Yeah Lid I just dont know what to think. This was the first game I had to shoot with 2.8. The other 2 or 3 were at 4 or 3.5 and I they looked OK. That is why I have to wonder if it was me and the 2.8 or the body still.
p.14 #19 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Whatever you guys do, please don't allow my results to make you guys start second-guessing your own results. It's entirely possible that my camera's AF will work properly after one more trip to Frankenstein's laboratory, and I haven't ruled out my 70-200/2.8L IS as a confounding factor.
Regards,
Jeff
Dec 22, 2007 at 12:58 PM
Red Grainger Offline • • [X]
p.14 #20 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Jeff,
When all is said and done, are you feeling that the only way Canon can deal with these issues is to shorten their product cycle and introduce a new series 1 body (Series 1 Mark IIIN)? I need another reliable body so have just sold my Mark III and bought a 40D. I won't touch the 1Ds Mark III until I am certain that the autofocus is not also an issue in that body. Fortunately the only limiting parameter for the publications I deal with is the requirement to use a D-SLR of 10 or more megapixels. The 40D meets that requirement and seems to be able to focus. So for the time being I'll sit tight but I don't like the options Canon is giving me. Unless they telegraph their intentions to issue a new series, absent these autofocus issues, many pros will be turning over their much loved glass for sale. I recently read an interesting observation on another forum, the reader likened Canon's current attitude to that of the Mercedes Benz automobile about 10-15 years ago. Once a company begins to slip, it can be hard to stop.
Have a good Holiday, Cheers-Red
