p.10 #1 · ['NEW Fix' UPDATE!] - MkIII AF still broken
I look forward to your test shots with your camera after the fix and hopefully, to Rob's reply. I wish there would be an immaculate gathering of opinions so I can buy one with peace of mind.
p.10 #2 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Yakim Peled wrote:
I look forward to your test shots with your camera after the fix and hopefully, to Rob's reply. I wish there would be an immaculate gathering of opinions so I can buy one with peace of mind.
Happy shooting,
Yakim.
I agree completely with Yakim. I want a new camera, with better ISO capabilities and better focusing than my mark2. But until I'm convinced they are both better I'll stick with my older models.
p.10 #3 · ['NEW Fix' UPDATE!] - MkIII AF still broken
For those of you who prefer not to wade through the last 4 pages of very inefficient information transfer, here are the images that I tried to present last night, in order:
p.10 #7 · ['NEW Fix' UPDATE!] - MkIII AF still broken
And finally, #4 (this is a 50% screen grab, note how the rocks are sharper on the left side than the right [a reversal in the effect]; it is glaringly obvious when viewed at 100%, thus a good example of how shots that look good viewed at screen size often are not):
Edited by Jeff on Dec 19, 2007 at 06:58 AM GMT (Reason: typo)
p.10 #9 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Where was the focal point for #4 Jeff?
The left hand rock pillar is obviously the sharpest place but equally the right hand pillar looks to be far nearer to you than the centre and left and at f2.8 I would expect that given the minimal DoF.
p.10 #10 · ['NEW Fix' UPDATE!] - MkIII AF still broken
jkurkjia wrote:
Sounds like we won't know anything for sure until you get the body back and test a few more lenses, including other bodies, to confirm exactly what is going on. I'll assume you will report back with your findings in this thread rather than start a new one; this will help to give us all an opportunity to help out (assuming we can do it in a positive manner).
Joe, late last night I also looked through 30D images using the 70-200/2.8L IS, and it did not exhibit this odd 'plane of focus' effect (nor did the 1Ds). I'm starting to believe that there is some sort of strange incompatibility between the 1D MkIII and IS, which is visually exacerbated the longer the glass. I can't fathom what the cause could be, but the IS mechanism moving around in the lens would seem to be the most likely culprit. Whether vibration/mirror slap or some sort of harmonic resonance is related I have no idea. For that matter, it could still be my particular lens, but I'm starting to doubt it due to the lack of similar effect using both the 30D and 1Ds (I will test the 30D to confirm that tomorrow).
I had intended on referring Chuck Westfall to this thread in my next communication to him, but after last night that would be rather embarrassing for FM.com (for obvious reasons), so I guess I'll have to continue to communicate with him one on one.
Regards,
Jeff
PS: I've requested that Fred delete the 'errant posts' from this thread, because as of last night it (rapidly) became difficult to find the signal through the noise. Hopefully I won't have to start a new thread and re-invent the proverbial wheel...
p.10 #13 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Jeff wrote:
I'm starting to believe that there is some sort of strange incompatibility between the 1D MkIII and IS, which is visually exacerbated the longer the glass.
p.10 #14 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Jeff wrote:
Joe, late last night I also looked through 30D images using the 70-200/2.8L IS, and it did not exhibit this odd 'plane of focus' effect (nor did the 1Ds). I'm starting to believe that there is some sort of strange incompatibility between the 1D MkIII and IS, which is visually exacerbated the longer the glass. I can't fathom what the cause could be, but the IS mechanism moving around in the lens would seem to be the most likely culprit. Whether vibration/mirror slap or some sort of harmonic resonance is related I have no idea. For that matter, it could still be my particular lens, but I'm starting to doubt it due to the lack of similar effect using both the 30D and 1Ds (I will test the 30D to confirm that tomorrow). ...Show more →
Jeff, can you take a shot with your 1D3 and 70-200 with IS on and then another of the same scene with same settings with IS off?
Surely this would prove whether or not the issue is due to IS....
p.10 #15 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Alistair Watson wrote:
Where was the focal point for #4 Jeff?
The left hand rock pillar is obviously the sharpest place but equally the right hand pillar looks to be far nearer to you than the centre and left and at f2.8 I would expect that given the minimal DoF.
That's a good question Alistair, and I made very sure of this one last night before I posted it (I should have included the loupe, but ran out of time, what with all the errant rhetoric being slung about).
The focus was on the tree in the center, and both sides of the rock were roughly equidistant to my position. Even if what you suggest were true, you should be able to move 'backward' on the right hand pillar of rock until you find the focal plane, but you cannot. There is absolutely nothing in focus in the right half of the frame, but there should be. At that distance to subject, the left side of the frame exhibits the depth of field one would expect with this lens and f/stop combination; the right is inexplicable for normally-operating equipment, in my opinion.
Now, turn this picture vertically, and let's analyze this in relation to the oft-repeated (by myself and many others) 'running sequence using AI Servo'. (Lots of 'ifs' here, so forgive the following) Now imagine the bottom half of the image being the sharp portion, and the top part being the out-of-focus portion. You would be able to precisely find a plane of focus near the runner's feet, where it is very easy to determine due to the grass. Some of the runner's legs/torso may be sharp, as you near the halfway point of the frame. As you move toward the shoulders and face, it becomes harder (and in some cases impossible) to find where the plane of focus goes through the 'meat' of the image. Nothing near the face is obviously in focus, and what is 'most in focus' is still quite soft.
I cannot tell you how many times I've seen this exact effect, and it can be seen in many of RG's sequences. Many people have noted not being able to find something in a given image that is critically sharp, with no obvious cause. Many of these 'runner' sequences are heavily scrutinized for front- and backfocusing, and using the face (the primary subject) it is often difficult or impossible to determine, because nothing in that half of the frame is sharp.
It is thus entirely possible that much of what has been attributed to misfocusing (a la the submirror) was actually this 'plane of focus' effect in concert with the (pre-Blue Dot) jumpy AF, making it very, very difficult to isolate as a discrete problem. Whether or not the 'ghosting' issue is the same source is anyone's guess. Time will tell now that repaired cameras are starting to become more numerous, and I admit that these effects are sometimes very hard to see unless viewed at 1:1 or printed very large.
Respectfully,
Jeff
PS: Writing the above has reminded me of several instances in which I manually focused the MkIII, only to end up with a soft image. Going to have to scratch my head on this one, and see if I can find examples...
p.10 #16 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Yakim Peled wrote:
I wholeheartedly hope you are wrong.
Me too, Yakim, me too...
DavidP wrote:
Hmmm, I wonder if RobG was shooting with IS on or off during his tests?
I've wondered the same thing, but one would think 'IS off', but I don't think he ever said. If he'd return my e-mail I could ask him, but it looks like that won't happen...
Alistair Watson wrote:
Jeff, can you take a shot with your 1D3 and 70-200 with IS on and then another of the same scene with same settings with IS off? Surely this would prove whether or not the issue is due to IS....
I could if it weren't in Virginia, and due to XMas it may be awhile before I get it back in my hands, unfortunately. Again, I find it vexing that neither the 1Ds nor the 30D ever exhibited this effect.
p.10 #17 · ['NEW Fix' UPDATE!] - MkIII AF still broken
The thing is... if this is caused by IS moving in ways it shouldn't be moving... Can you imagine how hard it is for the camera to focus? Regardless of the focal plane itself being crooked in the resulting image... the AF sensors are having to deal with that too...
So.. what happens if you were to turn off IS? Would photos of the runners etc... all start to show up in focus?
Assuming Rob and/or others did all their testing of AI servo with IS on... perhaps some tests with it off are in order...
p.10 #18 · ['NEW Fix' UPDATE!] - MkIII AF still broken
Interesting points Jeff.
I don't have the original CR2 file for this but this is the image I took for dvarnav since he likes shooting walls in AI-Servo mode!
My new blue dot 1.1.3 firmware 1D3 with 500/4 IS 'on' on a Wimberley head. As far as I remember f4 and sharp across the frame. Perhaps I should try it with a 2.8 IS lens.
p.10 #19 · ['NEW Fix' UPDATE!] - MkIII AF still broken
DynoMoHum wrote:
The thing is... if this is caused by IS moving in ways it shouldn't be moving... Can you imagine how hard it is for the camera to focus? Regardless of the focal plane itself being crooked in the resulting image... the AF sensors are having to deal with that too...
True, I never really considered that aspect of it... Maybe that AF system is better than we think!
Alistair Watson wrote:
I don't have the original CR2 file for this but this is the image I took for dvarnav since he likes shooting walls in AI-Servo mode! My new blue dot 1.1.3 firmware 1D3 with 500/4 IS 'on' on a Wimberley head. As far as I remember f4 and sharp across the frame. Perhaps I should try it with a 2.8 IS lens.
