Has anybody devised a sure-fire test method to reproduce these issues?
I'm hearing all sorts of opinions from amateurs and professionals who all seem to say different things. Personally, I can't say I'm 100% happy with the servo performance on the Mark III but I'm still convinced that it's me.. or my 300 2.8 needs a calibration/service - haven't had a chance to test this extensively.
I still get the shots I (largely) need however.
People coming from the 40D/30D loved the 1D III out of the box! Those coming from the 1D II complained.
Vincent Laforet continued to make fantastic images of the Olympic Games in Beijing last year with his 1D III bodies. We didn't hear a word from him about the AF issues (mind you, he's in bed with Canon to some degree so this may or may not affect things here).
The other thing to consider is while Canon has released firmware and fixes for these cameras to resolve *known* issues, there are always many unknown issues and bugs that are also solved in firmware and software releases that we never hear about. I see a lot of this in my other line of work (IT network infrastructure) with our Cisco equipment) where firmware releases do fix issues that are not made publically available in release notes/documentation.
My 1D III had its flange depth adjusted when it went in for its AF adjustment.
Paul B wrote:
Obviously, anyone with "evidence" (i.e. a Canon employee, which I'm not) would not share the info for obvious reasons. But I've seen examples of similar things in this arena and it's how things get played, for better or worse. And my intention is not to lay blame on someone (i.e. Canon. I can't imagine Canon continuing to advertise on RG's site with this sort of thing going on. Certainly doesn't mean it will always be that way but when a big political issue like this develops and stays for some time, budgets tend to get cut when they are perceived as benefiting - rightly or wrongly - the wrong message.)
My main point, though, is that there seems to be an automatic assumption that this produces more traffic which produces a net financial increase. The financial piece, itself, is much more complex (i.e. firms support analysts in various ways and those can be cut off.) And then there's the question of how much a given firm cooperates with an analyst in providing info on future products. The latter is critical in terms of providing a given analyst a technical heads-up on info, which in turn draws more people to a given site or research firm (because that firm is perceived by the public as having greater "expertise" in the area than its competitors.) So I think it's a bit questionable to assume that someone like RG is doing this all simply to drive traffic (if that were true, why are reviewers such as DPR and others, not also blasting the MkIII to "drive traffic" and make lots of money?) ...Show more →
Ok, no evidence of conspiracy. Much more simply, do you have evidence that Canon had advertising on RG's site which was subsequently withdrawn (for whatever reason). If not, it's more than questionable to be making the assertion that Canon financially "punished" Galbraith.
EvilZardoz wrote:
Vincent Laforet continued to make fantastic images of the Olympic Games in Beijing last year with his 1D III bodies. We didn't hear a word from him about the AF issues (mind you, he's in bed with Canon to some degree so this may or may not affect things here).
Personally I am quite happy with my photo take from here, although I am very disappointed with the performance, or lack thereof, of some of my Canon gear. The fact that many integral moments and photos were missed entirely because of camera malfunctions, has made me really reconsider, as many of the sportsshooters in the industry, if I should make the switch to black lenses.
EvilZardoz wrote:
Has anybody devised a sure-fire test method to reproduce these issues?
I'm hearing all sorts of opinions from amateurs and professionals who all seem to say different things. Personally, I can't say I'm 100% happy with the servo performance on the Mark III but I'm still convinced that it's me.. or my 300 2.8 needs a calibration/service - haven't had a chance to test this extensively.
I still get the shots I (largely) need however.
People coming from the 40D/30D loved the 1D III out of the box! Those coming from the 1D II complained.
Vincent Laforet continued to make fantastic images of the Olympic Games in Beijing last year with his 1D III bodies. We didn't hear a word from him about the AF issues (mind you, he's in bed with Canon to some degree so this may or may not affect things here).
The other thing to consider is while Canon has released firmware and fixes for these cameras to resolve *known* issues, there are always many unknown issues and bugs that are also solved in firmware and software releases that we never hear about. I see a lot of this in my other line of work (IT network infrastructure) with our Cisco equipment) where firmware releases do fix issues that are not made publically available in release notes/documentation.
My 1D III had its flange depth adjusted when it went in for its AF adjustment....Show more →
I hear you- in my experience the AF issues came up in all sorts of weird situations. I took it out for 3 days straight of nature and wildlife shooting out on the Channel Islands with my 300 2.8 IS, all in servo mode, it performed like a true champ. Sequences were well in focus. Then, on a relatively warmer day (this was back in the winter... that day it was.... 70 degrees?) back on the mainland, with my 24-70 (with which it never had a problem with), it started front focusing in one-shot mode. And then, when I got home, that problem went away.
It came up intermittently... worked fine most of the time, but it was just a pain to have to live with that inconsistency and uncertainty. I have backup gear but you know, really, your primary gear should at least have to be consistent, because those shots you miss, could be the ones you need.
My pronlems with RG's reviews is his complete lack of info on the CF setting he used.
This is the most customizable camera Af system ever delivered to the photographic community and he did not mention once what or how to set the camera up. It was some time before the Canon white paper was released after his initial review.
So it is partly Canons fault for many reasons.
Poor QC and system testing prior to shipping bodies. Given how fast and accurate the AF is it is totally intolerant of any defect or mis calibration. Totally.
Not enough info available to users about the most complex AF system ever released regarding set up and CF use.
Having just completed a major 6 day airshow in very similar conditions so the previous one but with the 1Dmk3 and 1Ds Mk3 the AF accuracy is so far superior to the mk2N as to leave it wallowing in the background.
Fast fighter jets with low contrast paint schemes against a blue sky is the hardest AF target ive ever photographed and the MK3's are excellent.
Not perfect but there will never be a perfect AF system and the large percentage increase in sharp shots with the 3's over the 2n is all the evidence i need that its the best AF system ever released when set up and operating properly.
Paul B wrote:
I think it's a bit questionable to assume that someone like RG is doing this all simply to drive traffic (if that were true, why are reviewers such as DPR and others, not also blasting the MkIII to "drive traffic" and make lots of money?)
No need.
The first thing most folk do when seeing a "bombshell" like RG's opinion is visit every other site - like DPR, like here - to see what support (or not) exists for his opinion.
So whether there's anything on DPR about this or not, they'll still get the traffic (and very likely more traffic - that's the thing about discussion forums).
The more I dig into his photos and review them the more convinced I become that he doesn't think about the custom functions in relation to the shoot. In fact at best he can't understand them if these examples were to be an objective test and at worst he or whoever took the shots intentionally selected the settings to get the worst possible focus through a sequence.
First he mixes up the III4-1 setting as already discussed. Then if you take the 15 minutes to download the massive files and review the shots you find more issues with the results and his conclusions. For example from the time stamps he's consistently getting 10 fps in shots including the sequence he references in the "miswired" comment where the 4-1 setting is moving focus to a higher contrast area. The only way the camera can get 10 fps of athletes running straight at you is with III at 1 or 2 which means he's told the camera that continuous focus isn't the priority in tracking. In fact his fps is so consistently at 10 fps that he must have the setting at 2 which basically means you've told the camera that the focus of the first and subsequent frames isn't the priority. This is also consistent with many of the other sequences in which the first shot of the sequence is not in focus. Add to the mix his apparent mental fixation on the MKII and it's very likely that he's also got III-2 at increased sensitivity which was a standard setting for the MKII in sports.
All this means that the camera will move to a higher contrast focus area very quickly, not that it will track focus more quickly but that it will lose focus more quickly as it did in many of his sequences. So he's basically made focus priority a complete non-priority for the camera and then bashed it for it's emphasis on fps over focus. With III-3 at 0 I never get more than 8 fps in sports settings and often get 6-7 fps.
Finally, all of the subjects are in very high contrast situations. That is what track and field subjects are on bright sunny days, high contrast with bright colors and logos. If you want to optimize your focus in af servo sequences you don't use the settings he has. You slow down III-2, have III-3 at 0, and if you really want to control who's in focus have III-4 at 0 and just release the focus and refocus if you lose the main athelete with III-8 at 0. Telling this camera that focus isn't your priority and then letting it select the focus point with III-4 at 1 and expansion points selected on a bright sunny day shooting athletic events isn't optimizing your settings. It's a recipe for oof images in a sequence of fast moving, high contrast subjects.
Not trying to be argumentative, but to clarify my previous post against your response I should have been a bit clearer. Canon has told me on two separate occasions that the fix is not about Servo - they were very emphatic about that the second time I spoke with them because as they noted that was spreading all over the discussion boards. I was told it was simply about calibrating the outer focus points and the tech gave me a series of tests to do to see if my camera needed to be sent in and all the tests involved shooting from Single Mode.
I seemed to read into what RG was writing that he seemed to have been told the same thing by Canon and that's where I was coming from. Perhaps I made the wrong assumption.
Cheers,
JD
mark fadely wrote:
Jon,
The latest fix was all about fixing the tracking and focus accuracy of all the points. Canon didn't want to openly admit that they were fixing the overall performance of the ai-servo autofocus system. Probably just to save face from their prior denials of actual problems. Anyway, fact is, they figured out a fix to turn the camera into a great performer that can even outdo the mighty MKIIn which I also own.
If I didn't own the camera and had read RG's article I would assume the very same thing, Jon. But there are a lot of hidden issues at play here and when people without first hand knowledge make assumptions and conclusions based on heresay and inaccurate information it gets very confusing for everyone following the thread.
I am still hoping that some people can benefit from this thread. Your continued questioning makes perfect sense. If Canon is saying to you that the fix only involved the outer points then that's what I would believe too. I really don't what to make of all the various reports from Canon, RG, and others but I'm just very glad I've got superb camera now and I wanted to share the info with everyone here. So it makes you wonder who to believe.
But here are the facts:
1. I bought my MKIII in Feb '08 and was never satisfied with the performance of the ai-servo AF. It was only about 70% as good as my MKIIn. I returned it to Canon twice to have them fix the problem and it was never any better so I just lived with a poorly functioning sports camera. I shoot action with center point only usually at f2.8 and always in ai-servo. I am constantly using the camera in it's weakest area o performance.
2. I sent in my MKIII for the latest "outer point" AF fix not knowing what to expect.
3. I received my MKIII back from canon with a completely different functioning center point focus. In ai-servo it is now better than my MKIIn. I have no idea how the outer points work cause I've never used them.
So what would you think if you had been through this scenaro. I am just one of the many users that had the same experience. Just check the N&W forum to see all the glowing reports from BIF shooters using 500, 600, and 800mm lenses.
Here's a couple of examples of shots my MKIII could never get before the fix -
busy backgrounds with small moving subjects were impossible before the fix:
Not trying to be argumentative, but to clarify my previous post against your response I should have been a bit clearer. Canon has told me on two separate occasions that the fix is not about Servo - they were very emphatic about that the second time I spoke with them because as they noted that was spreading all over the discussion boards. I was told it was simply about calibrating the outer focus points and the tech gave me a series of tests to do to see if my camera needed to be sent in and all the tests involved shooting from Single Mode.
I seemed to read into what RG was writing that he seemed to have been told the same thing by Canon and that's where I was coming from. Perhaps I made the wrong assumption.
Cheers,
JD
mark fadely wrote:
Jon,
The latest fix was all about fixing the tracking and focus accuracy of all the points. Canon didn't want to openly admit that they were fixing the overall performance of the ai-servo autofocus system. Probably just to save face from their prior denials of actual problems. Anyway, fact is, they figured out a fix to turn the camera into a great performer that can even outdo the mighty MKIIn which I also own.
If I didn't own the camera and had read RG's article I would assume the very same thing, Jon. But there are a lot of hidden issues at play here and when people without first hand knowledge make assumptions and conclusions based on heresay and inaccurate information it gets very confusing for everyone following the thread.
I'm not sure if this has been mentioned or not in the ten or so pages of this thread - but regardless of who Rob Galbraith is or how much influence he supposedly has in the photographic community - statistically his experience is ONE DATA POINT. Right?
One data point that is not necessarily any more valid than anyone else's I would submit. His data point is just more public and expressed via a highly trafficked website.
bauerman wrote:
I'm not sure if this has been mentioned or not in the ten or so pages of this thread - but regardless of who Rob Galbraith is or how much influence he supposedly has in the photographic community - statistically his experience is ONE DATA POINT. Right?
One data point that is not necessarily any more valid than anyone else's I would submit. His data point is just more public and expressed via a highly trafficked website.
Actually I don't think it's even one valid data point. Apparently the "thousands" of photographs were taken with different cameras, different photographers and with no explanation of the custom function settings. The "samples" were also taken at the widest apertures with long lenses at relatively close distances judging by the size of the subjects in the frames. With such a wide degree of variability I don't see how there's even one data point in terms of his conclusion - which is simply that he thinks the MKII performed better.
Thanks for the pamphlet. Very clear explanations and scenarios for a newcomer to the 1DIII like me. Nice to have the PDF to keep on the iPhone along with the User Manual for quick reference.
Well to push this to 11 pages: I had the Mk III (now have the 1Ds MkIII) and do have the D3. The performance and focus differ somewhat from Canon to Nikon. However kudos to Canon that continue to fix their AF (at no cost to me).
I cannot imagine that Nikon can't, after all this time, improve their AF performance. But all this is from someone who never had these problems. Also I assume that I will have to take many shots for one hero. Remember AF is, statistically, never in focus-just less out of focus in some settings. So this is a question of tolerances and conditions.
Maybe the problem is there's no really standardized test settings? Obviously RG has a use pattern, and consequent testing protocol, that the Mk III has trouble passing. But YMMV and so, until there are apples to compare with apples, forget the conspiracy theories and goodnight.
I will add my thoughts to this thread, a bit late, but work has been busy so I haven't had too much FM time!
I have owned 5 1D3s, 2 before the first blue dot fix and 3 after. I can sum up my experience as the 1D3s were much faster to acquire focus than my venerable 1D2Ns, especially in lower light, but all 5 of them never tracked as well as my 1D2Ns for burst shooting in any light.
A month and a half before this latest 1D3 fix was announced I happened to try the Nikon D3. Simply put I was stunned how well the D3 and x focussed. Yes, initial AF acquisition is a tad slower than the 1D3 but once it has lock, burst shooting is extremely accurate. Like Lee Woolery, I sold my own small fortune of Canon kit and jumped with both feet and I haven't looked back.
Since moving to the D3 and x, I have borrowed 2 1D3 and 300/2.8 IS' and without going into detail on my findings, the performance of these new fixed bodies for AF tracking certainly didn't make me think cr4p, I have made the wrong decision in switching.
While I have read RG's comments on the 1D3 AF since his first article and also read comparisons between the 1D3 and the D3, all this stuff is just information, nothing more, it's just information. I would never be so bold as to stand up and proclaim that the D3 is better than the 1D3 overall, but what I will say, is that for what I shoot (sport, motorsport and aviation by the way) and the way I shoot, the Nikon D3 is a big improvement over all of the 1D3s I have owned, and to me that equates to better. That may not be the case for gear reviewers and other people who have shot with both, but it is the case for me.
Plus, it is great to be back with full frame again.
Alistair Watson wrote:
I will add my thoughts to this thread, a bit late, but work has been busy so I haven't had too much FM time!
I have owned 5 1D3s, 2 before the first blue dot fix and 3 after. I can sum up my experience as the 1D3s were much faster to acquire focus than my venerable 1D2Ns, especially in lower light, but all 5 of them never tracked as well as my 1D2Ns for burst shooting in any light.
A month and a half before this latest 1D3 fix was announced I happened to try the Nikon D3. Simply put I was stunned how well the D3 and x focussed. Yes, initial AF acquisition is a tad slower than the 1D3 but once it has lock, burst shooting is extremely accurate. Like Lee Woolery, I sold my own small fortune of Canon kit and jumped with both feet and I haven't looked back.
Since moving to the D3 and x, I have borrowed 2 1D3 and 300/2.8 IS' and without going into detail on my findings, the performance of these new fixed bodies for AF tracking certainly didn't make me think cr4p, I have made the wrong decision in switching.
While I have read RG's comments on the 1D3 AF since his first article and also read comparisons between the 1D3 and the D3, all this stuff is just information, nothing more, it's just information. I would never be so bold as to stand up and proclaim that the D3 is better than the 1D3 overall, but what I will say, is that for what I shoot (sport, motorsport and aviation by the way) and the way I shoot, the Nikon D3 is a big improvement over all of the 1D3s I have owned, and to me that equates to better. That may not be the case for gear reviewers and other people who have shot with both, but it is the case for me.
Plus, it is great to be back with full frame again.
Thanks for bringing your perspective to the thread. I would be interested to hear more details about how the "fixed" MKIIIs compare with the D3 that you are now comfortable with? And how the "fixed" MKIIIs compare with your old MKIIIs. In other words, was there a significant difference in focus accuaracy with the "fixed" cameras?
