mark fadely wrote:
Yep, like David says, they are electronic. It makes sense that as the camera cycles though its software loop the MF check could throw off the system. It is a bug in the software but I'm convinced it's there. It should be able to be fixed by a firmware update since it must be software related. It is a strange quirk, but it makes a lot of sense to me that this is what's happening.
If that is the case, I hope they update the firmware soon to take care of it. I use the 85LII, and the last thing I need is to keep track of changing a CF when I change lenses. Although...I don't have any of the superteles with electronic MF, so I suppose I could register a set of CF for sports/wildlife and a different set for other stuff. But that's no excuse - fix the glitch, Canon!

apdieb wrote:
I just used these settings as mentioned above... Didn't do any real tests since I don't have any moving subjects.. However, I did point it at various objects and moved quickly backwards and forwards... The AF seemed to be much more rock solid..Perception anyway.. I haven't looked at any images... but the viewfinder subject didn't appear to hunt at all..

Wow.. I really need to do some tests this weekend. Never thought to turn off tracking functionality..




So have you had a chance to test it with the CF changes and action? I am going to play with this and see if I notice any change. Right now if I could just "dial" it in I would be stoked.

If it doesn't pour down rain Sat...I'll give it a good test. So far, I have only shot a couple of weddings with these settings. They seem good...but no real serious movement...

I'll be covering the Texas Rangers opening day ceremonies on Tue...Hope to get a good test there as well.

****UPDATE*****

Well, after my Err99 situation I had during a baseball tournament a few weeks back, I received my MKIII back from Canon, saying they fixed the AF charge and box (whatever that means) and other adjustments (not very specific). So.. I got to test it all this week with yet another baseball tournament I was covering, and I can honestly say that so far.. its fixed and working just as described, but this is only after around 6k clicks this week, but I was putting it through all the tests I could. I was getting around a 96-98% keeper ratio which is better than I got with my 1dmkIIn and wayyyyyy better than I was getting when I used this MkIII before sending it in to canon. Another interesting point is that before I sent it in to canon I was using a +9 ratio for all my lenses to correct a horrid backfocus problem, now everything is set to 0 and it works fine.. so far so good, but it sucks that I still have to worry about it in the back of my mind when I am shooting!

Charles

Without editorial comment.... The interesting thing is how many different things Canon is doing to "fix" these cameras. We have realigning and re-calibrating focus points, correcting lens to sensor distance, mirror stoppers, mirror boxes, AF charge, etc. etc.

Hrow wrote:
Without editorial comment.... The interesting thing is how many different things Canon is doing to "fix" these cameras. We have realigning and re-calibrating focus points, correcting lens to sensor distance, mirror stoppers, mirror boxes, AF charge, etc. etc.

Yes, agree, and this "headless chicken" approach does not instill confidence in their "fix" procedure by a long shot.

Colin

Yakim Peled wrote:
I had good success with non moving subjects in One shot mode.

If everytime you send the Mark III in they do something different to fix the problem, then sooner or later they might accidentally fix it. 4 times for my Mark III, and still not replaced or working properly. My Mark II N gets it right just about every time.

Charlemagne wrote:
****UPDATE*****

Well, after my Err99 situation I had during a baseball tournament a few weeks back, I received my MKIII back from Canon, saying they fixed the AF charge and box (whatever that means) and other adjustments (not very specific). So.. I got to test it all this week with yet another baseball tournament I was covering, and I can honestly say that so far.. its fixed and working just as described, but this is only after around 6k clicks this week, but I was putting it through all the tests I could. I was getting around a 96-98% keeper ratio which is better than I got with my 1dmkIIn and wayyyyyy better than I was getting when I used this MkIII before sending it in to canon. Another interesting point is that before I sent it in to canon I was using a +9 ratio for all my lenses to correct a horrid backfocus problem, now everything is set to 0 and it works fine.. so far so good, but it sucks that I still have to worry about it in the back of my mind when I am shooting!

Charles

Good news Charles...

Question: You said +9 due to backfocus... Correct me if I am wrong, doesn't + (positive) adjustments correct FRONT focus issues?

BTW: I did some tests today out in bright sun 2pm here in TX at my son's soccer game. Overall, I was happy with my results...Still not a keeper rate I'd like, but not terrible. All shots benefited from a tad of USM...and still exhibited a propensity to front focus a tad...I went to making on the fly M/A again and before I knew it the game was over. I REALLY need to do a more controlled test (and not of my son...I get to excited and forget what I am doing).

Canon 10D wrote:
Yakim Peled wrote:
I had good success with non moving subjects in One shot mode.

Well, I finally got my first 2-game outing with the 'new' MkIII, and I must say it performed quite well. About 94% 330/350 images) in focus disregarding user error, so perhaps even a bit higher hit rate.

Mixed sun and clouds, 60 degrees F, sun to my back most of the time. AI Servo at medium tracking sensitivity, center-point with left-right assist enabled. Looking forward to some cycling shooting today, we'll see how it does with faster subjects.

Regardless, this is more like it.

-Jeff

Here's one for the heck of it, though not at f/2.8:

Jeff, is this a new, new one that replaced the new one that replaced the old one?

Jeff, congrats! I hope you will get the cycling shot at the same rate.

No Henry, this is my original new one, the BD of 12/07, #558xxx. Apparently, when it works, it works...

I've gotten some one-shot oddities out of it, and some AI Servo issues in very specific lighting, so perhaps today the light was just 'right'?

Ron, I shot about 1500+ pictures today of the Oredigger Classic, and although I haven't gone through them, I noticed very few problems, except for when the clouds came in and the light got flat. I'll post some when I've had a chance to go through them.

Jeff wrote:
No Henry, this is my original new one, the BD of 12/07, #558xxx. Apparently, when it works, it works...


Arrrrrrrrr........

Why do we always put a reservation when talking about this camera? This is so frustrating.

Happy shooting,
Yakim.

hey jeff, the wee ones benefit from a lower perspective. get your backside down on the pitch, man

nice timing!

jbfaulconer wrote:
4 times for my Mark III, and still not replaced or working properly. My Mark II N gets it right just about every time.

And you're not the only one who has said that. Canon really seems to be struggling with this camera. Which is a shame for that price tag.

Here's the real kicker for me. I didn't care about file transfers to FTP sites, live view on the LCD, or a number of other built-in functions. The highly regarded AF performance of the 1D series is about the only reason I decided to finally shell out $4500 in cash for a camera body. Especially the Servo focus.

And guess what..... the AF performance is the one thing they screwed up on this new camera.

There has to be more people who are just as disappointed about that as I am.

Jeff wrote:
No Henry, this is my original new one, the BD of 12/07, #558xxx. Apparently, when it works, it works...



Yakim Peled wrote:
Arrrrrrrrr........

Why do we always put a reservation when talking about this camera? This is so frustrating.


Huh?

apdieb wrote:


Question: You said +9 due to backfocus... Correct me if I am wrong, doesn't + (positive) adjustments correct FRONT focus issues?

BTW: I did some tests today out in bright sun 2pm here in TX at my son's soccer game. Overall, I was happy with my results...Still not a keeper rate I'd like, but not terrible. All shots benefited from a tad of USM...and still exhibited a propensity to front focus a tad...I went to making on the fly M/A again and before I knew it the game was over. I REALLY need to do a more controlled test (and not of my son...I get to excited and forget what I am doing).





You are right about it supposedly fixing front focus, but my lenses were all backfocusing like champs with the MKIII, I tried to use the negative settings, but that made it worse.. so I used the positive settings and that is how I got to +9... I know.. Weird.. but maybe that is why my MkIII just didn't work.. it was THAT screwed up! LMAO!

Charles

Hey Jeff.. Looking good... maybe they actually have the real "Fix" for this camera now.. and about farging time! I only put it through one tournament, but it was out in the 90 degree florida sun, so I would say it probably is a go.. we will see when I do some track this friday! wooohoo!

Charles

Charlemagne wrote:
apdieb wrote:


Question: You said +9 due to backfocus... Correct me if I am wrong, doesn't + (positive) adjustments correct FRONT focus issues?

BTW: I did some tests today out in bright sun 2pm here in TX at my son's soccer game. Overall, I was happy with my results...Still not a keeper rate I'd like, but not terrible. All shots benefited from a tad of USM...and still exhibited a propensity to front focus a tad...I went to making on the fly M/A again and before I knew it the game was over. I REALLY need to do a more controlled test (and not of my son...I get to excited and forget what I am doing).





You are right about it supposedly fixing front focus, but my lenses were all backfocusing like champs with the MKIII, I tried to use the negative settings, but that made it worse.. so I used the positive settings and that is how I got to +9... I know.. Weird.. but maybe that is why my MkIII just didn't work.. it was THAT screwed up! LMAO!

Charles

Gotcha...

I completely understand. I am all over the place with my M/A testing as well. The thing that bugs me the most is that controlled tests/adjustments with charts, LCD diagrams, etc...don't equal those of real world use. I am becoming more and more convinced that the AF sensors are locking on objects with more texture/contrast in the frame regardless of the point chosen. Seems like for me no matter what adjustments I make, the tendency (not always) is to lock on the grass in front of the subject (if it's in the frame). This is with center point manually selected and AI_Servo. I have no idea why grass is my major contributor...but all I know is that when it's predominately in the frame, my camera wants to grab it out in front...regardless of my AF point being squarely and consistently on the subject.

All these adjustments seem to be helping...but I am convinced there is something inherently wrong with the way the AF is interpreting the scene and we're having to make adjustments to fool the camera into actually misfocusing based on where it thinks it should be vs. where we want it.

The more I read and write on these threads, the more I realize that a person that hasn't experienced the issues might think we're crazy. I know much of it sounds like a run-on babbling rant. hehe

I did not want to start a new post about this, but I am having some weirdness with my "Blue Dot" MKIII.

It seems to have major AF issues in AI Servo mode using low MM lenses, like the 24-70 F2.8 L or the 24-105 F4 L. If I have it in "Center point" and have the "surrounding assist points" on, I get ALOT of problems with it. If I turn those off, it seems better but still inaccurate. With higher mm lenses like 70-200 or even a 100-400mm I seem to have far better luck.

Has anyone else experienced this problem? I am just wondering if it something stupid I am doing. My 40D and 5D do not have this problem (the 40d does not have the assist points anyway).

Thanks.

Yes, many of us have experienced just what you are going through, as the 55 pages of this thread will illustrate. Lots of suggestions buried within these pages...

Charlemagne wrote:
Hey Jeff.. Looking good... maybe they actually have the real "Fix" for this camera now.. and about farging time! I only put it through one tournament, but it was out in the 90 degree florida sun, so I would say it probably is a go.. we will see when I do some track this friday! wooohoo!

Charles


Not so fast, Charles. I'm only about halfway through the 1800 cycling images I shot yesterday, and although I got a lot of good ones (witness the shotgun approach), there's a lot of crap in there, too. I'll get a statistical number posted when I have one, but let's say it's nowhere near 90%...

I hope yours works with all your shooting as well as mine does for soccer!

-Jeff

apdieb wrote:
I am becoming more and more convinced that the AF sensors are locking on objects with more texture/contrast in the frame regardless of the point chosen.

Now that is an impression that my 40D gave me. It focused on edges that were nowhere near the chosen AF sensor. However, it's hard to imagine how it can happen. BTW there was no cure for my 40D - Canon reckoned it was within spec.

- Alan

Alan321 wrote:
apdieb wrote:
I am becoming more and more convinced that the AF sensors are locking on objects with more texture/contrast in the frame regardless of the point chosen.

Now that is an impression that my 40D gave me. It focused on edges that were nowhere near the chosen AF sensor. However, it's hard to imagine how it can happen. BTW there was no cure for my 40D - Canon reckoned it was within spec.

- Alan




Sorry to hear that about your 40D. It is ridiculous that OOF is now officially within their specs. As I am sure you know that spec represents a range but perhaps you weren't aware that that range goes from "not even close" to "way the f**k off".