fredmiranda.com
Login

Moderated by: Fred Miranda
  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Canon Forum | Join Upload & Sell

1              3              35       36       end
  

Archive 2010 · RG on 1D Mk IV autofocus performance

  
 
John--G
Offline
• • • •
Upload & Sell: Off
p.2 #1 · RG on 1D Mk IV autofocus performance


scowl wrote:
You can test this custom function by pointing the camera at something too close for the lens to focus on. The green light will blink and the camera will continue to shoot OOF shots, but more like two or three per second instead of ten.


It's even easier to test. Take a MkIII or MkIV, set it to AI Servo, focus priority, point it at the ground and press the shutter. It will blaze away taking OOF shots. Canon 1-series has always did this which is why I never understood the whole "focus priority" thing.






Feb 11, 2010 at 01:43 PM
brainiac
Offline
[X]
p.2 #2 · RG on 1D Mk IV autofocus performance


Wow. This is an absolute train wreck for Canon.


Feb 11, 2010 at 01:46 PM
Nick Nishizaka
Offline
• • •
Upload & Sell: Off
p.2 #3 · RG on 1D Mk IV autofocus performance


Ouch...


Feb 11, 2010 at 01:59 PM
jeremy_clay
Offline
• • • • • •
Upload & Sell: Off
p.2 #4 · RG on 1D Mk IV autofocus performance


*sigh* here we go! If anyone has one that no longer works because of this review, please pass it over and I'll take it off your hands for a fair price.


Feb 11, 2010 at 02:00 PM
Nick Nishizaka
Offline
• • •
Upload & Sell: Off
p.2 #5 · RG on 1D Mk IV autofocus performance


I should add, I am having my MKIV exchanged for a new one after my LCD went dead after shooting in 27-32 degree weather...cold, windy, and some snow.

Should have easily handled that. My MKIII was doused with water a few times and went through a few fierce snow storms and it kept on going. A few hours in my house with the battery out revived the LCD.

I know another MKIV user that also had the same problem. Sent it into Canon and they said they don't see what the problem is and shipped it back to him.

I'm not taking that chance. I've only had this 2 weeks. I am getting it exchanged...

I hope it doesn't snow too much at the Olympics for the sake of all those photogs with MKIV's!

Edited on Feb 11, 2010 at 02:04 PM · View previous versions



Feb 11, 2010 at 02:02 PM
SoundHound
Offline
• • • • •
Upload & Sell: On
p.2 #6 · RG on 1D Mk IV autofocus performance


Oh BOY!

Well, for sure, those folks shooting the sunny Olympics will be the real field laboratory. Obviously this is the best Canon can do. Probably not good enough since they already have a firmware focus upgrade right out of the box and Rob can't see any difference.

Rob's evaluation is just so neutral and dispassionate. No agenda just get the data (a really, REALLY lot of work) and analyze it while putting his reputation on the line-again! Nice to know that my D3s is just as I found it-better AF than I have a right to expect and a generally fine performer under all but the strangest of situations. Canon probably has hit men searching for Rob.

Evidential, Mk IV, single point AF is really NG (while fine on my old Mk III). That's mostly how I (have to) use it. I just sent Rob $10. He saved me the turnaround cost of buying and selling a Mk IV + lot of, familiar, hassle!

Hassle: because, even if I was/am satisfied with the performance (and my Mk III was just fine out of the box), I am compelled to get the upgrades and paperwork (remember those colored dots and so many pages?) to I can protect my resale. It was during one such "upgrade" that they really screwed up my Mk III and it went back (again) for a silly, obvious problem.





Feb 11, 2010 at 02:02 PM
scowl
Offline

Upload & Sell: Off
p.2 #7 · RG on 1D Mk IV autofocus performance


John--G wrote:
It's even easier to test. Take a MkIII or MkIV, set it to AI Servo, focus priority, point it at the ground and press the shutter. It will blaze away taking OOF shots. Canon 1-series has always did this which is why I never understood the whole "focus priority" thing.


I just did this with my Mark III. Mine shoots only about three frames per second of OOF shots instead of the machine gun ten frames per second when shots are in focus. When you move the camera between things it can focus on and things it can't focus on, the drop in frame rate is very apparent.



Feb 11, 2010 at 02:07 PM
Cementjungle
Offline
• •
Upload & Sell: Off
p.2 #8 · RG on 1D Mk IV autofocus performance


mark fadely wrote:
Wow, he said his BB shots sucked and he used a 70-200 @ f6.3 - something is wrong with that for sure.


Perhaps he should drink less coffee.



Feb 11, 2010 at 02:13 PM
timgangloff
Offline
• • • •
Upload & Sell: Off
p.2 #9 · RG on 1D Mk IV autofocus performance


Hmmn, this will no doubt get interesting. When I first got my mk4 a few weeks ago, I was a bit worried that the hype might just be that, hype. Obviously, I thought it was a risk worth taking and was an early adopter. I bought my mk3 when some were still having troubles (actually, bought 2 and still have one), but was I guess fortunate in that mine seemed to work wonderfully. Still does. But, I really wanted the ability to shoot some candid sports video and still do what the 1d series can do that my 5d2 can't, so I got the 1d4.

So far, my results have not been anything like what RG reports. I shoot a lot of basketball (high school and middle school levels) and mostly strobed and have lots of galleries up that one could look at if so inclined and bored enough. When strobbing a game, I'll typically shoot about 100-120 frames. With my mk4 and the 70-200 2.8, and I'll end up tossing a few to focus issues or what not and probably end up posting about 70. Of this 100-120, some are tossed because I got the refs butt and not the play I was watching a nano second before pressing the button. Most deletions are these kind of issues.

Admittedly, of the ones that I post, the 70 might not all be worthy of praise from an action and or framing perspective, but they are in focus and thus are posted in the hopes someone might buy one. Occasionally, they do. The point is that only a small percentage of my shots get tossed due to focus errors for which I can blame the camera. Most of the time, the focus issue is my fault. Most, but not all.
Sometimes, with basketball, the action will present itself suddenly and I will be forced to pretty much point and shoot. Sometimes I (the camera) nails it and sometimes not. These types of shots are pretty difficult and the players are probably moving fairly fast and I am not spending too much time focusing on the player before hitting the shutter button, so I may in fact be focusing on the fans or another player. So, some of these are those that get tossed due to focus are clearly a weakness of the photographer being unable to track the action properly or anticipate correctly. Sometimes I am slow.

But, this mk4 is not perfect either. Sometimes, I think I am tracking the player well and when I examine the photo, I'll see that the camera missed it. Sometimes I don't know why. Could another player crossed my path so quickly I did not even see them and that threw the focus off a smidgen? Maybe. From my experience, I've not found a camera that nails the shot every time and clearly, this camera will miss some that I think it should have nailed. However, I think it does do a better job than any I've used so far (1d, 1d2, 1d2n or 1d3).

I do think it does a great job of tracking a person and oftentimes, I'll get an image of a person directly in front of my target, obscuring most of my target. The person in front of the image is out of focus and the largely obscured target is still in focus. The camera has essentially followed the subject even when the subject is (mostly) hidden.

However, as good as I think the mk4 is, you have to realize that it is not perfect. You will miss shots. I am unaware of any camera that is perfect. And despite all of the rumblings about Nikon being better now, my friend who shoots along side me with his Nikon D3 averages about the same number of keepers per game as I do.
I am continually amazed that with many of my shots when I am following the action with my center AF point and then snap the shot, I'll get an in-focus image even though the red lit AF is not on my subject as I intended, but the camera continued to follow my subject.

The custom AF settings can also negatively change your results too. If you have a uniform that is tough to focus on (like all black with small numbers or something) and you allow the surrounding AF points to assist, you may well end up focusing on something that you did not intend. You need to experiment and learn what works best for you, your style of shooting and the environment you are shooting in.




Edited on Feb 11, 2010 at 02:24 PM · View previous versions



Feb 11, 2010 at 02:14 PM
dolina
Offline
• • • •
Upload & Sell: Off
p.2 #10 · RG on 1D Mk IV autofocus performance


I think RG and I share the same problem. We do not know how to use the body well enough to customize it to AF properly in a particular situations.

Canon allowing a "strike 2" would damage the Canon brand among pros and amateurs too much that no one would take them seriously anymore.



Feb 11, 2010 at 02:17 PM
Beni
Offline
• • • • • •
Upload & Sell: Off
p.2 #11 · RG on 1D Mk IV autofocus performance


I think that the winter Olympics will tell the tale as to whether RG is on his own with this. I have no doubt that the pro's will not put up for another mkIII type fiasco and the response this time will be far far faster. Let's wait and see...


Feb 11, 2010 at 02:20 PM
FretNoMore
Offline
• • • •
Upload & Sell: Off
p.2 #12 · RG on 1D Mk IV autofocus performance


Well, "interesting" is the word...

I can only say my number of in focus shots has increased a lot with the Mark IV over the Mark III, very few of the misses can be blamed on the camera but rather on me missing the target. But it's early yet, I've only shot a few indoor games and maybe my Mark IV will get worse once I learn how to use it.



Feb 11, 2010 at 02:32 PM
pjbuehner
Offline
• • •
Upload & Sell: On
p.2 #13 · RG on 1D Mk IV autofocus performance


: pjbuehner wrote:
I wish someone could come up with an objective battery of tests to assess AF (AI servo). That is the one thing always lacking in reviews and for these cameras also so crucial.


scowl wrote
These are an objective battery of tests -- give the cameras to experienced photographers in the real world and see how many keepers they walk away with. As long as there are humans behind the cameras, you have to keep the human element in camera tests otherwise the tests are meaningless for anything but remote cameras.


I am sorry but that isn't objective. I also disagree that an objective test would be meaningless. If you put a camera's AF point on a target and you can objectively prove to me that it missed focus 7 of 10 times then I
...Show more

Edited on Feb 11, 2010 at 02:41 PM · View previous versions


Feb 11, 2010 at 02:38 PM
fernsidekid
Offline

Upload & Sell: Off
p.2 #14 · RG on 1D Mk IV autofocus performance


Can I ask about method of shooting
My preferred style is back button focus then shorts bursts of 3-4 shots

I rarely or never have shot a burst of 10 shots as decribed for the runners and skaters in this test where they appeared to have been tracked for some time

I would normally aim for a action skating shot or burst
Then another burst for the arms raised celebration shot
(runners more than skating for me)

Do people shoot more extended bursts than I?

RG appears to shoot in a much more continuous fashion and is the AF set up for this? or even capable of this?



Feb 11, 2010 at 02:39 PM
Mike Mohrmann
Offline
• • •
Upload & Sell: Off
p.2 #15 · RG on 1D Mk IV autofocus performance


I apologize ahead of time as I didn't follow all that closely Galbraith's episode with the 1D3. But does he shoot these tests alone, or does he have others try to disprove or replicate his findings?


Feb 11, 2010 at 02:46 PM
ragebot
Offline
• • •
Upload & Sell: Off
p.2 #16 · RG on 1D Mk IV autofocus performance


From what RG wrote at the end of page 2:

"Be sure your viewing of the photos includes some time spent at a magnification level of 100% in your preferred photo browser. If you look at the pictures exclusively at 50% magnification or less then you'll miss seeing the true softness of some of the out of focus files. "

I confess to being as much of a pixel peeper as the next guy, but when I see terms like "100%" and "true softness" my eyes start to glaze over.

Can someone explain to me what the difference between "true softness" and "softness" and "false softness" is.

Just for the record my 1d4 seems fine. I took a head shot of a gator and am currently in the process of editing the image in CS. One thing I am trying to do is clone out a branch in front of the gator's snout and I noticed a small termite on the branch with lots of detail; but was not able to see "true softness".



Feb 11, 2010 at 02:48 PM
scowl
Offline

Upload & Sell: Off
p.2 #17 · RG on 1D Mk IV autofocus performance


pjbuehner wrote:
I am sorry but that isn't objective.


The only goal of autofocusing is to get the human photographer shots that are in focus.

I also disagree that an objective test would be meaningless. If you put a camera's AF point on a target and you can objectively prove to me that it missed focus 7 of 10 times then I find that very meaningful. I don't mean to discount the value of the human interface but if a test could first eliminate the human variable, then we could see what does or does not work in a given AF system.

I think that's what Canon is doing and I think that's why their AF is failing in the hands of photographers. I'm sure their cameras score tremendously well in Canon's controlled objective scientific autofocusing tests. Rob's experiences with Canon showed their engineers were baffled when they saw their cameras failing out in the real world.

Just as an example, there were a lot of "experienced photographers" who landed on both sides of the fence with the mark III issue and there are also many pros out there who have praised the mark IV...how does one make sense of that?

Photographers take photos of different things. They shoot differently. They move the camera differently. They have different reaction times. They have different expectations. They have different lenses. They shoot in different conditions. The number of variables becomes overwhelming. A good autofocusing system must compensate for all these things.



Feb 11, 2010 at 02:51 PM
jhom
Offline
• • • •
Upload & Sell: Off
p.2 #18 · RG on 1D Mk IV autofocus performance


I'll have to reread the RG article again, but did I miss RG reporting that the lenses he used were individually adjusted to each camera body?

Jim



Feb 11, 2010 at 02:51 PM
David Manning
Offline
• • •
Upload & Sell: Off
p.2 #19 · RG on 1D Mk IV autofocus performance


"true softness" is that softness that only becomes apparent at 100% views. In other words, it's insignificant in real life photography.

ragebot wrote:
From what RG wrote at the end of page 2:

"Be sure your viewing of the photos includes some time spent at a magnification level of 100% in your preferred photo browser. If you look at the pictures exclusively at 50% magnification or less then you'll miss seeing the true softness of some of the out of focus files. "

I confess to being as much of a pixel peeper as the next guy, but when I see terms like "100%" and "true softness" my eyes start to glaze over.

Can someone explain to me what the difference between "true softness" and "softness"
...Show more



Feb 11, 2010 at 02:56 PM
scowl
Offline

Upload & Sell: Off
p.2 #20 · RG on 1D Mk IV autofocus performance


Mike Mohrmann wrote:
I apologize ahead of time as I didn't follow all that closely Galbraith's episode with the 1D3. But does he shoot these tests alone, or does he have others try to disprove or replicate his findings?


During the Mark III fiasco Rob was in continuous contact with dozens of professional photographers who shared their experiences and recommendations with him. He also did testing with Canon engineers using their certified "good" cameras and lenses. He certainly does not do his testing in a void.



Feb 11, 2010 at 02:57 PM
1              3              35       36       end




FM Forums | Canon Forum | Join Upload & Sell

1              3              35       36       end
    
 

Welcome back
Log in to your account