Definitive 10D AF test

TangoMan

Veteran Member
Messages
1,889
Solutions
3
Reaction score
639
Location
DJ
Last week I shot about 2000 shots on assignment.

Rodeo, moving or static subjects, poor lighting (800 iso, 1/90, 2.5 or 1/60, 2.8), lenses: 50mm 1.8, 100mm/2, 100mm/2.8 macro, central sensor, both "one shot" and "servo" according to situation.

About 200 shots were in full sun, with bright colorfull subjects and the results were marginaly better. (100ISO, 1/750-1/1500, 2.8)

Results: Half way through, I had to trow out 70% for misfocus. The camera loved to focus way behing the subject. It occasionaly focused in front of it, and sometimes dead on it.

I ended up having to prefocus whenever possible: shoot on AF, blow up the picture 10x to see if it's in focus, repeat, repeat, and eventually accept it as "as good as it can be in the limited time in have in my hands".

So, a manual focus camera with an optical viewer unable to tell you if the focus is right, used nearly all the time on manual exposure mode: 1500 US$. What a bargain. Fortunately there was an histogram and a digital sensor producing great quality for the 800 speed.

That's a real world test and it says "not good enough." Sure, many shots were discarded for motion blur, but when the crowd is sharp and the static subject of your shot is blurred, it's called AF failure. I want a better camera. I have better things to do that go through piles of pictures discarding AF failure casualties...

--
I loved curves, long before Photoshop appeared :)
 
That behavior never bothered me too much before, because the pictures I sell were shot in studio at f8-11 and the other pictures... well I didn't really care that much. I had the 10D since it was first available and know it pretty well. Had I not decided to make that kind of photo, where focus is both critical and difficult, I would have remained satisfied with my camera.

So... You may feel safe by saying that MY camera behave that way, and it's true. But should you some day ask more of YOUR camera, you might very well call it definitive for you too.

Enjoy your good enough 10D :)

TM

--
I loved curves, long before Photoshop appeared :)
 
I would have tought that the lack of samples and precise measurements would have kept them away, but, well... I really don't mind. What's a forum for after all?

--
I loved curves, long before Photoshop appeared :)
 
it's been up for 30 minutes or so in prime time....

I always dread seeing this kind of post because so much energy and effort is usually put into responses by the pro and con groups here.. maybe it IS getting old and everyone is just tired of it... that would be a good thing.

I'm just waiting to see what comes out next. I can live with my 10D and D30 even though both have some minor faults.. they are still the best two cameras I have ever owned... Well, the F1 really was a great film camera.

Jim Radcliffe
http://www.image36.com

The ability to 'see' the shot is more important than the gear you use to take the shot.
 
I just wanted to give an honest evaluation of how the gear worked under some conditions to help others decide if the camera will be good enough for them.

There might have been 80 responses by now if I had worded things the way others do, just to stir up emotional responses. But it's not my purpose.

It's been a while since I came to the forum and I really appreciate the slower pace and the lens forum.

Good job Phil!
 
ISO 800 or 1600, servo, 100mm f/2.8 and 200mm f/2.8. About 20% misfocused shots, including, just like you, some pictures where the crowd is sharp and the static subject is not. And this is from a 10D body that went 3 times to Canon for AF ajustment...

Although I like these:






Last week I shot about 2000 shots on assignment.
Rodeo, moving or static subjects, poor lighting (800 iso, 1/90, 2.5
or 1/60, 2.8), lenses: 50mm 1.8, 100mm/2, 100mm/2.8 macro, central
sensor, both "one shot" and "servo" according to situation.

About 200 shots were in full sun, with bright colorfull subjects
and the results were marginaly better. (100ISO, 1/750-1/1500, 2.8)

Results: Half way through, I had to trow out 70% for misfocus. The
camera loved to focus way behing the subject. It occasionaly
focused in front of it, and sometimes dead on it.

I ended up having to prefocus whenever possible: shoot on AF, blow
up the picture 10x to see if it's in focus, repeat, repeat, and
eventually accept it as "as good as it can be in the limited time
in have in my hands".

So, a manual focus camera with an optical viewer unable to tell you
if the focus is right, used nearly all the time on manual exposure
mode: 1500 US$. What a bargain. Fortunately there was an
histogram and a digital sensor producing great quality for the 800
speed.

That's a real world test and it says "not good enough." Sure, many
shots were discarded for motion blur, but when the crowd is sharp
and the static subject of your shot is blurred, it's called AF
failure. I want a better camera. I have better things to do that
go through piles of pictures discarding AF failure casualties...

--
I loved curves, long before Photoshop appeared :)
--
Mskad.

 
Last week I shot about 2000 shots on assignment.
Rodeo, moving or static subjects, poor lighting (800 iso, 1/90, 2.5
or 1/60, 2.8), lenses: 50mm 1.8, 100mm/2, 100mm/2.8 macro, central
sensor, both "one shot" and "servo" according to situation.

About 200 shots were in full sun, with bright colorfull subjects
and the results were marginaly better. (100ISO, 1/750-1/1500, 2.8)

Results: Half way through, I had to trow out 70% for misfocus. The
camera loved to focus way behing the subject. It occasionaly
focused in front of it, and sometimes dead on it.

I ended up having to prefocus whenever possible: shoot on AF, blow
up the picture 10x to see if it's in focus, repeat, repeat, and
eventually accept it as "as good as it can be in the limited time
in have in my hands".

So, a manual focus camera with an optical viewer unable to tell you
if the focus is right, used nearly all the time on manual exposure
mode: 1500 US$. What a bargain. Fortunately there was an
histogram and a digital sensor producing great quality for the 800
speed.

That's a real world test and it says "not good enough." Sure, many
shots were discarded for motion blur, but when the crowd is sharp
and the static subject of your shot is blurred, it's called AF
failure. I want a better camera. I have better things to do that
go through piles of pictures discarding AF failure casualties...

--
I loved curves, long before Photoshop appeared :)
Shooting moving subjects at 1/90th or 1/60th?? No wonder they weren't sharp-you may want to go back and study some more.
 
None.

After reading a thread from Isaac explaining how Canon has two kind of AF systems, one with low precision for Elans, D-30-60, 10D and rebels, and one with high precision for the Eos 3, 1N, 1V, 1D and 1Ds, I am hoping that I will have better luck with a body of the higher AF grade.

One thing I noticed though, when the Elan 7 was offered as a "better" camera than the Elan 2, was that it's AF speed appeared to move the lens faster, yet it hunted a bit before locking, probably in focus. I purchased the Elan 2 which usually stopped dead on focus at the first try, and I wondered how can Canon pretend that the camera is newer and better when it's offering an Af performance that is worse... Hmmm "Not satisfied? Buy a EOS 3" might be the reason behind that strategy... Or is the AF really worse? I felt it was.

Anyway, the 10D AF isn't good enough for me now and I hope the 1D replacement will be.

DavidP, is the 1D up to the task of shooting a rodeo? What percentage of OOF do you get roughly?

(my own percentage migh have been 50%. Frustration makes it appear higher, although the bad lighting is also part of the problem, and as mishkin would have it, the light was severly green in some parts of the arena, despite a careful custom balance in the center of it)
Which other cameras have you used to shoot this sort of assignment?
How did they perform?
--
I loved curves, long before Photoshop appeared :)
 
I can tell motion blur from OOF. If you can't, stop studying and go shoot
 
The 1D seems to be the favorite of the action shooters. Perhaps they will get a successor this fall.
 
The 1D is quite capable of performing well in rodeo situations. The photographer, however, is another story. ;)

I'm not sure how many I got that were truly out of focus. Probably not that many. A lot suffered either from motion blur or just not getting the right composition. I was in pretty poor light, though. ISO 1600, f/2.8, and I think 1/400.

Actually, I just checked, and it was 1/320.
DavidP, is the 1D up to the task of shooting a rodeo? What
percentage of OOF do you get roughly?
--
The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers
 
I assume you can easily tell when the background is what was focused. LOL. Easily distinguished from motion blur.

On your shots with good AF, how did the images look at only 1/60 or 1/90?

I would've bumped up the ISO to 1600. Maybe even 3200 in that kind of lighting.
I can tell motion blur from OOF. If you can't, stop studying and go
shoot
--
The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and tax it out of business?' -- Will Rogers
 
I love to hear that your 1D is what I expect it to be.

To reduce the motion blur rate, maybe the photographer can try yoga or meditation (or medication?) before the next rodeo :) Or if it's synchronised rodeo, sit on the next bull. Motion will cancel out. :p

As for the composition, it's a no brainer for me. When in AF, the central sensor get a grip on what it can and I crop later for composition. When in MF, things are more relax. Prefocus, precompose, just pan if needed...

If you have any tips, keep them coming :)

TangoMan
I'm not sure how many I got that were truly out of focus. Probably
not that many. A lot suffered either from motion blur or just not
getting the right composition. I was in pretty poor light,
though. ISO 1600, f/2.8, and I think 1/400.

Actually, I just checked, and it was 1/320.
DavidP, is the 1D up to the task of shooting a rodeo? What
percentage of OOF do you get roughly?
--
The Lowest Paid Concert Photographer Around
http://www.neonlightsimaging.com/artshow/final.htm
Photography -- just another word for compromise

'Since we can't keep crime in check, why don't we legalize it and
tax it out of business?' -- Will Rogers
--
I loved curves, long before Photoshop appeared :)
 

Keyboard shortcuts

Back
Top