Are these pics perfectly focused (dof)?

[jooka]

Got few weeks ago 10d and 24-70/2.8 L set. I'm bit concerned about autofocusing and dof (at 2.8) cause i think i'm getting sometimes bit out of focus shots, like focus is too back of the subject rather than where it should be. So, i'm posting few pics for you guys to check out and prove me wrong.

On both photos focus was on auto and was locked on the book or the cd case. These are 100% crops. No post processing. Jpeg files packed with setting 7 on Photoshop.

Do you see these as perfectly focused shots?

http://www.pbase.com/image/29927447

1)
Shooting Mode
Aperture-Priority AE
Tv( Shutter Speed )
1/90
Av( Aperture Value )
2.8
Metering Mode
Center-weighted averaging
Exposure Compensation
0
ISO Speed
100
Lens
24.0 - 70.0 mm
Focal Length
50.0 mm
Image Size
3072x2048
Image Quality
Fine
Flash
Off
White Balance
Auto
AF Mode
One-Shot AF
Parameters
Contrast Normal
Sharpness +1
Color saturation Normal
Color tone Normal
Color Space
sRGB


http://www.pbase.com/image/29927499

2)
Shooting Mode
Aperture-Priority AE
Tv( Shutter Speed )
1/180
Av( Aperture Value )
2.8
Metering Mode
Center-weighted averaging
Exposure Compensation
0
ISO Speed
100
Lens
24.0 - 70.0 mm
Focal Length
50.0 mm
Image Size
3072x2048
Image Quality
Fine
Flash
Off
White Balance
Auto
AF Mode
One-Shot AF
Parameters
Contrast Normal
Sharpness +1
Color saturation Normal
Color tone Normal
Color Space
sRGB

[Bruce Hamilton]

Do you see these as perfectly focused shots?

The little X in the upper left corner is razor sharp. :lol:

[Cadwell]

The little X in the upper left corner is razor sharp. :lol:

ROFLMAO :lol: :lol:

[jooka]

For some reason pics didn't appeared on the thread itself so there are now linked.

[Cadwell]

To answer your question though, they look fine to me. :) The first one is perhaps a little soft.

[jooka]

Thanks for the input!

Is this something i should be worried about?

http://www.pbase.com/image/29928583

Same lens at 2.8. Autofocused. If i do it manually it locks right on middle...

[Cadwell]

Thanks for the input!

Is this something i should be worried about?

http://www.pbase.com/image/29928530

Same lens at 2.8. Autofocused. If i do it manually it locks right on middle...

You mean that the DOF is smaller in front of the focus point than behind? That's normal on a 10D. Says so in the manual on page 162.

[jooka]

Thanks for the input!

Is this something i should be worried about?

http://www.pbase.com/image/29928530

Same lens at 2.8. Autofocused. If i do it manually it locks right on middle...

You mean that the DOF is smaller in front of the focus point than behind? That's normal on a 10D. Says so in the manual on page 162.

So if i do it manually and get almost tack sharp shot in the middle, this is considered as normal behaviour when using autofocusing (DOF is smaller in front of the focus point than behind)?

And I will have to re-read the page 162.

Thanks again for clarifying this subject for me.

[Pekka]

Thanks for the input!

Is this something i should be worried about?

http://www.pbase.com/image/29928583

Same lens at 2.8. Autofocused. If i do it manually it locks right on middle...

Do you see this problem in normal shooting?

That image seems to be withing 10D autofocus system limits. Perhaps slight backfocus - but the test you use is not proper one. The AF rectangle you see in viewfinder is smaller than the actual area which is used by AF. See http://photography-on-the.net/forum/showthread.php?t=10526 for good test procedure - just be aware of the 1/DoF accuracy rule.

10D AF accuracy is 1/DoF, which means the when you focus to a spot, the actual focus may be anywhere withing depth-of-field of the open aperture of your lens (in this case 2.8). This does not mean it always is off center, it may be spot on in many cases - amount of light matters here, and contrast of focus point.

If you suspect backfocus, the Canon Finland repair is handled by http://www.jastekniikka.fi/ they can test it very accurately, they have the new test system installed. Both lens and camera can be checked and adjusted.

[jooka]

Normal shooting pics (when compared to manually focused) are bit soft.

Take a look at this photo. Took couple of shots with af and manual from same subject (left side manual, right side af)

http://www.pbase.com/image/29929377

Do you think i should send my 10d to repair?

[jooka]

10D AF accuracy is 1/DoF, which means the when you focus to a spot, the actual focus may be anywhere withing depth-of-field of the open aperture of your lens (in this case 2.8). This does not mean it always is off center, it may be spot on in many cases - amount of light matters here, and contrast of focus point.

Is this written in the manual or how do you know it (i.e where did you got this information)? Is AF accuracy for example better in 1DmkII or is it same?

I mailed to JAS yesterday. They said that if you use smaller aperture, focusing problem disappears because of larger dof... however they also said that they can repair whole package (lens and the body).

What kind of repair (how massive?) this really is? I mean, do they have to explode whole camera into bits and pieces and same with lens or is it just software kind of thing?

[kb244]

The focus may seem fine, but i think you are more worried bout how sharp the lens can actually produce on a focused subject. Or something to that extent, I only breifed thru this thread by the way.

[Jemmind]

The little X in the upper left corner is razor sharp. :lol:

Bruce you crack me up!

Julie

[Pekka]

10D AF accuracy is 1/DoF, which means the when you focus to a spot, the actual focus may be anywhere withing depth-of-field of the open aperture of your lens (in this case 2.8). This does not mean it always is off center, it may be spot on in many cases - amount of light matters here, and contrast of focus point.

Is this written in the manual or how do you know it (i.e where did you got this information)? Is AF accuracy for example better in 1DmkII or is it same?

This info is from Canon's technical director Chuck Westfall and it is accurate. All 10D focusing points are NP (Normal Precision, 1/DoF). 1D Mark II has 7 HP (High Precision, 1/3 DoF) focusing points and the rest are NP.

I mailed to JAS yesterday. They said that if you use smaller aperture, focusing problem disappears because of larger dof... however they also said that they can repair whole package (lens and the body).

What kind of repair (how massive?) this really is? I mean, do they have to explode whole camera into bits and pieces and same with lens or is it just software kind of thing?

AFAIK the lens is adjusted by hardware (I think one small component is soldered) and camera is adjusted with computer (there are lots of settings which are available only to repair). Of course these are things that they can answer more accurately :)

[jooka]

This info is from Canon's technical director Chuck Westfall and it is accurate. All 10D focusing points are NP (Normal Precision, 1/DoF). 1D Mark II has 7 HP (High Precision, 1/3 DoF) focusing points and the rest are NP.


I'm not familiar with this kind of techical information. Could you tell me more what does this "1/DoF" really stands for, what does it mean like when compared to 1DmkII's "1/3 DoF" for example?

[Pekka]

This info is from Canon's technical director Chuck Westfall and it is accurate. All 10D focusing points are NP (Normal Precision, 1/DoF). 1D Mark II has 7 HP (High Precision, 1/3 DoF) focusing points and the rest are NP.


I'm not familiar with this kind of techical information. Could you tell me more what does this "1/DoF" really stands for, what does it mean like when compared to 1DmkII's "1/3 DoF" for example?

It's said most concise in

http://bobatkins.photo.net/info/faq30/eos3af.html

"What do we mean by "normal-precision" vs. "high-precision" AF? A "normal-precision" AF sensor produces focusing accuracy within the depth of focus for the lens' maximum aperture. A high-precision" AF sensor produces focusing accuracy within 1/3 the depth of focus for the lens' maximum aperture. Therefore, when activated, the high-precision AF sensors have up to 3 times the precision of the normal-precision sensors. "

What this means in practice is that

- in all cases a faster lenses focus more accurately because focus is done in open aperture, regardless of aperture you chose to shoot with.

- when you have 2.8 lens and focus few meters away the subject may be off-focus but still in depth-of-field (i.e. not totally blurry).

- in 1D series you should use any of the the seven hp AF points if you need best focusing accuracy.

Also, Chuck once explained that np definition does not negate the possibilty that with np AF the focus CAN BE perfect. The definitions are "worst case" definition, i.e how off it may be. in worst case

Note also that the focus rectangles you see in camera viewfinder are smaller than the actual focus areas. In my previous link there is a link to photo which shows the real AF areas of 10D.

PS. If you are in Helsinki we can meet and compare our 10D's - mine is quite well calibrated and should be by the specs.

[jooka]

Sadly we can't meet since i live in Oulu but thanks for the offer, Pekka!

I'm still thinking whether i should send camera to repair or not.

This is the photo which got me thinking something is wrong. It's 100% crop and focus was set on "the face" of the this wooden "creature". Take a look...

http://www.pbase.com/image/29998064

[Pekka]

Try the test outlined in http://photography-on-the.net/forum/showthread.php?t=10526