could f4 really be that shallow?

[stanpatyrak]

I am new to this site, and hope I have figured out the right place to post this. I recently upgraded glass to the 24-70f2.8L. I have been playing around with the new toy, trying to get used to the lens before a late January wedding job.

Help me out here if you can - fast glass is something new for me. I took this pic at f4. Both men are in focus, the woman in the middle appears soft. Could f4 really be too shallow? Maybe I was too close to the subject? (I was about 4 feet away).

I had this issue a few times - in fact, there were occasions where when shooting a couple, one person was soft and the other was sharp. (No doubt due to being a few inches in front of or behind of the sharp subject).

Shot with a 5d.

Thanks for the feedback.

[goforphoto]

It is hard to tell anything from such a small image. What focus point were you using ? what were your other camera settings? f4 doesn't really give us much to go on.

[stanpatyrak]

Thanks for the reply. Not sure if I can post an image much larger due to image size rules.

Used central focus point (targeted man on right) and then recomposed since the man on the left joined the image. shutter at 60.

Since I had this issue a few times, I am assuming that f4 on a 5d might be too shallow. But, not completely sure.

[goforphoto]

usually 24-70mm L takes sharp images. are you sure it wasn't just camera shake?
oh you are allowed an image 800x800px as long as the file size isn't over 100kb try reducing for web at 72 dpi.

[stanpatyrak]

maybe this will help: http://img72.imageshack.us/img72/4716/img1048xi2.jpg

[goforphoto]

The center is a little soft I can see that now. You may be right. Have you tried the same type of shot at smaller apertures? And how did they turn out?

[lil_miss]

I'm having similar issues with my Sigma 24.70 F2.8. Took some photos on teh weekend at F6 and even that had blurred parts.. Only got the lens new a week ago.. hoping its not a dud :( :(

[stanpatyrak]

Well, a week ago I took a family portrait, two rows of people. f5.6. Shot was a winner. So, I guess I am finding that f4 is too shallow, even for two people side by side - given that one may be slightly in front of or behind the other.

[tim]

Seem to me you've answered your own question. I'm suprised it's so noticable at F4, I do couple shots at anything down to F2, but group shots are F8 or narrower. Since the guys are in focus and the girls aren't it has to be DOF. Check out DOFs at www.dofmaster.com

Here's one for comparison for you. The first is the whole shot, the attached is a 100% crop. It was taken at F2.8 at around 2 meters from the couple. Both are acceptably sharp, but the girl is less so. I focused between them.

Whole shot
http://mrwild.co.nz/unprotected/wp/sj/SJ_347.jpg

100% crop attached

[goforphoto]

Still this is the kind of problem you'd see with a 50mm 1.8 not a 24-70mm 2.8 L

[stanpatyrak]

well, here is a variable - the 5.6 that turned out great was taken outside (flash was used, but as fill) and the f4 shot was obviously inside, flash dependent. Both ISOs were at 100.

[tim]

I've read the 24-70 F2.8 L does often need to be calibrated when it's new.

[goforphoto]

Did the lens search for focus or snap right to it? or were you manually focusing?

[stanpatyrak]

It is a snappy lens, and I have no issue auto focusing. I am beginning to wonder if a few factors are involved - I could have been too close and ISO 100. I am looking at the pics that did turn out ok and I have one of 5 people, two layers deep, at iso 200 that looks great.

[tim]

DOF is meant to be half a foot if you were at 50mm, or quarter of a foot if you were at 70mm. If this is an isolated incident i'd not worry about it, just learn from it. You could always try http://www.focustestchart.com/chart.html

[stanpatyrak]

Seems somewhat isolated. Out of the 20 shots of couples or groups, 3 came out like this. I guess it could have been a number of things, including slight movement. Christmas with family will be a great time to work this out a bit more. They don't mind me pointing a camera at them for the day we are all together. Well, they mind - but I am feeding them so they will have to deal with it.

A few days ago I did the test you mentioned and found slight backfocusing. Called canon and they told me that the test could be flawed by a number of factors, including metering mode used. They agreed to take a look at it, but I have a wedding in a few weeks - so it is going to have to wait. I am one of those 5d owners that had dust in the viewfinder straight out of the box, so they will be taking care of that at the same time (they want the camera anyway).

I guess in the long run that would answer any issue I am having, but your sample pic at 2.8 helps. Me not staring at a 100% crop also helps!

[D. Craig Flory]

One thing to keep in mind is this ... what is being focused on ? Here is a site on depth of field.

http://www.ephotozine.com/techniques/viewtechnique.cfm?recid=63

[Hellashot]

If that shot was a 24mm, DOF=2.1 feet

If taken at 70mm, DOF=0.23 feet

DOF is very dependent on subject distance and focal length.

[Titus213]

What Hellashot said. Yes, your DOF could be that shallow. And that DOF is split, .11 ft in front, .12 ft in back.

Check DOF here: http://www.dofmaster.com/dofjs.html

[DocFrankenstein]

You gotta remember that tim is shooting with a smaller sensor than you do.

When he's shooting at f/4 it's the same as if you're shooting at 6.4

IMO on full frame you'd really want f/5.6 and up to get both of them in focus if you're printing an 8*10

The DOF chart is useless because it's very relative. If you're shooting landscapes, then you have one criteria... if you're shooting a couple and both of their eyes have to be in focus when printed at 300DPI with decent size... it's a whole other story.

[D. Craig Flory]

You gotta remember that tim is shooting with a smaller sensor than you do.

When he's shooting at f/4 it's the same as if you're shooting at 6.4

IMO on full frame you'd really want f/5.6 and up to get both of them in focus if you're printing an 8*10

The DOF chart is useless because it's very relative. If you're shooting landscapes, then you have one criteria... if you're shooting a couple and both of their eyes have to be in focus when printed at 300DPI with decent size... it's a whole other story.

Good analysis ! Another example is when doing a large group and you need every person sharp.

[stanpatyrak]

Good point. Thanks so much for taking the time to help me out with this.

[tim]

Good point doc. Even with smaller sensors you have to be very careful of DOF, when i'm not being careful I do get quite a few images with insufficient DOF. The little screen on the back of the camera is very deceptive.