2 examples of what I feel are unacceptable 100% crops.

Now it very well may be me that is creating the unacceptable images, but hopefully someone can tell me why they appear so blurry. Most all of my soccer shots were like this. All were handheld, some with lower shutter speeds, some with higher. Straight from the camera. Canon 100-400 lens on a 40D

HTTP response: NOT FOUND | MIME changed to 'image/gif'

HTTP response: NOT FOUND | MIME changed to 'image/gif'

i'd assume that you are not focusing on the face, that's why the face is soft.

i'm not sure if this is correct, but i think with what I've learnt so far, in physics and camera, I've put together this little theory.

1st - most of use uses center focus, thus the focus is not on the face.

2nd - because of that, the face, more often that not, is not on the same plane as the center focus plane.

i'm assuming that, imagine and x,y,z coordinate system, x being lens to subject, y being // to x axis, and z as the ground to sky. when the camera AF locks onto a point, the point is relatively locked to the xy plane, resulting in on that plane be in clear and sharp focus. the rest of the subject not in that plane will be OOF,

given that we're shooting wide open most of the time, that plane, is really really thin, in close relation with the DOF and the aperture setting, the thickness of the plane is increased, so to speak.

not sure if this is the right theory, but at least, that's how i explain it.

one area of the racket is razor sharp, that's where i was focused on, the face, being behind the racket's plane is a little OOF

HTTP response: NOT FOUND | MIME changed to 'image/png'

I do indeed focus on center target vs focusing on the face. However I would estimate that I was 80-100 feet or so away which should have given me a 3-5 foot range for sharp DOF. IS was on in both this instances shooting in servo multishot mode. Set to the longer mode to avoid searching for focus and shorter range.

carpenter wrote in post #5569600
I do indeed focus on center target vs focusing on the face. However I would estimate that I was 80-100 feet or so away which should have given me a 3-5 foot range for sharp DOF. IS was on in both this instances shooting in servo multishot mode. Set to the longer mode to avoid searching for focus and shorter range.

---------------
Below are the actual numbers using your info - You think you have 3-5 feet when in actually you really only have less that 2 feet and that's in a perfect environment - now take info account things like real world camera shake - IS not fully engaged.

Subject distance 80 ft
Depth of field
Near limit 78.3 ft
Far limit 81.7 ft
Total 3.38 ft
In front of subject 1.65 ft (49%)
Behind subject 1.72 ft (51%)

Then there's the fact you were shooting at wide open...or close to it.

I wouldn't be happy either. Your shutter should have been fast enough to prevent camera shake.

Here are two examples of what my lenses can do; as a comparison.

First the lowly Canon 100-300 I got for $190 on ebay. The lens is a GREAT budget sports lens.

Straight from camera; just resized. Both shot wide open at f/5.6

HTTP response: NOT FOUND | MIME changed to 'image/gif' | Redirected to error image by FLICKR

HTTP response: NOT FOUND | MIME changed to 'image/gif' | Redirected to error image by FLICKR

HTTP response: NOT FOUND | MIME changed to 'image/gif' | Redirected to error image by FLICKR

HTTP response: NOT FOUND | MIME changed to 'image/gif' | Redirected to error image by FLICKR

Palladium wrote in post #5569927
---------------
Below are the actual numbers using your info - You think you have 3-5 feet when in actually you really only have less that 2 feet and that's in a perfect environment - now take info account things like real world camera shake - IS not fully engaged.

Subject distance 80 ft
Depth of field
Near limit 78.3 ft
Far limit 81.7 ft
Total 3.38 ft
In front of subject 1.65 ft (49%)
Behind subject 1.72 ft (51%)

I thought about that as well, but none of the pictures came out sharp. In the second example the subject was walking.. with those setting I would have had 2.24 feet in front of him. Shouldn't that be more than enough to get a sharp picture? I am pretty steady with my hands even with the larger lens on. Is it possible that the servo is a bit slow to focus? Possible that the lighting just wasn't right for these as they do seem a tad underexposed??
Perhaps I am expecting too much from the 100% crop as the subjects don't really fill up the frame much so facial features at 100% would be far less sharp??

Make sure you set your focus point to be in a position to be on the face. I use the top middle and rotate it as I go from portrait to landscape.

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

Your original subject is small and dark. Both would make the shot seem less sharp.

I'd try some more shots before I gave up.

George

carpenter wrote in post #5569600
I do indeed focus on center target vs focusing on the face. However I would estimate that I was 80-100 feet or so away which should have given me a 3-5 foot range for sharp DOF. IS was on in both this instances shooting in servo multishot mode. Set to the longer mode to avoid searching for focus and shorter range.

You are using DOF calculations and 100% crops - this is an error.

The underlying and very subjective detail in DOF is the circle of confusion. This is basically how OOF a shot can get before it is noticeable in a given print size viewed at a given distance.

If you are staring at 100% crops on a 30 inch monitor, the standard CoC values will not be adequate because you will be scrutinizing the image at a much greater enlargement and closer distance than the DOF calculator expects.

Don't forget, there is only one plane of infinitismal thickness that is at maximum sharpness. The DOF calculator is just giving you a ballpark as to how far you can get from that plane before the softness becomes detectable.

I've actually set up my own calculator to give me the size of the blur disk at a range of distances from the focus point after I punch in aperture, focal length and focus distance. This is handy because then I can judge the DOF based on my own requirement for CoC at a given shot. Also I can see how fast I get to large disks (very blurred backgrounds) reltive to the background distance.

JeffreyG wrote in post #5570425
You are using DOF calculations and 100% crops - this is an error.

The underlying and very subjective detail in DOF is the circle of confusion. This is basically how OOF a shot can get before it is noticeable in a given print size viewed at a given distance.

If you are staring at 100% crops on a 30 inch monitor, the standard CoC values will not be adequate because you will be scrutinizing the image at a much greater enlargement and closer distance than the DOF calculator expects.

Don't forget, there is only one plane of infinitismal thickness that is at maximum sharpness. The DOF calculator is just giving you a ballpark as to how far you can get from that plane before the softness becomes detectable.

I've actually set up my own calculator to give me the size of the blur disk at a range of distances from the focus point after I punch in aperture, focal length and focus distance. This is handy because then I can judge the DOF based on my own requirement for CoC at a given shot. Also I can see how fast I get to large disks (very blurred backgrounds) reltive to the background distance.

What he said! :p

Honestly, I dont see anything wrong with 100% crops coming from a lens being shot wide open. I think too many people expect too much out of their cameras now-a-days. Apply some USM and call it a day.

These two images were shot with a MkIII and 400 f/2.8L IS. Both are working perfectly and the combination makes a total north of $10,000. It's pretty much the pinnacle of sports photography today. They both may look super sharp but the 100% crops dont look very much different than what youve posted here. I can post up 100%s later when I get home if you really want me to. I think that what youve posted are typical results from using a 40D and a zoom. They also may appear to look a little softer due to lack of overall contrast and being a little on the dark side (at least in the 100% area)

HTTP response: 404 | MIME changed to 'image/gif' | Byte size: ZERO | PHOTOBUCKET ERROR IMAGE

AdamLewis wrote in post #5570931
Honestly, I dont see anything wrong with 100% crops coming from a lens being shot wide open. I think too many people expect too much out of their cameras now-a-days. Apply some USM and call it a day.

Hmmm....I actually agree with the OP that the crops are just a touch softer than what I would call OK......BUT.....only if the crops are at the exact point of focus.

I think the problem is that the OP was looking at anything within a normal DOF calculation using 0.030 or 0.019 CoC numbers (depending on format) and thinking the sharpness would be perfect.

I can say that if I focussed on the 100% crop faces shown above with my 300/4, even with the 1.4X TC on and wide open the result would be sharper.

OP need to reshoot at the same type subjects with focus on face and see what it looks like.

So, Adam I take it your 1D3 wasn't plagued by that AF "phenomenon" as Canon calls it, huh? Those look tack sharp!