HELP! 10D v 5D close-up at 100%

[warrenb]

Seriously, I need someone to tell me if I'm crazy. My new 5D came out of the

box today and I was less than impressed with the images. Soft focus everywhere.

I was beginning to think I was a worse photographer with better equipment. Then

i did my little experiment. A few weeks ago I did a closeup with my 10D and

100/2.8 USM macro and took a photo of a coin. Today I repeated the conditions

(as close as possible) with the same lens/exposure with the 5D. Here are the

results. Please tell me what to do. I'm balled up in a corner right now sucking

my thumb in the fetal position with a serious case of buyers (or suckers)

remorse.

[kiro]

were these both shot in raw? jpg? what were the camera settings?

[warrenb]

both raw, both processed with same, saved parmeters. same exposure.

[beauh44]

Hi Warren, No need to ball up! If you got it from a place like B&H you can exchange it or have Canon service take care of it under warranty. Just curious but can you manually focus and get a good shot?

[warrenb]

didn't try a manual focus. got it at B&H.

[KenPapai]

f/32 ehh?

 

Still your results are weird -- why not shoot the lens at f/11 and check the 100% crop outcomes?<div></div>

[gdanmitchell]

The 5D normally produces very sharp images when combined with excellent lenses and

appropriate camera settings.

 

At f/32 on a DSLR you'll normally get significant defraction effects and far less than optimal

sharpness. Try a more normal f/8 or possibly f/11.

[KenPapai]

One more 100% crop from a 10D at 54mm, f/8, ISO 200<div></div>

[bobatkins]

The 10D actually has a higher pixel density than the 5D. The 10D has 6MP, but it's 1.6x. If you just scaled up the sensor until it was full frame like the 5D, you'd end up with 15.36MP.

 

The 10D has, therefore, higher native resolution than the 5D, but the image is cropped of course.

 

If you want the Canon canera with the highest native resolution sensor, it's currently the Digital rebel XTi. It has the same pixels size (native resolution) that a fullframe 25.6MP camera would have.

 

See http://www.bobatkins.com/photography/digital/canon_eos_5d_or_20d.html

 

BTW at f32 diffraction should be killing sharpness, especially with the 10D's smaller sensor (see http://www.bobatkins.com/photography/technical/diffraction.html)

[beauh44]

Something told me to hang on to that 10D! ;-)

[warrenb]

OK, maybe my test was flawed, but does this seem like a good enough image quality from a 5D? 24-105/4L f/16 ISO100 Sharpness increased on camera, unsharp mask, tripod, mirror up, etc.

[warrenb]

Here it is. sorry.

[warrenb]

Why won't this go inline?

[PuppyDigs]

Warren, normally I use MF for macro. AF tends to be unrelieble. Nevertheless, I've viewed

hundreds of 5D images at 100% and they blew away my 10D in terms of detail and

sharpness. Looks like your 5D needs AF calibration.

 

Also consider that photos are not normally viewed at 100%. My 5D landscape images printed

at 12 x 18 always looked sharper, more detailed and had fewer artifacts than my 10D. Why?

The 10D images need more uprezzing (sorta enlargement) than the 5D files, so softness and

defects are more evident.

[lester_wareham]

I am surprised there is any sharpness if you shot at f32, try at f11. Bob has maid the point about pixel density.

[mrstubbs]

Forget that suckers remorse stuff. Try going back and making the frame again with your 5d. In camera...reduce contrast two notches..increase sharpness to six or seven...reduce saturation two. Then compare it with the best you can muster from the 10d. My bet? ....the 10d would go rusty pretty quick.<div></div>

[mrstubbs]

again

[valo_soul]

To Bob (or anyone else who can explain the pixel density issue):

 

I understand that this would apply to cropping a 5D image to get the same FOV of a 10D and image would yield a lower quality final image. However, does this mean that completely filling the frame with both cameras on the same subject, the 5D image would be less sharp than the 10D, just larger? Like if you simply used the same lens on both cameras, but walked closer to the subject with the 5D to compensate for the crop factor, would the 10D image actually contain more detail in a 100% crop? Well, any crop for that matter.

[lester_wareham]

Correct this pixel density point only applies if the subject is at the same magnification. If the use reduces the magnification on the crop camera the full frame should have the advanatge.

 

I think it is assumed Warren has just cranked the magnification all the way up to 1:1. Looking at the shots that is clealy not the case, he seems to have almost but not quite adjusted for the relative magnification.

 

Anyway, shooting at f32 is unlikely to give optimum results on either camera.

[warrenb]

OK, how bout this one at 100%? 24-105/4L f/16, 1/8, mirror-up, tripod, etc. Does this look like 5D sharp?

[mike_earussi1]

Why do you insist on using such high apertures? For optimum sharpness you should be shooting at f5.6-8.

[warrenb]

Optimum, fine, but even at f/16 don't you think this should be a little sharper?

[mike_earussi1]

Out of curiousity why not shoot one at f8 to see if there is much difference?

[sitemistic]

Whatever the aperture, if I had just spent what a 5D cost and the photos looked like that, I'd

send it back. There is clearly something wrong.

[warrenb]

Thank you Jim. That is what I am doing. I just wanted to make sure that I wasn't completely crazy or stupid (although, some reading these threads may think I am both).