Camera Labs

This is a quick practical test, triggered by discussion of diffraction limiting in the Canon 7D thread.

Setup:
Canon 50D on aperture priority mode
ISO400 fixed
Lens is Sigma 150mm f/2.8 macro with 1.4x TC attached. This is my typical macro configuration. Therefore it is effectively a 210mm f/4 lens.
Camera is mounted on a tripod with wireless shutter release.
Manual focus on the centre line of the text in the image. According to the lens window, this is at 45cm, for an optical magnification of approximately 0.8x with the teleconverter in place.

Starting wide open, I took shots at one stop intervals throughout the range offered by the lens.

Following are the resized shots, as well as 100% lossless crops from the camera produced jpeg.


f/4


f/5.6


f/8


f/11


f/16


f/22


f/32

It is no surprise at maximum aperture, the DoF is quite thin, and in this test case only the centre line is mostly in focus. The lines either side disappear into an unreadable blur rapidly. As the aperture is reduced, more and more text becomes sharper. At the smallest aperture, even the extreme lines are quite clear even if they are softer.

But that is only the bigger picture, where the resize down hides the smaller differences. What happens to the fine detail? Following are the crops of the "th" from the centre of the image.


f/4


f/5.6


f/8


f/11


f/16


f/22


f/32

At f/4, the DoF is so thin it is hard to tell the sharpness. As the DoF increases up to f/11, to my eyes the sharpness doesn't really change much. There is a slightly higher degree of softening at f/16, and getting noticeably worse at f/22 and f/32.

I couldn't quantify the degree of softness at the smallest apertures, but to me the softening effect is much smaller than that of the DoF gains. My conclusion is that diffraction softening at small apertures should not be a barrier in itself to using them for increased DoF.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Thanks for researching it

_________________
I take pictures so quickly, my highschool was "Continuous High".

f/5.6 looks sharpest to me, but I do agree that the difference in sharpness isn't noticeable until f/16 when some softening is visible. Even so, it would be practically impossible to notice unless you were viewing at 100% and had a side by side comparison with another shot.

_________________
PhilipGoh.com

The texture is much more defined at f/11 / f/8 than it is at 16.

_________________
Canon EOS 5DmkII + BG-E6 + Canon EOS 40D + BG-E2N + Canon EOS 33
Canon 17-40 F/4L USM + Canon 24-70 F/2.8L USM + Canon 28mm F/1.8 USM + Canon 70-200 F/2.8L USM IS + Canon 85mm F/1.8 USM
Canon Speedlite 580EX II + Canon Speedlite 540EZ + 2 x Nikon SB-80DX
Cactus V2s Wireless Trigger - 5 x Cactus V2s Wireless Reciever

MY FLICKR!

Hi popo,

Great work.

I think it's fair to say that from theory f/8 on the 50D is only just into the diffraction softening area whereas at f/16 the size of the Airy disk is about two pixels wide. Comparing the f/5.6 and f/8 images shows no perceptible difference in sharpness but the depth of field is certainly shallower at f/5.6 as expected:

..............f/5.6.....................................................................................................................f/8

Comparing f/8 and f/16 we have:

..............f/8........................................................................................................................f/16

I'd have to agree that while diffraction is starting to soften the central part of the image that effect is small and the increase in depth of field at more than compensates at the extremities.

Comparing f/8 and f/32 we have:

..............f/8........................................................................................................................f/32

To my eye the f/32 image is much softer in the centre and as soft at the extremities.

My only question relates to the in-camera defaults used for the JPEGs. Like all modern DSLRs the 50D has a default "Sharpness" step when producing its JPEGs similar to applying sharpening during post processing. As I mentioned in this post in the 7D thread it's my personal belief that "image sharpening can help a little if the image is just starting to get soft but with a potential trade-off in introduced artefacts". I wonder if that's what we are seeing in the f/16 shot as compared to f/8. It's possible that image sharpening is recovering most of the sharpness in the centre and because the depth of field softening at the extremities has dropped at f/16 image sharpening is more effective at the extremities in that shot than it was when at f/8.

Of course you can blow that possibility right out of the water if you can tell us that in-camera sharpening was turned off. But maybe that doesn't matter as however they were obtained your shots show that operating a full stop dimmer than the diffraction limit (f/16 in this case) gives more depth of field benefits than any diffraction induced softening while two stops dimmer than the diffraction limit results in an unacceptably soft image if one has to crop near the 100% level. As expected if the final image is viewed smaller than at a 100% crop diffraction softening is no longer visible as the diffraction limiting f-number increases in direct proportion to the reduction of the image.

I'll add a pointer to this thread from the 7D thread so that potential purchasers can benefit from your research. Many thanks for putting the time in to ground the theory with real world results.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

I just checked the camera setting... it's on standard which has a sharpening level of 3 on a scale up to 7.

Even if that may have some effect on the pure theory side, being practical there will likely be some sharpening applied. This still serves as a practical example.

Actually, I just had an idea to gauge the detail loss.

f/8

f/8 half resolution

f/8 quarter resolution

f/32

The middle two have had information thrown away by resizing down to half and quarter linear size respectively, and resizing back to original. Note the f/8 at half resolution looks pretty good still, only slightly softer than the original sample. I suspect if I applied a little sharpening, it could look practically as good again. Remember due to the bayer pattern of the sensor, the actual resolution may be half that of the output image (for red and blue), hence resizing to half doesn't lose that much information.

At a quarter linear resolution we're really throwing away information now. Putting aside the blocking due to the resizing up, we have lost fine detail there, and very loosely speaking, it is comparable to the f/32 image. Note the camera output seems a bit duller at the smaller two apertures, which may make it feel softer than it is. A touch of extra contrast would take that out.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Very interesting read. Thanks for posting it.

For me, f/16 is clearly beginning to loose clearity in these examples.

_________________
Thomas (beware: Nikon-fanboy and moderator!) My Lens Reviews, My Pictures, My Photography Blog
D800+assorted lenses

+1

And thanks for the experiment Popo your observation is quite useful

_________________
Soon.....

tnx popo&others for your research and discussion,im just a newbee and this is an eye opener to me and very informative....and guys pls tel me what is the meaning of one stop or four stop etc?..regarding aparture..

_________________
D90/Nikkor 18-105mm VR/Nikkor 50mm f1.8

On my 18-105 lens f/4 is the sharpest.

_________________
"If I want a 'snapshot'... i will use my phone.."

My Photos:
http://mistycreek.deviantart.com/

Thanks for this popo. Looking at this test, diffraction doesn't seem all that threatening. Wonder how the T2i/7D performs in a test like this...

_________________
Flickr

a_ron, see here.

eVolutioN, in theory the 7D might show diffraction at a pixel level slightly sooner, but as a very crude estimate it would be equivalent to around 1/3 stop so hardly any difference in the bigger picture.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

popo, again tnxx for the info&reference...really helpssss

cheers

_________________
D90/Nikkor 18-105mm VR/Nikkor 50mm f1.8

Thanks, so I'm guessing barely noticeable difference compared to these tests right?

_________________
Flickr