Posted by David Spindle on April 12, 2013 at 6:30 PM
I have been using my new D7100 for about a week. With my D90 (12mpx) camera I could routinely make use of f/22 for macro work. But with the D7100 my images begin losing sharpness a little by f/11 and by f/22 they are just awful. Thinking perhaps my Sigma 150 macro lens might not be compatible, I tried my Nikon 70-200 f/4 at 200mm for a close up and...same exact results.
Don't get me wrong. This is a wonderful camera, and for portraits, landscape and wildlife it's fantastic. It is even terrific for close up and macro.... at f/8 or larger.
But at f/16 or f/22 it's so diffracted as to be unusable.
Has anyone else experienced this? Is a 24 megapixel camera to much resolving power for small apertures?
Thanks in advance for any advice!
Dave
Posted by David Bostedo on April 12, 2013 at 7:03 PM
Dave - Assuming your lens is OK, that is most likely diffraction softening. It's a property of the lens aperture (of all lens apertures, really), and is always there no matter what camera body you use. It's there for all lenses too, and always has been for any type of photography. In fact, you got the exact same softening with your D90 - your sensor just wasn't able to resolve it.
Whether or not a sensor resolves the diffraction softening, and at what aperture it becomes visible, is dependent on the size of the pixels. A higher mega-pixel count on a smaller sensor means very small pixels. So diffraction can begin to show at 100% views at fairly small f-numbers (large apertures). So that sounds bad for small high mega-pixel sensors. However, with higher mega-pixel cameras, that's a much bigger magnification when viewed at 100%, so diffraction effects are not worse, just easier to see because you can blow the image up a lot larger. And since smaller sensors have more DOF, you probably don't need as small an aperture for a given shot.
Posted by David Spindle on April 12, 2013 at 7:38 PM
Thank you, David
That's the clearest explanation I've heard. Let me make certain I understand. If I take two images at f/16, one with the D90 and one with the D7100 with the same lens, then even though at 100% the D90 image appears sharp and the D7100 image appears soft.....if I were to print them as 9X13 prints, then, in theory, they should both look the same?
Posted by Brad Mangas on April 12, 2013 at 9:00 PM
Hmm, my NIKKOR AF-S 70-200mm f/4G ED VR on my D800e is extremely sharp even at f/22. You have to look at an image above 100% to even begin to notice diffraction. The only thing that noticeably changes with aperture is dof.
Posted by David Bostedo on April 12, 2013 at 10:58 PM
...then even though at 100% the D90 image appears sharp and the D7100 image appears soft.....if I were to print them as 9X13 prints, then, in theory, they should both look the same?
Dave - That depends on whether or not 9x13 is large enough to see the difference; But yes, that's correct. Try viewing the D7100 image so that objects in the image are the same size on your screen as the D90 image at 100%. That will give you some idea of the difference. The best way to check though is to make prints, as you mentioned.
...my NIKKOR AF-S 70-200mm f/4G ED VR on my D800e is extremely sharp even at f/22.
Brad - Some lenses may be a bit better than others due to the design of their aperture and optics, but diffraction is unavoidable - it's physics. The D800E is crazy sharp to begin with, so perhaps that's changing your definition of what sharp really means. You will see softening at f/22 compared to, say, f/8 though.
All of that being said, diffraction is only something to be avoided if you have options. If you need/want the DOF or small apertures for shutter speed reasons, that's still usually going to give you your best image, regardless of diffraction. There are folks who get amazing results shooting macro at f/22 all day, for instance.
The interesting thing in the article to me, is that wider apertures, like f/2.8, are significantly worse because of non-diffraction effects, than f/22 is because of diffraction. But it seems most folks are content to shoot that wide open, while avoiding f/22 because of diffraction.
If you want to test yourself, you could always lock down a tripod in really good light; Find something flat and detailed to shoot so DOF isn't a concern; Use live view manual focus, so auto-focus isn't a concern; And shoot a lens at various apertures and compare for yourself.
A scientist friend of my and myself have done some testing of diffraction using my nikkor 200mm micro and my D700 and his D800, by f16 the D800 was suffering badly were as my D700 i can use at f32 with no sign of it, we came to the conclusion that pixel density also comes into the equation.
Ian
Posted by Greg Lessard on April 13, 2013 at 7:19 AM
Great info! Thank you!
I have noticed similar issues between my D90 and D7000 for the past couple of years and assumed that it was the amount of extra pixels on the D7000 that were causing the problem. I must admit that the D7000 is superior to the D90, but the D90 is still my favorite camera. I think Nikon outdid themselves with that camera. I particularly like that the d90 never attracts dust inside the camera, while the d7000 seems to suck it up like a vacuum. I wish Nikon would replicate that D90 cleanliness on its future cameras!
Posted by Nick Baldwin on April 13, 2013 at 8:15 AM
This has been very informative. I just purchased both a D7100 and a new 70-200mm f/2.8 VRII. but have not yet tried any settings of f/22. (they replace an aging D300 and older version 70-200f/2.8 VR)
I do have a question though. If one is using a prime lens does that cut down on the diffraction. A friend told me that with regards my digiscope. So I ceased using the 20-60x eye piece and bought a 30x and was using a prime 20mm f/1.7 on my GX1.
thanks again for the info. with you permission I will copy and paste to Word so that i can read again from time to time.
Happy shooting,
Posted by David Bostedo on April 13, 2013 at 12:59 PM
A scientist friend of my and myself have done some testing of diffraction using my nikkor 200mm micro and my D700 and his D800, by f16 the D800 was suffering badly were as my D700 i can use at f32 with no sign of it, we came to the conclusion that pixel density also comes into the equation.
Ian - Your lens is producing the same amount of diffraction blur no matter which camera you attach it to. That can't change. All that changes is whether or not you have a sensor capable of capturing it. The D700 as 1/3 the resolution of the D800, so that makes some sense - the D700 wasn't capable of resolving the exact same diffraction blur that the D800 picked up at f/16.
Please keep in mind that you said the D800 was "suffering" from it. In this case, so was the D700 - but because it's a less capable sensor, it wasn't able to resolve it. Comparing both cameras at the pixel level as though that's equal footing is wrong, and is extremely unfair to the D800 when talking about things like this.
Posted by David Spindle on April 15, 2013 at 9:52 AM
Thanks again for all the help, everyone. It is much appreciated!
David,
Regarding testing lenses, I hadn't read your reply yet but this weekend I tested all my lenses, no flat subject or manual live view focus though. I may set up again and do that. Never the less I now have a much more comfortable feeling about using each lens with regard to diffraction softening. By the way, the big winner in my lens set was the Nikon 16-35.
Testing also brought out other quirks that were interesting in each lens. The oddest was that two lenses have one particular f/stop that produces a significant increase in contrast. On my 70-200 f/4 VR for instance it's always f/11 regardless of focal length. Strange, but at least I know it's there!
Discussion
9 repliesDon't get me wrong. This is a wonderful camera, and for portraits, landscape and wildlife it's fantastic. It is even terrific for close up and macro.... at f/8 or larger.
But at f/16 or f/22 it's so diffracted as to be unusable.
Has anyone else experienced this? Is a 24 megapixel camera to much resolving power for small apertures?
Thanks in advance for any advice!
Dave
Whether or not a sensor resolves the diffraction softening, and at what aperture it becomes visible, is dependent on the size of the pixels. A higher mega-pixel count on a smaller sensor means very small pixels. So diffraction can begin to show at 100% views at fairly small f-numbers (large apertures). So that sounds bad for small high mega-pixel sensors. However, with higher mega-pixel cameras, that's a much bigger magnification when viewed at 100%, so diffraction effects are not worse, just easier to see because you can blow the image up a lot larger. And since smaller sensors have more DOF, you probably don't need as small an aperture for a given shot.
Here's a good thread on NPN discussing diffraction effects.
And another relating to macro.
That's the clearest explanation I've heard. Let me make certain I understand. If I take two images at f/16, one with the D90 and one with the D7100 with the same lens, then even though at 100% the D90 image appears sharp and the D7100 image appears soft.....if I were to print them as 9X13 prints, then, in theory, they should both look the same?
Dave - That depends on whether or not 9x13 is large enough to see the difference; But yes, that's correct. Try viewing the D7100 image so that objects in the image are the same size on your screen as the D90 image at 100%. That will give you some idea of the difference. The best way to check though is to make prints, as you mentioned.
Brad - Some lenses may be a bit better than others due to the design of their aperture and optics, but diffraction is unavoidable - it's physics. The D800E is crazy sharp to begin with, so perhaps that's changing your definition of what sharp really means. You will see softening at f/22 compared to, say, f/8 though.
All of that being said, diffraction is only something to be avoided if you have options. If you need/want the DOF or small apertures for shutter speed reasons, that's still usually going to give you your best image, regardless of diffraction. There are folks who get amazing results shooting macro at f/22 all day, for instance.
Here is a good article with actual data tests for various Nikkor lenses on D3's, D800s, and D700s. If you read it, you'll notice that resolution (not exactly sharpness, but very closely related) peaks somewhere between f/4 and f/8 regardless of camera or lens.
The interesting thing in the article to me, is that wider apertures, like f/2.8, are significantly worse because of non-diffraction effects, than f/22 is because of diffraction. But it seems most folks are content to shoot that wide open, while avoiding f/22 because of diffraction.
If you want to test yourself, you could always lock down a tripod in really good light; Find something flat and detailed to shoot so DOF isn't a concern; Use live view manual focus, so auto-focus isn't a concern; And shoot a lens at various apertures and compare for yourself.
A scientist friend of my and myself have done some testing of diffraction using my nikkor 200mm micro and my D700 and his D800, by f16 the D800 was suffering badly were as my D700 i can use at f32 with no sign of it, we came to the conclusion that pixel density also comes into the equation.
Ian
I have noticed similar issues between my D90 and D7000 for the past couple of years and assumed that it was the amount of extra pixels on the D7000 that were causing the problem. I must admit that the D7000 is superior to the D90, but the D90 is still my favorite camera. I think Nikon outdid themselves with that camera. I particularly like that the d90 never attracts dust inside the camera, while the d7000 seems to suck it up like a vacuum. I wish Nikon would replicate that D90 cleanliness on its future cameras!
I do have a question though. If one is using a prime lens does that cut down on the diffraction. A friend told me that with regards my digiscope. So I ceased using the 20-60x eye piece and bought a 30x and was using a prime 20mm f/1.7 on my GX1.
thanks again for the info. with you permission I will copy and paste to Word so that i can read again from time to time.
Happy shooting,
Ian - Your lens is producing the same amount of diffraction blur no matter which camera you attach it to. That can't change. All that changes is whether or not you have a sensor capable of capturing it. The D700 as 1/3 the resolution of the D800, so that makes some sense - the D700 wasn't capable of resolving the exact same diffraction blur that the D800 picked up at f/16.
Please keep in mind that you said the D800 was "suffering" from it. In this case, so was the D700 - but because it's a less capable sensor, it wasn't able to resolve it. Comparing both cameras at the pixel level as though that's equal footing is wrong, and is extremely unfair to the D800 when talking about things like this.
David,
Regarding testing lenses, I hadn't read your reply yet but this weekend I tested all my lenses, no flat subject or manual live view focus though. I may set up again and do that. Never the less I now have a much more comfortable feeling about using each lens with regard to diffraction softening. By the way, the big winner in my lens set was the Nikon 16-35.
Testing also brought out other quirks that were interesting in each lens. The oddest was that two lenses have one particular f/stop that produces a significant increase in contrast. On my 70-200 f/4 VR for instance it's always f/11 regardless of focal length. Strange, but at least I know it's there!