I have read the D800 Technical Guide at the Nikon site concerning diffraction.
My question is, if I change the image size from 36x24 (36MP) to 30x20 (25MP) will this lessen the diffraction when shooting above f 11?
I am not sure if it is the pixel size or the MP amount that effects the diffraction.
-------------------

My thoughts on the matter would be, by reducing image size(Down-sampling), you would gain some sharpening and would counteract any diffraction, but would limit overall print size. But, as many could tell you from their experiences, a 25mp image would still be more then enough to print very large from....

Just my 2 cents,

Chris

Visit my Nikonians gallery.

Hi!

Diffraction - the point at which the properties of the Universe (Laws of physics) render our efforts to increase resolution fruitless. Any system manipulating waves such as visible light, x-rays, or electrons, will ultimately be diffraction limited. The Wikipedia entry for diffraction is 'illuminating'. For camera lenses I would also recommend:

http://luminous-landscape.com/tutorials/resolution.shtml

My own experience in microscopy was that every day was diffraction limited when the quality of the optics, the object, and the lighting conditions were high enough.

In camera lenses diffraction occurs when the light waves go past an edge. This edge is the blades of the diaphragm. The effect on resolution is dependent on the diameter of the hole formed by the edge. See the Wikipedia article for the mathematics. The loss of resolution will not be seen if other aberrations are greater. The loss of resolution will be obscured if the resolution of the recording medium is lower than the change caused.

So with the best lenses and the best high definition film or digital sensor, will resolution be diffraction limited? Of course. It has been for years! Or rather for ever, but camera lenses are just late arrivals.

It was a major factor in my work photographing fabric weaves and fibres in close-up. Technical Pan film rated at 1/4 ASA, and Nikkor 55 mm f2.8 Micro and 105 mm f4 Micro were our standard tools. With any three dimensional weave depth of focus at smaller apertures had to be balanced against loss of resolution to obtain the most detailed pictures. f 8 was the 'normal' setting, f 5.6 preferred, and f 11 only used in desperation. That was over 25 years ago.

Today with a Nikon D800E and the best lenses such as the two above, or modern Nikkor Micro lenses, or Zeiss Makro lenses, the resolution of the recording system reveals more than even the Technical Pan film could. Thus diffraction is more frequently limiting. To avoid it I do not go below f 5.6. However as ever resolution may have to be sacrificed to depth of focus. A less good lens with optical aberrations, focus shift if not using LV, mirror vibration, etc etc all remove resolution so diffraction only becomes limiting at smaller apertures.

Thus downsampling, because it reduces resolution, will mean diffraction becomes limiting at a smaller aperture. It is still there but the resolution has already been thrown away. I was content at about one stop smaller on a D3X than on the D800E, and two stops smaller on a D2X.

Bring on the 60 MP FX camera on which nothing less than f 4 is acceptable to the resolution freaks!

Roderick

Visit my Nikonians gallery.

To my knowledge, changing the image format on a D800 merely crops the image and does not down sample anything. It therefore has no incidence on diffraction.

Cheers,
Bernard.

----

frame4zoom for Nikonians

As stated, diffraction has been a law of optical physics since lenses appeared. That said, far more photographic defects are the results of numerous technical and technique failures by photographers than from diffraction. Unless you are doing specialized work such as macrophotography, DOF, adequate shutter speeds and proper focus are much more important. I hesitate to put it in the same category, but too often, issues of diffraction are similar to the ago-old question of how many angels could dance on the head of a pin. My advice is, take photographs, seek and share beautiful images and worry less about diffraction. Don't shoot as many brick walls or test charts either.

Alan
Waterloo, IL, USA
www.proimagingmidamerica.com

Visit my Nikonians gallery.

Good advice Allan

Visit my Nikonians gallery.

Alan's right on the money. Diffraction gets yards and yards of concern among gadgeteers but in normal photography done by photographers it's a minor problem that can be corrected easily in Photoshop.

Russ Lewis
www.FineArtSnaps.com

Hi,

I was very concerned about this before I got my D800 as I felt I would be limited to f/8. In practice though I have taken many macro shots using f/11 - f16 and it hasn't been an issue for me, although I don't make large prints. Good technique is more important.

Enjoy!
Randy

The practical side of diffraction on the D800 is because of the large number of pixels, diffraction becomes an issue at about one stop wider than the D700/D3 - and that's about the same as the D300. Diffraction starts to be measurable at around f/8, but is rarely a problem until you get to f/11 or higher.

Diffraction impacts the sharpest area of your image - not the soft areas. If lens quality, DOF, motion, or anything else softens your image, the impact of diffraction is outweighed by the softness (in normal aperture ranges). Diffraction is most relevant at or near the focus plane.

The tradeoff on diffraction is increasing DOF. DOF should drive your composition with diffraction issues being a secondary consideration. With macro in particular, you often are struggling for adequate DOF and stopping down to diffraction impacted apertures. But composition needs would justify stopping down and accepting some diffraction impact. With landscapes, you may find that DOF tables show a small impact of moving from f/11 to f/16, and the diffraction impact may not be acceptable. With wide lenses even f/8 may provide plenty of DOF.

I don't worry about diffraction at all below f/8. I willingly use f/8 to f/11 as needed knowing the diffraction tradeoff is minor. I try to only use f/13 to f/22 only if needed with composition driving that decision. And I very rarely go beyond f/22 for any reason - even with macro.

As for your original question about resizing, the high resolution images of the D800 can be sharpened with a higher intensity and small radius before downsizing to recover edges impacted by diffraction. Then after resizing you can do a second pass of more typical sharpening as needed. This is probably a matter of taste for individual images.

Eric Bowles
Nikonians Team
My Gallery
Workshops

Nikonians membership — my most important photographic investment, after the camera

Diffraction is purely a function of photosite size, not count.

That says that a larger sensor, with the same pixel count, will result in a smaller aperture yielding the same diffraction performance characteristics. In other words, 36mpx FX will be superior to 36 mpx DX, in terms of "diffraction limited" resolution at the smallest possible aperture. And Medium Format would be better still.

But larger sensors require smaller apertures to maintain the same depth of field so in the end it can be a net push . That is the conundrum of ever increasing sensor density and sensor size for macro and landscape photographers, and even wildlife photographers shooting TCs.

Assuming you are asking about shooting the 1:2 format vs full sensor, 1:2 will not improve things because the photosite size is unchanged. You are probably better off shooting full frame and getting a bit closer or using a longer focal length to similarly fill the frame. The only argument for 1:2 here is that in principle you can shoot a slightly smaller aperture and get the same depth of field, nullifying the difference. But most people would probably not use that much precision in selecting an aperture for a given shot, and the computations are a bit fuzzier than most people consider (keep reading).

The issue is not if diffraction will destroy the image from denser/higher resolution sensors. The issue is simply that the additional cost of those sensors results in very diminishing (or no) returns. That cost is not only the camera but all the storage, and the more limited shooting frame rate if that is important.

Just as an example, at f/22 or F/32 a D800 likely results in little additional resolution over a D700. It might be a mistake to upgrade a D700 to a D800 just to go shoot images at f/22 or smaller.

Due to the lower true (color) resolution of the Bayer filter, and some other issues, there may be more real world benefits to shooting A D800 at small apertures, and even at F/22, than the simple diffraction formulas suggest. I think it is a complex issue that is not well modeled by standard diffraction formulas.

The incremental cost of higher resolution and sensor densities (and digital storage) is declining. At this point in time a D800 might make sense over a D600 simply for the added feature set- those features may make it worth the ~$1000 msrp price difference. I think that is an important consideration in the context of the poor modeling of the diffraction calculations. Same cab be said for the decision to shoot 1:2 to try to solve any of these problems.

_________________________________
Neil
Nikonians Team
My Gallery

>DOF should drive your composition with diffraction issues being a
>secondary consideration.

Thanks Eric

I think diffraction effects are over-rated. I recently shot some mountain scenery with rockpools and wildflowers in the close foreground and mountains in the background. I needed to get it all sharp. f11 was not enough. Even with techniques like hyperfocal distance being used, too much of the image was not in focus.

f16 provided enough DoF and the images were way better, viewed in their intirety. The sharpest part has theoretically been degraded by diffraction but I can't really see this unless I zoom in to ridiculous levels at which the overall context of the shot is lost.

I would hate to see a situation where less experienced photographers read a thread such as this and afterwards feel inhibited about using apertures smaller than f8

Visit my Nikonians gallery.

I am a Certified Public Accountant, who will be retiring in June.

As an accountant, I have never been good with di fractions.

Ooohhh.

"Today is the tomorrow that yesterday you spent money like there was no"

Visit my Nikonians gallery.

>> I would hate to see a situation where less experienced photographers read a thread such as this and afterwards feel inhibited about using apertures smaller than f8

I think a lot of people have trouble with the idea that these cameras will not deliver the highest possible resolution in all circumstances.

Resolution is just one aspect of an image that can be traded off for others, as is the case for most technical aspects of an image. This is the part that creates so many problems and misunderstandings.

_________________________________
Neil
Nikonians Team
My Gallery

Thank you all for your replies. I posted my question due to what I read in the Nikon Tech. Guide. I have reaquired a 105 Macro lens and will be using it with small apertures, probably below f11. Guess I will have to experiment and see what I get.

Diffraction ordinarily gets to be more of a problem for a given aperture as you decrease the size of the format. The reason is simply that if the in camera image is smaller, you have to enlarge it more for the same size final print, so the Airy disk created by diffraction will be larger in that print.. It may get a bit more complicated than that for digital images because the size of the pixels may be comparable to that of the Airy disk. But in general you want to increase the size of the format if you want diffraction to be less of a problem, not decrease it.

I do some large format 4 x 5 photography, and diffraction is seldom a problem for me, I regularly use apertures such as f/32, which would be severely impacted by diffraction in 35 mm or smaller. My lenses often may be set as small as f/64. A 35 mm lens would never allow such a small aperture.

Leonard Evens
Professor Emeritus, Department of Mathematics, Northwestern University