Diffraction Effects (Example Photos)

Started Nov 27, 2011 | Discussions
Shop cameras & lenses ▾
DvD5 Contributing Member • Posts: 831
Diffraction Effects (Example Photos)

After all the argument in previous threads, I decided to try this out for myself. I shot images with my old D50 (6MP) and D7000 (16MP) at f/32 with my 55mm Micro Nikkor. Please forgive the nasty reflections—I used a flash because I wanted to eliminate camera shake as a variable. These were shot in RAW and opened with Adobe Bridge.

Here are the images out of the camera; D50 on top, D7000 bottom:

Here, the D7000 is resized to match the D50 (6MP), they are both sharpened the same amount:

I also took the same shot at f/8 on the D7000 and resized/sharpened. It is the bottom one in this photo:

Clearly, the D7000 at f/8 is much better than the other two images, but it also appears that the D7000 at f/32 is somewhat better than the D50 at f/32. Look at the small lettering on the white book “Ballantine Books”, or the gold lettering on the black book on the right “Harper Collins”.

The D7000 is about the same pixel density as the rumored D800; The D50 would be about 13.7MP if scaled up to FF. I know this is not exactly apples to apples, but give you a ballpark approximation of the differences between a D700 density and D800 (rumored) density...

Don

-- hide signature --
Nikon D50 Nikon D700 Nikon D7000 Nikon D800
If you believe there are incorrect tags, please send us this post using our feedback form.
Flashlight Veteran Member • Posts: 6,175
Re: Diffraction Effects (Example Photos)

Good work, results as expected. Some people, from these other threads, really should set their minds in a different mode.

-- hide signature --

Philip

Deleted-pending Senior Member • Posts: 2,665
Re: Diffraction Effects (Example Photos)

Thanks, some posters here don't have the guts or knowledge to show some real life application, like real scientists do.

Kaj E Veteran Member • Posts: 9,085
Re: Diffraction Effects (Example Photos)

Your test shots confirm theory.

The diffraction effect is more visible on a per pixel on the higher density pixels, but this image is sharper on the image downsized to same image size. This is due to the reduction of the limitations by the sensor resolution for the lens (diffraction) +sensor system for the D7000. Compare to my earlier post:
http://forums.dpreview.com/forums/read.asp?forum=1021&message=39933143

By opening up the lens to f/8 you reduce diffraction and get a sharper image.

-- hide signature --

Kind regards
Kaj
http://www.pbase.com/kaj_e
WSSA member

It's about time we started to take photography seriously and treat it as a hobby.- Elliott Erwitt

DvD5 OP Contributing Member • Posts: 831
Re: Diffraction Effects (Example Photos)

Kaj E wrote:

Your test shots confirm theory.

The diffraction effect is more visible on a per pixel on the higher density pixels, but this image is sharper on the image downsized to same image size. This is due to the reduction of the limitations by the sensor resolution for the lens (diffraction) +sensor system for the D7000. Compare to my earlier post:
http://forums.dpreview.com/forums/read.asp?forum=1021&message=39933143

[...]

Yes--I'm not surprised.

-- hide signature --
Cytokine Contributing Member • Posts: 661
Re: Diffraction Effects (Example Photos)

It is interesting that while the D7000 is resolving more detail, have a look at the texture on the green book cover, It is also is the most affected or changed by diffraction.

So diffraction affects high Mega Pixel cameras more than low MP cameras, as expected, but high Mega Pixel cameras still provide better resolution.

John

DvD5 OP Contributing Member • Posts: 831
Where does diffraction set in

Just for kicks, I did another test to see where diffraction really kicks in for the D700. I shot a dollar bill with my 105mm Micro Nikkor and put these together at 100% in PS.

Not much difference between f/8 and f/11.
Diffraction starts creeping in a bit at f/16
Really picks up at f/22 and f/32

Don
--
http://www.dvd5.zenfolio.com

DvD5 OP Contributing Member • Posts: 831
Re: Diffraction Effects (Example Photos)

Cytokine wrote:

It is interesting that while the D7000 is resolving more detail, have a look at the texture on the green book cover, It is also is the most affected or changed by diffraction.

On the second comparison, the D7000 is on the bottom. The green book on the far left is sharper with the D7000 than the D50.
--
http://www.dvd5.zenfolio.com

Cytokine Contributing Member • Posts: 661
Re: Diffraction Effects (Example Photos)

DvD5 wrote:

Cytokine wrote:

It is interesting that while the D7000 is resolving more detail, have a look at the texture on the green book cover, It is also is the most affected or changed by diffraction.

On the second comparison, the D7000 is on the bottom. The green book on the far left is sharper with the D7000 than the D50.

Yes, that's what I meant.

Thanks for the tests.

-- hide signature --
Cytokine Contributing Member • Posts: 661
Re: Where does diffraction set in

I suppose the benefit of more pixels, must have an end point, presumably that point is limited by the lens, if so, we cant be far away from the end game.

John

James Bligh Senior Member • Posts: 1,809
Re: diffraction is a gradual thing so you need some margin of error in prediction.

In D3 & D700 diffraction sets in around F16.

In D3X diffraction sets in around F8 & F11.

In D800 if it is 36 MP FF I guess diffraction will set in around F5.6 & F8.

BTW diffraction is a gradual thing (See Thom Hogan's D3X review.) so you need some margin of error in prediction.

http://bythom.com/nikond700review.htm

http://bythom.com/nikond3review.htm

http://bythom.com/nikond3xreview.htm

-- hide signature --

Decision, decision, we are the slaves of decisions we made inadvertently.

May God forgive us our imperfections!

Dust to dust, what on earth are we entitled to claim?

AV Janus Senior Member • Posts: 1,994
Re: diffraction is a gradual thing so you need some margin of error in prediction.

But since it depends from the size of the hole the light passes through, wouldn't it be also dependent on focal length of the used lens?

Is f8.0 aperture on 24mm the same diameter as f.80 on 135mm lens?
--
Rick Halle wrote:

" Keep in mind that tall buildings sway back and forth so they require faster shutter speeds."

 AV Janus's gear list:AV Janus's gear list
Olympus OM-D E-M5 Olympus OM-D E-M10 Olympus M.Zuiko Digital ED 9-18mm F4.0-5.6 Olympus M.Zuiko Digital ED 75-300mm 1:4.8-6.7 Samyang 85mm F1.4 Aspherical IF +10 more
Kaj E Veteran Member • Posts: 9,085
Re: diffraction is a gradual thing so you need some margin of error in prediction.

AV Janus wrote:

But since it depends from the size of the hole the light passes through, wouldn't it be also dependent on focal length of the used lens?

Is f8.0 aperture on 24mm the same diameter as f.80 on 135mm lens?

The diameter of the airy disk is directly proportional to (light wave length) X (focal length):(aperture diameter). Focal length/aperture diameter is actually the f-stop number for the lens. The size of the airy disk is in this way only dependent on f-stop number and wave length of light.

The focal length is already included in the f-stop number.
--
Kind regards
Kaj
http://www.pbase.com/kaj_e
WSSA member

It's about time we started to take photography seriously and treat it as a hobby.- Elliott Erwitt

James Bligh Senior Member • Posts: 1,809
Re: different distances lights travel through cancels away different diverging angles

AV Janus wrote:

But since it depends from the size of the hole the light passes through, wouldn't it be also dependent on focal length of the used lens?

Is f8.0 aperture on 24mm the same diameter as f.80 on 135mm lens?

If you see the diffraction limit calculator there is no variable such as different focal length lens type, instead it has a variable of Camera Type.

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

I am not a physicist, but I guess why this way. Parallel light rays which pass through a small aperture begin to diverge. Diverging angle will be steeper in F8 of 24mm compared with F8 of 135 mm. But the diaphragm of longer focal length lens is far from the imaging surface (sensor) compared with that of short focal length lens. So the different distances lights travel through aperture holes cancels away different diverging angles and the circle of divergence of rays in the imaging surface (sensor) will be same between different focal length lenses in same F number.

-- hide signature --

Decision, decision, we are the slaves of decisions we made inadvertently.

May God forgive us our imperfections!

Dust to dust, what on earth are we entitled to claim?

DvD5 OP Contributing Member • Posts: 831
Two extremes here...

Just to demonstrate two extremes of focal length, I performed a test with the 24mm PC-E lens and the 500mm f/4 G. You can see in both cases, there is a small difference between f/8 and f/11. f11 to f16 a bit bigger, then f/16 to f/22 a large step. The 500mm doesn't go to f/32, so couldn't test that.

So, it looks like the practice matches the theory again.

24mm PC-E (100%):

500mm f/4G (100%):

Don
--
http://www.dvd5.zenfolio.com

James Bligh Senior Member • Posts: 1,809
Re: What camera did you shoot these pictures with?

DvD5 wrote:

Just to demonstrate two extremes of focal length, I performed a test with the 24mm PC-E lens and the 500mm f/4 G. You can see in both cases, there is a small difference between f/8 and f/11. f11 to f16 a bit bigger, then f/16 to f/22 a large step. The 500mm doesn't go to f/32, so couldn't test that.

So, it looks like the practice matches the theory again.

24mm PC-E (100%):

500mm f/4G (100%):

We are talking about diffraction here. What camera did you shoot these pictures with? I think you have forgotten to mention it.

-- hide signature --

Decision, decision, we are the slaves of decisions we made inadvertently.

May God forgive us our imperfections!

Dust to dust, what on earth are we entitled to claim?

DvD5 OP Contributing Member • Posts: 831
Re: What camera did you shoot these pictures with?

James Bligh wrote:

We are talking about diffraction here. What camera did you shoot these pictures with? I think you have forgotten to mention it.

D7000 (mostly trying to make a point out of the pixel density on the rumored D800)

-- hide signature --
Lance B Forum Pro • Posts: 28,904
Realistic viewing distances.

James Bligh wrote:

In D3 & D700 diffraction sets in around F16.

In D3X diffraction sets in around F8 & F11.

In D800 if it is 36 MP FF I guess diffraction will set in around F5.6 & F8.

It depends on the print or image size and how far away you view it. It is a very broadbrush statement to say that diffraction sets in at a particular aperture as it depends on many factors like viewing distance, eyesight, print/display size etc. Look at Cambridge Colour's site for a calculator.

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

As written in the other thread with regards to diffraction limitations, a 36 Mp full frame (35mm) sensor would start to be diffraction limited at f/11 or more but not at f/8 given a 36" print/display, at standard viewing distances, which the article claims is 25cm or 10 inches and that is far too close for a 36inch print/display. A more realistic viewing distance for a 92cm or 36inch print/display would more likely be closer to 1mt or 40inches. At this size and vewing distance that equates to a diffraction limit of f22, which is plenty of DOF. Even at a viewing distance of 50cm or 25inches (which I think is too close) that equates to a diffraction limit of f13 which is still big DOF in my books, depending on camera to subject distances. A more realistic distance might be 75cm and therefore an aperture of about f16 diffraction limitation.

However, a more realistic normal print/display for most of us at say, A3+, or 485mm (20inches) and a viewing distance of say 50cm (25inches) results in a diffraction limit of f22. This size dsiplay relates to many computer screens and large home based printers.

Looking at my computer screen, it is 21inch and I sit at about 60-70cm away from it and I would say that is about as close as I'd want to sit to that size screen. Even a 24" screen would be about as close as I'd say most would want to be as well. So, at 24" display, and 70cm, then I would say that diffraction would not be an issue to at least f18. Even if we allow for a screen to show more detail, I think f16 would still be enough.

So, for a FF 36Mp sensor, it would seem that you can still shoot at high f numbers like f16 or more as the viewing distances are such for a given print size that it means that diffraction is not an issue and even at the cusp of diffraction limits, as the article states, there is more to it than straight diffraction as many other factors come into play. If we go over 36Mp for FF, then diffraction may be a limitation, but again it is very dependant on viewing distances.

BTW diffraction is a gradual thing (See Thom Hogan's D3X review.) so you need some margin of error in prediction.

http://bythom.com/nikond700review.htm

http://bythom.com/nikond3review.htm

http://bythom.com/nikond3xreview.htm

-- hide signature --
 Lance B's gear list:Lance B's gear list
Nikon D810 Nikon D500 Nikon AF-S Nikkor 16-35mm F4G ED VR Nikon AF-S Nikkor 14-24mm f/2.8G ED Nikon AF-S Nikkor 24-70mm f/2.8G ED +16 more
Deleted-pending Senior Member • Posts: 2,665
Re: What camera did you shoot these pictures with?

DvD5 wrote:

James Bligh wrote:

We are talking about diffraction here. What camera did you shoot these pictures with? I think you have forgotten to mention it.

D7000 (mostly trying to make a point out of the pixel density on the rumored D800)

-- hide signature --

did you use mirror lock-up ?

DvD5 OP Contributing Member • Posts: 831
Re: What camera did you shoot these pictures with?

FTH wrote:

DvD5 wrote:

James Bligh wrote:

We are talking about diffraction here. What camera did you shoot these pictures with? I think you have forgotten to mention it.

D7000 (mostly trying to make a point out of the pixel density on the rumored D800)

did you use mirror lock-up ?

No--I used a tripod with a flash. This should be very effective at eliminating any motion blur.

-- hide signature --
Keyboard shortcuts:
FForum MMy threads