[Tamia]

The thread K-5 IIs - are my lenses "good enough" over at the Pentax SLR Lens Discussion portion of the forum makes me realize how little I know about gear, in this case, how lenses will behave at different apertures. I understand controlling DOF and using hyperfocal distance, but when someone says something like "... don't stop beyond f8; after f11 the diffraction degrades seriously image quality" I don't understand WHY. Moreover, I don't know if this is specific to particular focal lengths or is a universal axiom. And if diffraction is so poor beyond a particular point, why bother having apertures that tight? (The nature of my questions makes it clear how little I know.)

Of course, this forum is filled with loads of knowledge, but I hope someone can point me toward one or two books or articles where I can begin to understand lenses the way you do.

Thanks!

[JinDesu]

Diffraction depends on the size of the pixel on the sensor. On a sensor with lots of pixels, the pixels must be smaller. This link may help to understand diffraction and how mathematics can tell us what aperture starts to show diffraction: Diffraction

As far as why bother having apertures that small:
Some people don't care about having ultimate sharpness. If I am taking an image that will then be resized to fit online - having my images blurred by 1 or 2 pixels from diffraction isn't going to matter. Instead, if it's a scene where I need to stop down for maximum depth of field or to reduce the light, then I'd rather have a slightly softer image than no image at all.

[jatrax]

I am sure someone with more technical knowledge will also jump in but a simplistic explanation of diffraction is that the edges of the aperture blades 'diffract' the light slightly. When the lens is wide open the percentage of diffracted light compared to non-diffracted is small and so it makes little impact on the image. When the lens is stopped down (f/22 or f/32) the percentage of diffracted light is significant.

As I understand it the diffracted light hits the sensor slightly differently than straight in light making the image soft looking. Different lenses have different aperture settings where the diffraction becomes significant, and I have been told that different sensor sizes make a difference as well.

And the reason for the f/22 settings is that sometimes that is what it takes to get the shot and you have to put up with the diffraction. All of photography is a compromise one way or another.

Diffraction is not something that just suddenly happens at a particular f-stop. It is always present it just gradually becomes more noticeable as you stop down.

Like I said that is the simple explanation, I look forward to seeing the technical explanation.

[bobpur]

Don't spend too much time reading about other peoples ideas and go out and take a lot pictures, you can't learn to make great photos by reading a book.
It takes a lot of time to learn about each lens as they are all different, you have to know your tools to get the best out of them.
Some lenses are only good at 8 or 11 others at 2.8 ect.
Sometimes it is a trade off between maximum sharpness and Depth of Field then use 11 or 16 or 22
Too much worry about Tech and not enough about Artistic, composition and color .

[Digitalis]

Diffraction is not something that just suddenly happens at a particular f-stop. It is always present it just gradually becomes more noticeable as you stop down.

exactly, It is also important to note that Diffraction is as much a sensor dependant as it is lens dependant - there are no 35mm lenses that perform well at f/22 or higher. With a takumar 50mm f/1.4 lens the image quality starts a bit on the soft side, but still usable at f/2.8 we see improvement, at f/4 this trend continues until we reach the optimum aperture of f/5.6, at f/8 the resolution is slightly lower (but still extremely good) at f/11 it lowers a bit more, at f/16 there is a observable loss of acutance, at a hypothetical f/22( Takumars can't stop down this far) a considerable amount acutance in the image will be lost.

Lenses follow a shallow bell curve*, there is a peak performance and then there are the F/stops on either side of that optimal point.

*Pancake lenses tend to have a bit of a linear characteristic in terms of resolution, with a well designed pancake lens (The DA70mm f/2.4 springs to mind)the image quality at f/2.4 is already extremely good, the only reason to stop that lens down is to increase DOF.

[Tamia]

Thanks for the quick, helpful replies -- very much appreciated.

JinDesu: Thanks for the Rockwell link. I've saved the article and will read it carefully. I never considered that the sensor would play a role here. And as for ultimate sharpness, it's something I strive for at some times but not others, depending on the subject and what I hope to achieve. And I couldn't agree more when you say "I'd rather have a slightly softer image than no image at all."

John/Jatrax: Your "simplistic explanation" really worked well for me. Thanks! Makes so much sense when diffraction is put the way you did. Point taken on how diffraction gradually becomes more noticeable. You're right when you say photography is a compromise somehow. That's part of what makes it so enjoyable, to my mind. Thanks again!

Bobpur: No, I don't spend a lot of time trying to meet other peoples' ideas about good shooting. It was the technical aspects of lenses and sharpness-softness which I was curious about, since I never considered a lens would be soft at tight apertures due to diffraction (blurring, yes, with slower shutter speeds). I agree with you entirely on the importance of practicing and striving for better images. It's a continuous learning experience! Thanks very much for your help.

Digitalis... The importance of sensors, again. Thanks very much for your excellent description about the Takumar 50mm f/1/4. The shallow bell curve description works very well for me. Interesting about the pancake lenses.


Once more, thanks to everyone for your help. I already feel as if I have a better grasp of the subject.

[wombat2go]

Learning how to use use to K-01 today, the 18~55 was used on a tripod to check diffraction at 55mm and at f/8 and f/38
I was expecting the diffraction blur, but also the loss of contrast is maybe more of an issue in these shots (camera jpg, 4MB)
F/38
https://www.box.com/s/40c8y0xravvhnexehpgt
F/8
https://www.box.com/s/qi4feewbr3vlzujvev6x

I will do some more at f/11 ~22 to see how far to stop down acceptably.
Sensor size:
On 6X7 b/w, f/22 is very sharp on the 22 inch monitor.
On 35mm I often use f/11 and f/16 with iso 400 and 800
On M43 DOF is not an issue so I usually just set one stop down and let the metering work.
Actually I made a f/4 fixed waterhouse for the M43 with Pentax 110 18mm and it seems to work - ..just(?) OK for a 22 inch monitor
M43 18 mm DOF at f/4 , no need to stop down
https://www.box.com/s/r48i1b3wgpjmo5h4l70i

[Just1MoreDave]

All of photography is a compromise one way or another.

I quoted this because I sometimes have made stupid choices for some arcane technical reason. Like stopping down for sharpness when I'm in a dim restaurant, when it might cause motion blur or high ISO noise instead. Some people here have said that they never use any of their lenses wide open, possibly fearing a lightning bolt from the photography gods. Learn more but don't let the knowledge get in your way.

[jatrax]

I sometimes have made stupid choices for some arcane technical reason.

Yep, took a whole series of wonderful images in fantastic 'golden' light last fall. Everything just perfect. Except I had it in my head I wanted to maximize DOF. So I stopped all the way down to f/22 on the DA 21mm. Let's just say the results were not what I was looking for. And there was no reason to do that, even at f/8 everything over 10' would have been in focus anyway. And I knew better, just had a mental block and wasted a great opportunity.