Advise on f-stop

[ketan]

Hi,
I recently tried to take some photos of shells and rocks etc.
I used Tamron 28-78 f2.8 on my 30D.
After seeing the photographs I realised that whenever I wanted to have sharp pictures and used apperture f22-f32 the time value were ranging around 1/50 - 1/15 at ISO 400.
The photos were not very sharp. In fact I got the same results when I used self timer to avoid shake.
What apperture works best for such photography when I am attempting to click from about 2 feet distance?
----------------------------------------------------------
f/32 - 1/30

---------------------------------------------------------------------
f/4 - 1/1600

----------------------------------------------------------------
f/2.8 - 1/4000

------------------------------------------------------
f/7.1 - 1/500

 

[Digital Matt]

ISO 100, a good sturdy tripod, and try again.

 

[ketan]

ISO 100, a good sturdy tripod, and try again.

I will try with again with same apperture range.
I had everything except I kept ISO 400 to have good shutter speed.
I hope that the distance of 2ft has no role in this. I mean should I keep it to say 5ft and then crop the photo to get this...
Thanks

 

[Deimodius]

Ketan, your lens should have a minimum focus distance, and if it doesn't have Macro capability, then being _closer_ than the MFD will make it harder to focus (even manually).

For example, on my Olympus E-500 main kit lens (14-45mm) along the ring where it attached to the body it says "0.38m/1.3ft - Infinity" which means the closest I can get with that lens is 1.3 feet before I can no longer focus properly.

The only other thing I can think is that if the shutter speed is too slow (because of the narrow aperture) you would be getting camera shake unless you have a tripod as Digital Matt suggested. In general, it is best not to hand hold a camera at shutter speeds below 1/60 (if you have steady hands), and that changes the longer your lens. If you were using a lens that went to 200mm, you would have to use a Shutter Speed of 1/250 or greater.

 

[Helen B]

Moving away and cropping would probably be worse. The shots show that the lens is capable of focusing on the rock, and the spec for the lens (if I'm looking at the right lens) says that the minimum focus distance is about 1 ft through the entire range.

Experimentation is always the best way to learn the capabilities of your lens, but here are some general comments:

Though the rule of thumb that Deimodius quotes is often used as a starting guideline for documentary work etc, it is not suitable when critical sharpness is required. Camera shake doesn't suddenly go away at some magic shutter speed - it just has less and less influence on the softness of the image, and as it does it looks less and less like camera shake. Like Matt says, use ISO 100, a good tripod and mirror lock up.

Lenses designed for good optical performance at close distances are comparatively rare. Don't expect the same sharpness at very close distances that you get at longer distances. Just because a lens has a 'Macro' designation does not mean that it is optically corrected for macro work - it may just mean that it can focus at fairly close distances.

What focal length setting were you using? It is usually best to zoom in for close-ups (there are many exceptions, especially with lenses that focus more closely at wide zoom settings). For the same magnification the depth of field at a fixed f-stop is the same at different focal lengths (this is not true for distant scenes). As an exercise compare the close-up performance of your lens at (say) 37 mm from 1 ft (or slightly more) vs 75 mm from 2 ft.



Finally, to what is probably the most important point:
Your lens might be capable of being stopped down to f/32, but diffraction will almost certainly be softening the image at f/32.

It is well worth learning the trade-off between depth of field and diffraction through practical experience. Diffraction causes what should be a perfectly focused point of light to become a small circle (greatly simplified explanation). As you stop down, the diameter of that circle increases. This will eventually have a visible effect on the image (possibly at f/5.6, depending on a number of factors). It means that as you stop down further the parts of the image that appeared to be in focus at wider apertures become softer (diffraction), and the parts of the image that were out of focus become sharper (depth of field) until there is a near-uniform sharpness or softness.

Your attitude of trying this for yourself in practice is the best. Keep going!

Best,
Helen

 

[ketan]

What focal length setting were you using?

All are at 75mm i.e. full length.

Your lens might be capable of being stopped down to f/32, but diffraction will almost certainly be softening the image at f/32.

Well, this is a new lesson for me and I will spend time and try to learn on th net.
Thanks a ton for taking out time.
Ketan

 

[Helen B]

Ketan,

I'll try to put some numbers to it, to give some idea of how big the effect could be. This is a very basic explanation, with plenty of approximations and simplifications. I can desimplify it if you wish!

As rule of thumb, diffraction turns what should be a point of light into a circle of light surrounded by darker rings of light, and the circle of light has a diameter that is approximately equal to the f-Number in micrometers (μm). The circle is called the Airy Disk.

Therefore at f/32 what should be a focused point of light* is about 32 μm in diameter. The 30D has 3504 pixels across 22.5 mm, ie 6.4 μm per pixel**. What should be a point of light is covering about a dozen pixels.

There is another way of looking at it, because you can do the same calculation on the other side of the lens. A 75 mm lens at f/32 has an entrance pupil diameter of 75/32 mm, or about 3/32 inches (75 mm is about 3 inches). If the subject is 2 ft away (24 inches) then the relative aperture is

24 * 32/3 = 256

That means that on the object side of the lens, the smallest detail that can be resolved by a 75 mm lens 2 ft away at f/32 is about 250 μm, or about ¼ mm.

*There are various lens aberrations etc that will also turn what should be a point of light into a disk or other shape.

**Compare this to the rough estimate of diffraction possibly beginning to have an effect on the image at about f/5.6.

Best,
Helen

 

[littlesandra]

Try ISO 100, 200.
Get a meter reading before shooting, and adjust accordingly then.

 

[ketan]

Helen,
Many thanks.
This seems like whole new world opened up...I found few things on
http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm
and
http://www.luminous-landscape.com/tutorials/understanding-series/u-diffraction.shtml
But what baffles me is that if the stops change from f/11 towards f/32 how come image start becoming softer!!!! (I always get confused that from f/11 towards f/22 is called step up or down). I swaer I saw many razor sharp images at f/32 in the books (learning to see creatively). How does one achieve that?:er: How to achieve the best compromise? On what factors does it depend on - focal length or there are some universal rules?
My search will continue...
Thanks again Helen for guiding.
Ketan

 

[Garbz]

The bell curve. Basically the lens is sharper at the centre than at the outside. Lots of quality problems occur in all but the finest aspherically shaped glass. This is what causes a multitude of problems at wider (lower) apertures. As you begin stopping down the edge stops having an effect and the far less curvy centre of the glass dominates the sharpness. But keep going smaller and diffraction elements start causing problems. The smaller the beam of light gets the worse diffraction. At one point if you get small enough (when the width of the aperture becomes exactly the wavelength of the light beam hitting it) the light wave will hit it and will entirely disperse on the other side. This means a plane wave turns into a big circular point source on the other side and the light bounces all over the sensor, and at this point the lens it totally useless. Fortunately f/32 is not small enough for that

 

[Helen B]

...
But what baffles me is that if the stops change from f/11 towards f/32 how come image start becoming softer!!!! (I always get confused that from f/11 towards f/22 is called step up or down).

Confusing? Yes, raising the f-Number from 11 to 22 (ie going from an aperture of f/11 to f/22) is 'stopping down' (two stops in this case). The aperture becomes smaller, and less light passes through it.

I swear I saw many razor sharp images at f/32 in the books (learning to see creatively). How does one achieve that?:er: How to achieve the best compromise? On what factors does it depend on - focal length or there are some universal rules?
My search will continue...

Well, keep reading and keep asking questions about things that are unclear.

Why can a picture taken at f/32 look sharp?
The rule of thumb that I gave above suggests that at f/32 the nearest anything in the image will be to a true point is a smear about 32 μm in diameter. Just suppose that you were using film, and you made a contact print (ie the print is the same size as the piece of film). A smear 32 μm in diameter would look like a point if we viewed the print from normal viewing distance. If you want to equate that to a digital print, it would be something like an 800 dpi print (these numbers are intended to give an indication of magnitude, not for film vs digital mudslinging).

If you go along with the idea that a 300 dpi print is good enough to appear smooth - ie you aren't seeing the individual dots at normal viewing distances* then you might estimate that you could double the diameter of the f/32 smear and still be just below the size at which we would see it as a circle instead of a point. If you double the f-Number, you double the diameter of the smear - so f/64 would be OK to use if you were going to make a contact print. And it is. If you use a large format 8x10 camera with the lens stopped down to f/64 and then make 8x10 contact prints they will be razor sharp.

Now suppose that you are making a 2x enlargement. Using a piece of 4x5 film to make an 8x10 print. To keep the size of the smear on the print to 64 μm, the smear musn't be more than 32 μm on the film - so f/32 will look razor sharp.

And so on, as the format size comes down. It's not quite that simple, but it is a start, to accompany the other stuff you are reading.

I use f/32 with full-frame 35 mm when I absolutely need depth of field above all else (or I used to, until focus stacking became practical). One thing you do in that case is to ensure that the entire picture is at equal sharpness or softness, and not to have sharper images nearby.

What is the practical limit?
With a three-dimensional subject such as your rock, the overall sharpness can increase as you stop down. The parts that were in perfect focus at f/2.8 may get sharper at first as lens aberrations are reduced then softer as diffraction takes over. The parts that are way out of focus just get sharper as you stop down as the depth of field increases. They may reach a balance where the focus blur equals the diffraction blur. That would be the limiting aperture in that particular case. Then it is a matter of whether or not you can accept that degree of blur. This is something to learn from experience. Every case is different - the best f-stop depends on the situation. It is not a fixed value, unless you are taking pictures of flat objects all the time.

Photo.net discussions can be lively, surreal, entertaining, arcane, confusing, informative and embarrassing all at once. Here are some threads on diffraction, in no particular order:

One

Two

Three

Enter at your own risk.

*'Normal viewing distances' - there is a can of worms.

Best,
Helen

 

[chris_arnet]

the higher the number, the smaller the aperture is going to be. which means less light for your sensor, which is why you have longer shutter speeds. also the bigger the aperture (the smaller the number), the less DOF (depth of feild) your going to have, which is why its mostly blurry everywhere besides your focus point.

 

[Garbz]

With a plate in a large format camera you can easily get a razor sharp image at f/64. That's where your magazines stuff come from.

f/32 is pushing it for 35mm film IMO just like f/22 is on digital. It's doable but won't be razor sharp.

 

[ketan]

With a plate in a large format camera you can easily get a razor sharp image at f/64. That's where your magazines stuff come from.

f/32 is pushing it for 35mm film IMO just like f/22 is on digital. It's doable but won't be razor sharp.

Would it be too generalised if I say that if I am using a camera 30D which is not a large format then I should attempt to restrict to an apperture bigger than f-22 ? May be if I am going for a shot from distance (mountain, sea etc) may be I can afford to stop down... Possibly these questions can better be explained by self study and experiments only...
ketan

 

[Helen B]

Ketan,

If you are going for a distant shot then you may not need to stop down for depth of field. That is when knowing the lens' optimum aperture ('sweet spot') becomes worthwhile. You might also find that focusing on infinity gives better results than using the hyperfocal distance.

The plate camera that Garbz referred to would be similar to the 8x10 and 4x5 cameras that I gave the numerical examples for. Many of us use those formats, but very very few use plates (I don't). If you made a stitched image from multiple frames, your 30D would start to behave like a large format camera, but there would be other trade-offs. It's a minefield of a subject, but f/64 on an 8x10 camera may not give you very much depth of field - it might be comparable to about f/5.6 to f/8 on your camera (a rough estimate).

Yes, it is worth finding the aperture at which diffraction starts to cause a noticeable effect with your camera and your lens. It is likely to be well below f/22. As the rough calculations I gave above show, the Airy Disk (the smear caused by diffraction) is approximately the same size as one of the photosites ('pixels') in your camera at f/5.6 - but the lens may or may not have reached the diffraction limit at f/5.6, especially when close up - the diffraction limit is different at different focal distances.

Best,
Helen