Small aperture sharpness limit

metalgear · 19th July 2005

Generally, stopping down a lens to a small aperture makes the eventual image sharper (and greater Depth of Field):

1. What is the best way to determine the F-stop for optimal sharpness of a lens?

2. On a digital cam body, is the CCD/CMOS sensor and its physical properties a limiting factor (e.g. Crop factor, pixel size and density) ?

LittleWolf · 19th July 2005

Originally Posted by metalgear

Generally, stopping down a lens to a small aperture makes the eventual image sharper

No, this is not true. Stopping down a lens reduces some abberations, but increases others (diffraction). The point of best sharpness depends on the specific lens. A "perfect" lens would actually be most sharp with the aperture wide open.

Quote:

2. On a digital cam body, is the CCD/CMOS sensor and its physical properties a limiting factor (e.g. Crop factor, pixel size and density) ?

Naturally, if details are smaller than the pixel size of the sensor, they cannot be resolved [1]. In practice, the sharpness is frequently more limited by the aberrations of the lens. It is not uncommon to see chromatic aberrations (colour fringes) of fairly decent lenses extend over several pixels.

[1] This is actually not entirely true, but the ways around this limitation are probably not relevant for practical photography.

Del_CtrlnoAlt · 19th July 2005

1. every lens have a sweet spot, some lens is f8, some f5.6, but also dependent on the make, the environment and also the camera... it does not mean that f32 or f22 is going to be the sharpest, it only let you have a better depth of field. if your focusing is off, no matter what aperture, its going to be blur.

2. actually the crop factor on DSLR, after i play with some primes, i notice that the 'distortion' is still there... meaning, if 35mm gives u this distortion on the film, on the DSLR it will also give u the same distortion, the crop factor just brings you closer to subject, thats all.. its not like if you have a 1.5crop, the 35mm will give you a 50mm range kinda view... its just bringing you closer to the distance of the 50mm... (anyway i might be wrong... cos i notice this when i taking macro...)

dRebelXT · 19th July 2005

What is the best aperture for my ES 28-135 IS?

singscott · 19th July 2005

There this thingy call MTF if you know how to use it will help in your quest of sharper picture.

Greater DOF does not mean sharper picture.

metalgear · 19th July 2005

Originally Posted by singscott

There this thingy call MTF if you know how to use it will help in your quest of sharper picture. Greater DOF does not mean sharper picture.

Was reading and learning about this, until I found out that some lenses are Design Standards rather than actual measurements. BTW are these MTF measurements a common industry standard for objectivity ?

Watcher · 19th July 2005

Originally Posted by metalgear

Generally, stopping down a lens to a small aperture makes the eventual image sharper (and greater Depth of Field):

1. What is the best way to determine the F-stop for optimal sharpness of a lens?

2. On a digital cam body, is the CCD/CMOS sensor and its physical properties a limiting factor (e.g. Crop factor, pixel size and density) ?

For 1, Like what singscott said, read the MTF graph. Please note that MTF is not a be all or end all of it but it is a great starting point

2) Yes, the pixel pitch is a limiting factor, as you step down the lens. If the pitch is too small, the diffraction limit affects the image earlier on a sensor with a smaller pitch

metalgear · 20th July 2005

was doing more research and found this program called imatest that claims the following:

You cannot measure a lens in isolation. It is a part of an imaging system that includes the camera's image sensor and RAW converter (which may sharpen the image), or film, scanner, and scanner software. Hence,

Measurements are relative. It's difficult to determine an absolute number for the lens alone. But you can compare lenses on similar cameras with great accuracy.
Camera and RAW converter settings, especially those that affect sharpening, noise reduction, and the tonal response curve, are critically important. Record them, and be consistent. If possible, use RAW files for best results..

It also gets interesting with the statement here - correct me if i'm wrong but i understand it to say that software sharpening can reduce differences between good and poor lens performance.

Remember, in evaluating lenses, use the results without standardized sharpening (the black curves and text). Results with standardized sharpening do, however, have some interest: they indicate what can be achieved after sharpening. But they tend to "flatten" differences between lenses

Given two lenses of similar MTF properties, do the figures not say anything else that could be material in lens choice ?

ratboy · 20th July 2005

when in doubt ... stop down 2 stops : )
( no scientific explanation )

Halfmoon · 20th July 2005

Originally Posted by singscott

There this thingy call MTF if you know how to use it will help in your quest of sharper picture. Greater DOF does not mean sharper picture.

MTF for Canon lenses is measured at f8..... so maybe shoot at f8-11??

singscott · 22nd July 2005

Wow so many question. I let me stress MTF chart is only a guide. Yes for canon they only measure the MTF on maxium aperture opening and F stop at F8. There are other MTF charts where they compare every F stops of the lens. Those where the days when camera magzines sure to have some MTF chart of some new lens to showcase.

The perfect lens with great sharpness will have graph lines in an all most straight lines on the top of it's MTF chart for most of it's F stops. So we don't have perfect lens. Then the sharpest image a lens can produce are often in it mid aperture setting, hence on most lens it the F8. By looking at the lens MTF chart. We can tell how good or sharp the lens is at that aperture or which apertures lens can produce sharpess image. When the lens aperture MTF chart can maintain a MTF graph almost in a straight line at a position higher then 0.8 or 85% of the MTF characteristic or line per mm. Then by comparing two MTF charts of two lenes with same focal length at the same aperture, you can see or guess which is sharper. Some manufacturer will showcase their lens MTF just look at their web site Lecia and Canon are some of those. I leave here with a link that you might find useful http://www.luminous-landscape.com/tu...ding-mtf.shtml

dreamtheatre · 22nd July 2005

Originally Posted by ratboy

when in doubt ... stop down 2 stops : )
( no scientific explanation )

: ) nearly duplicated this suggestion.......this is what I always do....that's why fast lenses are good...not becasue to use them at maximuinm aperture, but the sweet sopt will be still fast enough.....

mpenza · 22nd July 2005

Originally Posted by Del_CtrlnoAlt

2. actually the crop factor on DSLR, after i play with some primes, i notice that the 'distortion' is still there... meaning, if 35mm gives u this distortion on the film, on the DSLR it will also give u the same distortion, the crop factor just brings you closer to subject, thats all.. its not like if you have a 1.5crop, the 35mm will give you a 50mm range kinda view... its just bringing you closer to the distance of the 50mm... (anyway i might be wrong... cos i notice this when i taking macro...)

The "crop" means that the centre part of the lens is used. And any distortion that is still present in that centre part of the lens would still turn out in the pic. If the distortion is only at the edges of the lens, using a "crop" DSLR will mean that the distortion will not turn out in the pic captured.

+evenstar · 23rd July 2005

"1.6x /1.3x crop is only like doing digital zoom"---Quote by SAFRA Photographer of the Year

NMSS_2 · 23rd July 2005

"1.6x /1.3x crop is only like doing digital zoom"---Quote by SAFRA Photographer of the Year

that is assuming if the lens is not meant for digital camera.

digital lens should be of better quality than normal film lens, else all the images produced might have some softness to it.

EiRiK · 24th July 2005

Originally Posted by metalgear

Generally, stopping down a lens to a small aperture makes the eventual image sharper (and greater Depth of Field)

only half correct. stopping down will only produce greater depth of field, not increased sharpness. mid-apertures (eg. f/8) will give best sharpness.

ricohflex · 24th July 2005

Have read a complicated treatise of depth of field.
Boil it down to an everyday guide : from the max apeture of your lens , stop down 3 apertures.
towards the last two smallest apertures, too much diffraction.
pic quality goes down. according to measurements. maybe our eyes cannot see diff.
There is a f/64 club. they shoot at small apertures. Large Format fans.

Must check carefully the spec and review of your lens, especially if it is a special and costly lens.
Some brands bother to design / make the lens so that it is best at the largest aperture and within a certain focusing distance range.

e.g. have read Leica Macro 60mm f 2.8 is designed to function excellently at f2.8 at the macro distance.

Similarly, the Canon 200 F1.8L is excellent at widest aperture F1.8 at about (say) 20 feet (6m).

It makes sense because that's what you pay a lot extra for.

19th July 2005	#1
metalgear Member Join Date: Aug 2003 Posts: 590	Small aperture sharpness limit Generally, stopping down a lens to a small aperture makes the eventual image sharper (and greater Depth of Field): 1. What is the best way to determine the F-stop for optimal sharpness of a lens? 2. On a digital cam body, is the CCD/CMOS sensor and its physical properties a limiting factor (e.g. Crop factor, pixel size and density) ?

19th July 2005	#3
Del_CtrlnoAlt Senior Member Join Date: Feb 2003 Location: Outside the Dry Box. Posts: 16,342	1. every lens have a sweet spot, some lens is f8, some f5.6, but also dependent on the make, the environment and also the camera... it does not mean that f32 or f22 is going to be the sharpest, it only let you have a better depth of field. if your focusing is off, no matter what aperture, its going to be blur. 2. actually the crop factor on DSLR, after i play with some primes, i notice that the 'distortion' is still there... meaning, if 35mm gives u this distortion on the film, on the DSLR it will also give u the same distortion, the crop factor just brings you closer to subject, thats all.. its not like if you have a 1.5crop, the 35mm will give you a 50mm range kinda view... its just bringing you closer to the distance of the 50mm... (anyway i might be wrong... cos i notice this when i taking macro...) __________________ Logging Off Permanently. Those who need to contact me will know where to contact me.

19th July 2005	#4
dRebelXT Member Join Date: May 2005 Posts: 1,563	What is the best aperture for my ES 28-135 IS? __________________ My photog website! :) I am AF - A Fool..

20th July 2005	#8
metalgear Member Join Date: Aug 2003 Posts: 590	was doing more research and found this program called imatest that claims the following: You cannot measure a lens in isolation. It is a part of an imaging system that includes the camera's image sensor and RAW converter (which may sharpen the image), or film, scanner, and scanner software. Hence, Measurements are relative. It's difficult to determine an absolute number for the lens alone. But you can compare lenses on similar cameras with great accuracy. Camera and RAW converter settings, especially those that affect sharpening, noise reduction, and the tonal response curve, are critically important. Record them, and be consistent. If possible, use RAW files for best results.. It also gets interesting with the statement here - correct me if i'm wrong but i understand it to say that software sharpening can reduce differences between good and poor lens performance. Remember, in evaluating lenses, use the results without standardized sharpening (the black curves and text). Results with standardized sharpening do, however, have some interest: they indicate what can be achieved after sharpening. But they tend to "flatten" differences between lenses Given two lenses of similar MTF properties, do the figures not say anything else that could be material in lens choice ?

23rd July 2005	#14
+evenstar Senior Member Join Date: Jan 2004 Location: Singapore Posts: 5,453	"1.6x /1.3x crop is only like doing digital zoom"---Quote by SAFRA Photographer of the Year __________________ the SEARCH FUNCTION is there for a reason. USE IT!

19th July 2005	#5
singscott Member Join Date: Aug 2004 Location: www.whltelightphotographer.com Posts: 1,659	There this thingy call MTF if you know how to use it will help in your quest of sharper picture. Greater DOF does not mean sharper picture.

20th July 2005	#9
ratboy Member Join Date: Sep 2002 Posts: 702	when in doubt ... stop down 2 stops : ) ( no scientific explanation )

22nd July 2005	#11
singscott Member Join Date: Aug 2004 Location: www.whltelightphotographer.com Posts: 1,659	Wow so many question. I let me stress MTF chart is only a guide. Yes for canon they only measure the MTF on maxium aperture opening and F stop at F8. There are other MTF charts where they compare every F stops of the lens. Those where the days when camera magzines sure to have some MTF chart of some new lens to showcase. The perfect lens with great sharpness will have graph lines in an all most straight lines on the top of it's MTF chart for most of it's F stops. So we don't have perfect lens. Then the sharpest image a lens can produce are often in it mid aperture setting, hence on most lens it the F8. By looking at the lens MTF chart. We can tell how good or sharp the lens is at that aperture or which apertures lens can produce sharpess image. When the lens aperture MTF chart can maintain a MTF graph almost in a straight line at a position higher then 0.8 or 85% of the MTF characteristic or line per mm. Then by comparing two MTF charts of two lenes with same focal length at the same aperture, you can see or guess which is sharper. Some manufacturer will showcase their lens MTF just look at their web site Lecia and Canon are some of those. I leave here with a link that you might find useful http://www.luminous-landscape.com/tu...ding-mtf.shtml

23rd July 2005	#15
NMSS_2 Senior Member Join Date: Apr 2003 Location: Northen part of Sing a pore Posts: 2,010	"1.6x /1.3x crop is only like doing digital zoom"---Quote by SAFRA Photographer of the Year that is assuming if the lens is not meant for digital camera. digital lens should be of better quality than normal film lens, else all the images produced might have some softness to it. __________________ My Wordpress Blog Flickr

24th July 2005	#17
ricohflex Senior Member Join Date: Feb 2005 Location: sing Posts: 1,540	iris Have read a complicated treatise of depth of field. Boil it down to an everyday guide : from the max apeture of your lens , stop down 3 apertures. towards the last two smallest apertures, too much diffraction. pic quality goes down. according to measurements. maybe our eyes cannot see diff. There is a f/64 club. they shoot at small apertures. Large Format fans. Must check carefully the spec and review of your lens, especially if it is a special and costly lens. Some brands bother to design / make the lens so that it is best at the largest aperture and within a certain focusing distance range. e.g. have read Leica Macro 60mm f 2.8 is designed to function excellently at f2.8 at the macro distance. Similarly, the Canon 200 F1.8L is excellent at widest aperture F1.8 at about (say) 20 feet (6m). It makes sense because that's what you pay a lot extra for.