Devising a test for lens clarity

[Ryogenetic]

I have several lenses now, and I am looking to devise a test for which has the best sharpness. I did some google searches and there does not seem to be a lot of information. Initial thoughts, are setting an subject a set distance from a tripod based camera. Setting the camera to a timed delay to minimize movement and interference from me. Then snapping a pictures starting at the lenses min f-stop up to F11. Hopefully this will capture any aberrations, and allow me to setup some quick reference cards. If you can think of other things I am missing....

Thanks
Ryo

Shooting a DMC-GF2 (M4/3) w/ 14-42mm f3.5-5.6, 25mm f1.4, 45-150mm f4, and a 45-200mm f4.

[Glenn NK]

Clarity, or resolution?

This is an ambitious undertaking. A few links that are interesting and might be helpful:

http://www.ehow.com/how_8048968_test...phic-lens.html

http://www.canonrumors.com/tech-arti...o-test-a-lens/

http://www.prime-junta.net/pont/How_...st_a_lens.html

http://www.normankoren.com/Tutorials/MTF5.html

To name a few.

Without a rigourous, repeatable, standard methodology, it can be a worthless exercise in frustration.

Richard (aka rpcrowe) may be able to shed some light on this.

Glenn

[Black Pearl]

Why?

[Geoff F]

There are a number of depth of field test charts but their effectiveness can be somewhat erratic.

However, they may pick up consistent under or over focusing and give an indication of the focus depth, but only at that one lens length.

A very simple test for sharpness is to photograph part of a printed magazine page. Mount it on a wall and use a tripod plus mirror lock up and cable release.

Zoom in on to an area which is part type and part photo. Compare the results on your computer screen at various zoom settings.

Keep repeating the results under different lighting conditions.

[Glenn NK]

A bit of clarification from the OP would help. The thread title mentions clarity, the text talks about sharpness, and depth of field has been mentioned.

Fortunately no one has mentioned bokeh or colour (oops).

Glenn

[Dave Humphries]

Fortunately no one mentioned all the other abberations either.

Hopefully this will capture any aberrations, and allow me to setup some quick reference cards.

Ah, so now we also perhaps know the intent; to identify all the lenses 'sweet spots'.

Hi Ryo,

To be honest, there's enough to worry about getting the composition, exposure, shutter speed and DoF right for the subject, I fear if you start down this road (even if someone did it all for you and handed you the quick reference cards tomorrow), you'll drive yourself nuts and quite probably attach too much importance to minimising distortions and spoil your images in some other way.

Just as people frightened of high iso stick to low numbers and end up binning countless subject movement blurred shots as a result.

It'll also, unless your into this kind of thing, probably spoil your enjoyment of photography.

Just get out there and shoot, you'll soon learn (unless you have too many lens), which ones exhibit which artefacts at which focal lengths and apertures.
A starting rule, like life in general, is to 'avoid the extremes' because design compromises will always be worse at either end of aperture and focal length ranges.

cheers,

[Ryogenetic]

Sorry for taking so long to respond, I was reading Glenn's links. (BTW Thanks Glenn) Part of what I had in mind was Flat Field Testing from your second link. I think Geoff was eluding to this with his test. Color shift and the smoothness of the Bokeh were designed into my OP test, though not mentioned. I was unsure what to use as a subject(for color range), and backdrop(that can challenge the lens's bokeh). Without having any real controls, I was thinking that shooting at my normal times, and normal conditions will allow me to best capture the info. I normally shoot from 9am to 3pm, clear to partially cloudy, or using Incandescence from garage lighting. At the end of the day I guess I am looking for the sweet spot/ weakness of my lenses.

Ryo

[Glenn NK]

At the end of the day I guess I am looking for the sweet spot/ weakness of my lenses.

Ryo

Ryo:

Now that is a good quest.

It's interesting that lens tests (available on the 'net for many lenses) find the "sweet spot", how shall I say it, almost by accident.

Let me explain.

When a lens is wide open, the entire glass area is utilized - and at the edges where the light rays have to be bent/refracted the most, is where a lens often exhibits softness, particularly at the edge of the frame. The inaccuracies of the refraction process are called refraction losses. No lens is perfect in this respect, but some are better or worse - this is part of lens design.

Stopping down eliminates the outer edges of the glass area where diffraction errors are larger, and the resolution is seen to improve.

Further stopping down, further reduces refraction errors, but they are starting to be offset by diffraction errors. Refraction is the phenomenon wherein light passing near the edge of an object is bent out of alignment. As the aperture gets smaller refraction errors get worse. You can google refraction for more detailed explanations.

Now, to the lens tests: the testers measure the resolution of the lens from wide open to fully stopped down, and the results are published - one example - for my own EF 100 macro (it applies to all makes of lenses tested):

http://www.photozone.de/canon-eos/16...review?start=1

Look under the heading MTF (resolution). It's best at f/5.6 and f/8, not bad wide open at f/2.8, but look what happens at f/32. It falls apart because of diffraction losses.

The simple testing at different apertures yields where the lens is sharpest and where it's at it's worst.

You could do the same thing with a sheet of newspaper, but the standard test charts provide an easier and standard method of quantifying the results.

If you can find the MTF results for your lenses, they will tell you where your lenses are best.

But truthfully, I seldom use them (read post by Dave Humphries), because composition, lighting, etc are more important.

Well I do use them, but just when I'm buying a new lens - they give me an indication of how good/bad the lens can be. I've bought three lenses this way (EF 100 macro, TSE 24 II, and EFS 17-55 - all Canon).

Glenn

[Shadowman]

The final output is also something to consider. Will these clarity/sharpness image tests be viewed on the monitor or as actual prints and if so how large?

[Colin Southern]

Why?

My thought exactly.

Ryo - please don't take offence at this, but having been through the same thought processes many years ago, I came to learn that exercises like this are an exercise in futility, and do NOTHING to improve one's photography. Far to many people "try to save the pixels" and end up "ruining the photo" in the process (a bit like the old "can't see the forest for the trees").

Couple of quick notes ...

1. You're trying to evaluate lens sharpness - and yet an optimal sharpening workflow during post-processing will have a far greater effect on the sharpness of an image than the difference between the native sharpness of any two lenses.

2. Normally the sweet spot (in terms of sharpness) is 2 to 3 stops down from maximum aperture - but - if you're always going to shoot at that aperture then you'll end up with some pretty bland images due to lack of DoF control.

3. The difference in colour between any two lesses will probably be far less than the inaccuracies in white balancing and the colour profile of whatever is doing your RAW conversion. If colour accuracy is important, get a colour passport and create a custom profile for each lens / light source.

Hope this helps.