acutance vs contrast

[Jed Freudenthal]

Just in case, to practice what I preach:
The Theory of the Photographic Process, 3rd Edition (Hardcover)
C. E. Kenneth Mees and T. H. James
This is the tome Photo Engineer is talking about. This is the '66 edition now in Amazon
http://www.amazon.com/Theory-Photog...=sr_1_3?ie=UTF8&s=books&qid=1269377490&sr=8-3

This is not really an introduction. It is too scientific for many people.

At the introductory level is someting like:
http://www.luminous-landscape.com/tutorials/understanding-series/understanding-mtf.shtml

and Norman Koren has written a good introduction.

Jed

 

[gainer]

Unless you are able to determine your own MTF curves from experimental data, you depend upon manufacturer's data. If you pay enough, you may get the curve for the lens you bought. Otherwisw, you depend on manufacturer's manufacturing tolerances.

 

[Jed Freudenthal]

Unless you are able to determine your own MTF curves from experimental data, you depend upon manufacturer's data. If you pay enough, you may get the curve for the lens you bought. Otherwisw, you depend on manufacturer's manufacturing tolerances.

I never paid anything for the MTF data. I got them for nothing from internet or from data sheets. I buy a lens on the basis of these data. If a manufacturer doesnot supply the data, I will not buy the lens. And the same happens when people come to me for advice what lens to buy.
And when I recall right, a popular photography magazine published quality data for lenses. These data were a popular variation on MTF. But, in reality, the purchase is based on MTF data.

Every product has its manufacturing tolerances. For some products, the tolerances are given and guaranteed.

Jed

 

[gainer]

The MTF is a Fourier Transform of an autocorrelation function. As such, it contains both amplitude and phase functions of frequency. In photography, we use spatial frequency rather than temporal, so the idea that there can be a phase shift between the input and output of a lens, that varies with spatial frequency, is sometimes overlooked. I once had a Kodak booklet dealing with movie film characteristics and processing that showed such phase shift, but have lost it. IIRC, the quality called "bokeh" depends on the nature of this phase shift. Two lenses with the same amplitude spectrum can have different bokeh.

 

[Jed Freudenthal]

The MTF is a Fourier Transform of an autocorrelation function. As such, it contains both amplitude and phase functions of frequency. In photography, we use spatial frequency rather than temporal, so the idea that there can be a phase shift between the input and output of a lens, that varies with spatial frequency, is sometimes overlooked. I once had a Kodak booklet dealing with movie film characteristics and processing that showed such phase shift, but have lost it. IIRC, the quality called "bokeh" depends on the nature of this phase shift. Two lenses with the same amplitude spectrum can have different bokeh.

It is impressive how much power may be hidden in the MTF. But for the moment, a simple introduction on the subject, is more practical.

Jed

 

[gainer]

If you examine the entire MTF, the amplitude and phase variations with spatial frequency form separate plots. Nothing is hidden, but the phase plot is frequently overlooked.

 

[Jed Freudenthal]

If you examine the entire MTF, the amplitude and phase variations with spatial frequency form separate plots. Nothing is hidden, but the phase plot is frequently overlooked.

You are quite right. It is not hidden scientifically seen. It is hidden because the phase variations are left out n introductory texts on MTF.
I think that the reason is that they have not much practical value for the average photographer. For a manufacturer , however, I can imagine it is a different story.

The "bokeh" for a film-developer, I like, is found by testing and look at the result.I do not see a relationship betwween the phase variations and my appreciation for a certain "bokeh".

Jed

 

[gainer]

Well, now, if you did the practical testing and compared the results with phase distortions, you might see a relationship. I have read that the bokeh does not always correlate with amplitude response.

 

[Jed Freudenthal]

Well, now, if you did the practical testing and compared the results with phase distortions, you might see a relationship. I have read that the bokeh does not always correlate with amplitude response.

You are right. And I have considered this. But, in my situation, with a modern lens like the Sironar-S 150/5,6, and a symmar-s I see most of the variability in the high frequency spatial range in the film-developer response. From the theoretical point of view I expect this in the amplitudo as well as frequency variations. The question is now how to get reliable MTF measurements for that high frequency range. And we ( the university of Utrecht and me) got stuck in finding a relationship between my bokeh preference and phase shift part of the MTF measurement.

Jed

 

[gainer]

I gather from what I have read that if the phase shift is directly proportional to frequency there will be no distortion of an electrical signal. Bokeh refers to out-of-focus behavior of a lens. It might show up in the phase response of an out-of-focus lens. I don't know if that has been tested, but if not, it might be worth doing.

 

[Jed Freudenthal]

I gather from what I have read that if the phase shift is directly proportional to frequency there will be no distortion of an electrical signal. Bokeh refers to out-of-focus behavior of a lens. It might show up in the phase response of an out-of-focus lens. I don't know if that has been tested, but if not, it might be worth doing.

First of all, I am talking on film and developer effects, rather than lenses. [ Using different, but similar lenses, the image quality is primarily determined by the selection of the film developer, as determined from experimental evidence].

In the high frequency range the theoretical data are hard to link to the psychological data (preference on bokeh). The main bottleneck are the psychological circumstances (has been told to me). Testing seems the best solution.

And this has led to testing with 6 films ( Fortepan 200, adox 50 ART, FP4+, Fuji Acros 100, Fomapan 200, adox 100 ART) in the size 4x5", and developed in the catachol developer, I have published on the APUG site. The negatives were printed on 16x20".
The result of the test is that I preferred the adox 50 ART on landscapes in southern Europe ( Italy, France, Spain). And the Fortepan 200 was preferred for landscapes in the Netherlands, UK .
And it was not only my preference, but the preference of othe people too. The preference is not always the same. For Fuji Acros some people like it, some not.

It is a practical test. We do not have an explanation. May be the quality of the light plays a role?

Jed