Olympus 17/1.8 review

Started Nov 17, 2012 | Discussions thread
noirdesir Forum Pro • Posts: 13,401
Re: Macrocontrast vs. Microcontrast

Anders W wrote:

noirdesir wrote:

Anders W wrote:

Didn't you mean to say somewhere between the 100% and the 50% mark here rather than between 0 and 50%? If yes, then I am all with you.

I think there might be different definitions of microcontrast, as you said that "at least some people use it in reference to contrast at fairly low frequencies rather than very high ones", thus it could be contrast at resolutions between the 0 and 50% mark (but closer to the 50% mark) or it could be the contrast a bit above that.

Yes, but I was thinking about your example with the hypothetical MTF curves and Zeiss versus Leica. In that example, the "Zeiss" (with a presumed focus on microcontrast) beats the "Leica" (with a presumed focus on resolution) at contrast criteria higher than 50 percent whereas it is the other way around at contrast criteria lower than 50 percent.

Ah, I see where the misunderstanding is: when I said below 50% I meant to the left of the MTF50 resolution. Between 0 target resolution and a target resolution corresponding to the 50% contrast mark.

One further question: Would you agree with my assessment that, as far as Ming's images let us judge, the 20 has a better MTF curve than the new 17/1.8? I must admit, I am a bit disappointed at the showing of the 17/1.8 here. I hadn't really expected it to beat the 20 in this regard but I had hoped/expected that it would do about as well. If that had been the case, I would probably have been ready to exchange my 20 for the 17/1.8. Now, I am not so sure.

I'm not sure, my general impression from his review is that the two lenses are relatively close (with a few notable differences like LoCA and vignetting that favour the 17 mm), and closer than I though they would be. But evaluating MTF from a few crops is something I am not really comfortable with, I'd put more weight on his overall comments which I would assume to be based on a lot more (test) images than the ones shown.

OK. Then we simply have quite different perceptions. To my eyes, there is a very clear difference in contrast in favor of the 20 not only at very low frequencies (macro contrast) but also at higher ones. Look at the crops from the bottom-right corner, for example. The individual lines in the grid below the window is clearly visible in the shot from the 20, even wide open, whereas they nearly disappear into a gray patch in the corresponding shot from the 17/1.8. Similarly, the joints (lines) between the concrete blocs in the same corner crops are clearly visible in the shot from the 20 but hardly visible at all in the shot from the 17/1.8.

You see, I simply have very little confidence in a single test image being relevant and accurate. Show me 50 test images taken under varying conditions (and with the greatest care taken to equalise exposure, focus and processing not just for each image globally but also locally for each position in the image) and I might start to have some confidence in the conclusion drawn from them.

For example, if a lens shows vignetting, any crop from the corners will have received a different dynamic range compression during the raw conversion depending on the amount of vignetting making a contrast evaluation on the final image null and void.

Post (hide subjects) Posted by
(unknown member)
(unknown member)
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow