Micro-Contrast Matters


Today I came across an excellent article discussing micro-contrast by Yannick Khong. In it he discusses how sharpness is being over-emphasized by buyers and I couldn’t agree more. For a short period of time I was really wrapped up in technical reviews; before I started demoing and reviewing equipment on my own.

I’m a technical person at heart, but gaining access to a large amount of equipment caused me to form an appreciation for the intangible characteristics associated with how a lens draws a subject. Now, when I shop for a lens that I personally want to purchase and do not have time to demo, I look for examples of other people’s photos on sites like 500px to see what photographers have created with it, or for non-technical reviews that have photos with minimal editing, so I can see what I will have to work with. From there, if I’m sufficiently impressed, I’ll read about the build quality of the lens and any auto focus characteristics that maybe important.


Sadly, as Yannick Khong addresses, many manufactures are responding to the metric driven buying habits of consumers by producing lenses that are ultra-sharp and score well on DXO, but have very poor micro-contrast. This is a shame, because micro-contrast is largely where things like the Leica/Zeiss/Fuji look comes from, and consumers haven’t taken the time to appreciate its importance because it is difficult, if not impossible, to quantify. Relying too heavily on objective values provided by DXO mark scores or resolution charts, instead of your own subjective opinion of many examples of how a lens draws, might cause you to make an expensive mistake.

This entry was posted in Fuji lenses and tagged . Bookmark the permalink. Trackbacks are closed, but you can post a comment.
  • Foot of our stairs

    Im not convinced.

    • Jeffry De Meyer

      Click on blog there he does a comparison of an old Nikor to the otus.
      It is really clear in that one.

      To me this phenomenon became really clear when I bought al old 300mm and for some reason I didn’t have to touch my vibrancy slider to get colours to pop like I always have to when using modern glass

      • Foot of our stairs

        I think he has cherry picked the examples that support his point, I don’t think he makes a technical point with much scientific rigour, a lot of images made under different lighting, the Otus comparison example, with the Nikon, you also notice that the lady he photographs, switches from a smile which flattens the contours of her face and the angle of the face has changed subtly such the lilting is creating less glare and is falling flatter on her face creating less deep shadows and modelling.
        Nevertheless he has some moderately compelling evidence, at the very least you can conclude the perpetual quality of the older Nikon glass does not match the measurable results seen in allot of controlled testing in these examples, if this is down to “Micro contrast”I’m not sure it seem a term poorly defined

  • Jorge

    Isn’t that the hard rock cafe in Phily?

  • nwcs

    I read the blog article and some of his earlier (related) articles. I’m not convinced, but in a different way. Microcontrast is relatively easy to describe: it’s the ability of the lens to transmit subtle color tones. That is, it has a better contrast rendition. It can transmit 4 shades of red, for example, instead of smashing it to one shade by the time it gets to the sensor.

    His comparison photos are not very good. Most of the “do you see its” are really figure-ground visual perception issues. He sees it but I do not. But if I change my perception I can see it and then suddenly not. Just like faces/vases. He also isn’t convincing that mainly old film lenses have this “ability” and that most modern lenses are clinical. It doesn’t pass the smell test especially when he waxes poetic about how ED type glass removes binocular seeing and turns a lens into a monocular lens (all lenses are monocular!).

    That said, he misses a very important aspect of microcontrast and its elusiveness: the sensor. The bayer, x-trans, and similar methods of interpolation have a severe impact on microcontrast. Low-pass filters also play a small role in killing it. That’s one reason why some people love the Merrill style Foveon sensors even above the Quattro style. It’s full RGB 1:1:1 at every pixel. The quality of information is superb and gives the appearance of high microcontrast. The downside of the technology is that it falls apart very quickly after ISO 200 and it is very slow to process.

    Good post for giving me something to think about.

    • I think part of what makes an article good is how it changes or adds to your perception of an issue. I have found it very hard to differentiate between the technical side of photography and the art side of photography over the years, but I find my method of lens selection doesn’t fail me when making an expensive decision. I have purchased many new/old Leica lenses without regret, but in the past when I shot Nikon and Canon I would make my purchases based on technical reviews and almost always came away disappointed.

      With his examples I do notice some differentiation, but it took me a while to understand what the author was getting at, but it would have been more affective if he was more technical in his approach. Further, English on the internet has become a semi conversational language that lacks some of the nuance necessary to relay complex ideas. I for one do my best to word things well, but I often reread what I have written and think I could have found a way to have written it better, but sometimes you just have to move on…

      • nwcs

        You are right in that it is hard to handle both the technical and artistic side of photography. I’ve seen it in my other hobby, numismatics. When it comes to evaluating the condition of a coin it’s part technical and part aesthetic and the aesthetic part is always controversial and a source of a lot of contention.

        I think his writing was fine for me and I think I know what he was getting after but his images just didn’t seem to back it up. I’m not entirely sure he’s correct. For example, on one of his charts (in a later blog post) he states how the 180mm f2.8 ED lens (Nikon) is one that displays this characteristic. It’s more of the 3D and less technically/optically corrected. The thing is, it’s probably one of the most frequently used lenses in the astrophotography world because of its optical corrections, low coma, and reasonably controlled CA. But on his charts he puts astro lenses in the (what I’ll call his) no soul section.

        • I wrote him directly and I think his later articles are based on a dialog with a variety of people. Maybe he lost focus.

          • nwcs

            Pun intended? lol

  • So. . . that’s 20 minutes I’m not getting back. I just can’t see it. I’d love to see a double-blind test to see if people can really pick this out or it’s just confirmation bias.

    • ShawnM

      I lost more than 20 minutes.. He seems so obsessed with microcontrast that he confused it with distortion, CA, contrast, etc. Side by side blind test is really the only way to convince others.

      • nwcs

        While I’m waiting for the link to be approved he does attempt to do that with his blog post “Depth vs. Flat Lens Quick Comparison” but it fails IMO to demonstrate anything meaningful.

  • Hardcore_Fanboy

    I have read about this matter from few authors… but all that they could provided was like in these articles – captivating theory, but 0 evidence… I looked at those pictures for minutes, pixelpeeped, zoomed out, zoomed in and what not – I just CAN NOT see it…. I have both – sigma 50 ART and Voigtlander 58 f1.4 (what is highly praised by “new lens hater” crowd) and I just can not see it – how the new lenses are bad and microcotrasts or that mystical 3d pop… I can see that new lens is supersharp at f1.4, there fore I can use whenever I want and get 3d effect with bokeh where with old lens I would need to stop down for acceptable resolution and loose some of bokeh, but old lens advantages – there is none…nor in my pictures nor in these “tests”… I want to believe, but there is nothing out there 🙁 (funny that this old lense loving corwd ridicule pixelpeeping, megapixel wars, and sharpness… and yet “evidence” they provide for their arguments need to be pixelpeeped even on a deeper level and yet can not bee seen)

  • TheoBenschop

    The author of this inspiring article has a good point!

  • TheoBenschop

    PS I’m very, very, very happy with my Fujifilm lenses on a x-m1 body, on which body I’m able to push a little bottum!

    But after a research on the intermet of the internetindustry with aspirines I’ve understand that it’s going very, very, very quick in the dynamic photo-industry,

    where I’ve just on the internet find out that the rather expensive Fujifilm lenses cant’t be used properly on other brands for instance if you want a better video quality, but this will of course not say that I’m not very, very, very happy with my Fujifilm lenses!

  • TheoBenschop

    PS So before everybody is really addicted to this digital nonsens of the digital industries it is in the here and now always good to know that the world has done it for millions of years without this digital nonsens, for instance in the times when the not digital people came home from a nice not digital vacation, when they could for instance look with their own not digital eyes on a not digital screen of a not digital tv that not digital people where walking on the not digital moon, so the world will do it another millions of years without this digital nonsens!

    So that one wunders all the time when are the digital industries inventing just a normal private affordable not digital aeroplane for everybody, where the not digital people are willing to work for in not digital offices and other not digital places?

  • TheoBenschop

    PS So I’m just saying that the ones who don’t agree with mister Louis Ferreira

    – albeit that he is a technical person at heart, what of course for mister Louis Ferreira a little problem is, what he surely with his interest in lenses of Fujifilm will overcome –

    must be happy with the zombie results of the nonsens of the digital industries, which nonsens of these digital industries should of course have been stopped by the people in the time of programs like Word Perfect:

    thank you very much, but this is for the next 5000 years more than enough digital stuff!

    Which nonsens of these digital industries is in the here and now to be seen in the streets where the people are very busy with their socalled “smart” digital phone, which socalled “smart” digital phone is produced in a country where the people dont’t have the opportunity to be all day busy with a socalled “smart” digital phone, and where the people earn one dollar a day.

    But as long everybody is happy with one thing or the other, the world keeps on turning, so even with the nonsens of the digital industries!

  • Narretz

    Take this sentence from the article:
    > M43 lenses have awful to very good micro-contrast but it is harder to identify which ones since they are still figuring out ways to exploit that attribute at the sensor-level (they might have reached solution in the GX8).

    How can you take it serious after that? He raises more questions than he answers and openly contradicts his own theory (micro contrast is a property of the lens). What does the sensor have to do with it? Shouldn’t you be able to identify the micro contrast independently from that?

    • Chai Nagappa

      From my experience the sensor does have an impact on the micro contrast. Many in the Fuji X community prefer the rendering of the skin tones in the Bayer sensor (i.e., X10, X100, X-A1, X-A2) versus the X-Trans sensor (i.e, X20, X30, X100S, X-M1, etc.). There is a difference in micro contrast rendering between versions of the T-Trans sensor. Many prefer skin tone rendering of the X-E1/X-Pro1 over the X-E2/X-T1. Having said that, in my experience, the lens is the most important component.