Bokeh comparison between Sigma, Otus and Nikkor 58mm

Started 6 months ago | Discussions thread
TQGroup
Contributing MemberPosts: 791Gear list
Like?
Re: Bokeh comparison between Sigma, Otus and Nikkor 58mm
In reply to Fave Photog, 6 months ago

Fave Photog wrote:

TQGroup wrote:

David Whysong wrote:

TQGroup wrote:

David, your gear list shows you own some incredibly fine glass... surely you can see more in an image than just "sharpness"?

Perhaps you could help me understand your opinion better by precisely explaining what you mean when you say "the lens is not sharp"?

Thank you

Andrew

You were looking at my gear *wishlist*. I have a Nikon 180 AF, 85 1.4D, 85 1.8G, and a Sigma 35 Art. Those also qualify as nice glass, but it's not what you saw in the gear list.

Sharpness isn't everything one might want from a lens, but a sufficiently sharp focal plane is the most fundamental requirement of a lens.

Down-sampling is nothing more than throwing away resolution that the lens failed to deliver.

I love some of the pictures I've seen posted here from the 58, but I don't think its image quality is any better than the old Sigma 50 which costs about 4x less, achieves better sharpness, and provides similarly excellent bokeh.

I see a lot of people making claims related to bokeh quality of the Sigma vs. Otus vs. 58mm, but it's almost impossible to properly judge without a side-by-side comparison. Here we have one, and while there is a visible difference in bokeh, none appear perfect in this example. (On further inspection, I do see differences between the Zeiss and Sigma; I'd say the Sigma looks a bit better but it's a close call.) However the 58mm is noticeably unsharp even in this very low-resolution image!

David

I agree with a lot of what you write but I do see a dichotomy.

You very rightly say; "Sharpness isn't everything one might want from a lens, but a sufficiently sharp focal plane is the most fundamental requirement of a lens".

Then, your last sentence says; "However the 58mm is noticeably unsharp even in this very low-resolution image!"

There is no dichotomy in David's position. Where is the dichotomy in 'A lens should be sufficiently sharp at its focus point, however, the 58/1.4 is not sharp anywhere in the image at 1.4 or even at 2.0'?

Firstly, thank you for taking the considerable time and trouble to pull my arguments apart! As you will see, I believe, we are both "essentially right" but we are looking from two different and legitimate points of view...

To me the dichotomy is in ""Sharpness isn't everything one might want from a lens, but a sufficiently sharp focal plane is the most fundamental requirement of a lens".

The problem is the definition of a "sufficiently sharp focussed plane" (I prefer focussed plane to focal plane) for a specific image. Photography to me is about producing images that meet "requirements" at the very least and preferably exceed them in pleasing ways (in the eyes of the beholder). Therefore, a lens that at a particular F-stop that produces a pleasing "unsharp" blur can both meet and exceed requirements for that image. I think it is fair to say that the Nikon 58 was more intended for photography of people than the other kind of "heavenly objects" where the requirements for sharpness are obviously totally different!

In my prior post I wrote: "Perhaps you could help me understand your opinion better by precisely explaining what you mean when you say "the lens is not sharp"?"

Can you see my challenge? It is how to accurately interpret your concept of "sufficiently* sharp focal plane"! (* my underscoring and italicization)

It's safe to say that blurry (as seen in the 58/1.4 examples) is not "sufficiently sharp" by anyone's standard.

Sometimes, appropriate "blur" is very good and a lens that can produce smooth, gentle and subtle blur is better than a very sharp lens that has to be manually "backed-off" to produce the desired "blur" effect. many of the best photography taken are not "clinically: sharp by intention.

To me, the concept of "lens sharpness" is totally subjective

Lens sharpness is not a "concept" and there is nothing "subjective" about it, at least from a 'user' point of view. It's a reality of Physics determined by the lens' design and its quality of manufacture, and is easily measured by tests. Differences in sharpness are easily observable by the human eye.

If we can accept that a "user" may have different "outcomes", then we can both be right. I agree about the reality of our current and ever-changing knowledge of Physics. May I also point out that in my own lifetimes of an avid interest in Astronomy, many basic concepts have changed; eg where you a "steady-stater or a "big-bang" supporter in the 1960's?

Of course differences in sharpness, at a certain level, can be observed by the human eye. That is precisely why one often shoots less than perfect human skin in a deliberately "unsharp" way and why lenses that can do this are highly sought after by the cognoscenti.

and totally dependent on the situation being shot

Perceived lens sharpness is very dependent upon the lens' ability to discriminate between pure black and pure white (contrast), so any lighting situation (strobes, Sun, street lights, etc.) that would cause flare or lack of contrast will reduce perceived sharpness. All lenses suffer from flare and lack of contrast in difficult lighting situations to one degree or another. Some handle it much better than others.

Again, that's a function of the lens' design and is a trade-off determined by the stated goals of the lens designers and constrained by the lens' cost.

Which is why lenses are tested using uniform testing methods under ideal lighting in order to determine the sharpness of a lens. That way 'Lens A' can confidently and accurately be said to be sharper than 'Lens B'.

Situations that cause lens flare or other adverse lighting conditions are rare in real life situations, and can usually be avoided or lessened for the most part.

Amazingly, night shots of citiscapes, sunsets, contre jour photography, etc, etc are hardly rare in real life situations. Quite the opposite, in fact. Many "photographers" rather than "optics experts" know that the "best light" is often around dawn and dusk. Most of my best and most prized photos were taken in lighting conditions that you would suggest "can usually be avoided or lessened for the most part".

I buy my lenses for the real world not a theoretical optimum laboratory test world!

Further, these "chart tests" are over a nominal fixed distance that may or may not accurately predict the performance of a lens over a different distance. Bring on standardised optical benches!

So, if all lenses are tested under "uniform" conditions; why do (a) the different major camera companies produce MTF charts that are different and cannot be directly compared? And, (b) why do some "chart" testing sites reveal precisely their testing methodology and others shroud it in secrecy?

It is quite different to a lens "potential sharpness".

Lens sharpness is lens sharpness. The sharpness of a lens doesn't change because the user doesn't employ proper technique. Camera movement or vibration from the mirror is user error in producing a usable image and has nothing to do with the sharpness of a lens.

There's only one way to determine a lens' sharpness and that is to employ proper technique, yet you bemoan such proper technique (technique that amateurs strive to employ) below as being nothing but trickery and deceit utilized by testing sites to fool the new and naive.

In the real world of candid photography, street photography, sports photography, wedding / event photography, tripods and even flash are not possible and / or feasible. That is why VR and AF can be so handy. So, in real life situations where many photographers ply their trade, a lens with a decidedly superior MTF chart "sharpness: score may fail miserably against a much lower ranked competitor with superior low light performance achieved by a faster F-stop potential and very fast AF, juyst as one example.

The concept of "potential lens sharpness" in a laboratory setting under controlled conditions and at a distance of 40 X the focal length, etc, etc is of academic interest for the most part as virtually all good quality modern lenses are sharp enough for all but the most demanding assignments. And those who get those assignments know what they need and why they need it.

In fact, I am starting to believe that MTF, as in MTF Chart, is really an abbreviation of Meaningless, Titillating and FUDing

Meaningless: MTF charts are shot of a flat subject,

That's because the only way to properly assess field curvature or to determine if a lens has a de-centering issue is to shoot a flat subject.

Are you a photographer? If so, approximately what percentage of your photos are taken of "flat" subjects?

usually black & white

That's because the only way to accurately determine a lens' ability to discriminate contrast (the difference between pure black and pure white) is by shooting a black and white subject.

And what percentage of photographs, these days, are taken in black and white? Is it not possible that lenses tested in black and white might actually render differently when shot in natural colour?

over distances of 40 X focal length or so

That's because there needs to be a standard distance at which lenses are tested, and because all lenses should perform optimally at that distance, even macros.

I am amazed! I bought my Nikon 105 macro to shoot up close, real close, that is its purpose... yet you are suggesting that it should be tested for sharpness on subjects around 4 metres away? This does beggar belief!

from heavy tripods with remote releases,

Perhaps you would prefer the camera and lens be hand held by a 6 year-old while walking? The sharpest lens in the world would appear dismally blurry if that were the case.

... or a 64 year old while walking in a crowded market, shooting from the hip? Yes, absolutely, that is how I shoot street and the lenses I choose perform admirably. Strangely, these lenses usually do not top "chart" test scores....for example, the new Nikon 35 F1.8 G ED is superb in this situation while the nominally "sharper" bigger, heavier Sigma 35 ART is a distant second in this application.

BTW, I do own at least 6 tripods and use them whenever possible / appropriate...

mirror-up for 4 seconds under controlled and even lighting conditions, etc, etc.

I know! Imagine the gall of testing sites to employ such ridiculous methods when testing a lens for sharpness!

Honestly, now who actually takes pictures like that in real life!

Those who care enough about their photography to shell out $950 (Sigma 50/1.4), $1,700 (Nikon 58/1.4), or $4,000 (Zeiss Otus). One would hope, anyway.

Actually, lens testers and pixel-peepers, that who!

I have actually owned the Nikon 58/F1.4G for nearly 3 months and have yet to mount it on a tripod... silly me! But, that hasn't stopped it from rendering some great "candids" of my friends...

My real concern about a "fixation" on MTF charts is that, not only can they be (are?) misleading and potential be "gamed" by manufacturers, they take away the focus from taking "beautiful images", instead focusing on a culture of "my lens is sharper than yours; na na na na na na!"

Sufficient "sharpness" for the intended purpose is a basic, a given, it is so obvious that it is not even necessary to be stated. "Sharpness" over and above the necessary level required for an intended purpose can and does detract from other aspects of photographic technique and, by itself, does not enhance an image when viewed in conventional ways.

In some pursuits, critical sharpness will "never" be enough and forever increasing resolving power is the "holy grail", but let me assure you that, in most cases, that does not apply to people photography!

Once again, thanks for posting!

Cheers

Andrew

-- hide signature --

The Five 'Ps' of Photography:
*Proper Planning Prevents Poor Performance!*

 TQGroup's gear list:TQGroup's gear list
Nikon D600 Nikon D810 Nikon AF-S Nikkor 16-35mm f/4G ED VR Nikon AF-S Nikkor 300mm f/4D ED-IF Nikon AF-S Nikkor 24-70mm f/2.8G ED +18 more
Reply   Reply with quote   Complain
Post (hide subjects)Posted by
(unknown member)
(unknown member)
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow