No Cold Water

I'll just point out, following yesterday's discussion, that it has been my observation, over many years, that any time you start talking about aperture, in any context, the discussion eventually devolves into squabbling about the science and the math. It virtually never fails. But one thing I will toss out there is not to forget about the degree of precision in the device, which is a wild card we don't know. Yes, ƒ/1.2 might not be exactly one-third or exactly one-half of a stop different than ƒ/1.4, which itself might not be exactly ƒ/1.4. But your particular camera might not be setting ƒ/1.2 as the maker intended when it's set to ƒ/1.2, either.

It doesn't really matter—the general points still stand.

Here is the accepted f-stop scale for camera lenses, in third stops, with whole stops in bold red:

0.7, 0.8., 0.9, 1, 1.1, 1.2, 1.4, 1.6, 1.8, 2, 2.2, 2.5, 2.8, 3.2, 3.5, 4, 4.5, 5, 5.6, 6.3, 7.1, 8, 9, 10, 11, 13, 14, 16, 18, 20, 22, 25, 29, 32, 36, 40, 45, 51, 57, 64, 72, 80, 90

You'll notice it's only an approximate geometric sequence. And f-number is a dimensionless number that already takes into account the size of the entrance pupil. Each whole stop represents an approximate doubling or a halving of the amount of light.

Anyway, photography is just not that precise such that the picayune differences are worth arguing about. First of all, exposure is a judgement call—ask ten people to adjust overall exposure in any picture with sliders and you'll get seven different renderings, perhaps with two or three quite far off from each other. Secondly, the differences we're talking about represent relatively minor corrections in software. They also represented relatively minor corrections in the darkroom. Whether you are in the Third-Stop camp or on Team Half-Stop with regard to ƒ/1.2 lenses, the differences are slight and easy to compensate for.

Michael Fewster wrote: "Once upon a time in film days, superfast lenses had a point. High ISO speeds meant ISO 800. Even more relevant was the need for fast lenses to nail accurate manual focus. As cameras with internal light meters closed down apertures, manual focus became difficult. I think it was Minolta that first came up with focus at full aperture and then auto stop down to the required aperture when exposing. The brighter the viewfinder, the easier it was to manually fast focus. There was real street cred in having a fast lens on the front of your SLR. I don't see the point of such lenses on DSLRs."

An utterly unnecessary lens might be useful to calibrate your sense of optical perfection

I would defend fast-lens fans more than that—if people like their fast lenses, far be it from me to splash any cold water on their happiness. (I have plenty of my own enthusiasms and tastes.) I sympathized with what Zed said, that he bought his ƒ/1.2 lens "in part for the prestige of owning just one such utterly unnecessary lens." You go, bro—I get that. And of course maximum aperture is only one feature of any lens. It's the whole lens one has to look at. I'd say, if you like fast lenses, go for it, and if you don't, don't worry about it. All I'll add is that personally, just for my own self, I'd prefer it if the Fuji 56mm had the specs of the Viltrox 56mm, and the Viltrox had the specs of the Fuji. Moderation in all things.

Back at the ranch, ever the empiricist, I got out the Fuji 56mm ƒ/1.2 yesterday, and set the camera to square, and went out photographing. I even discovered that my X-H1 has eye-recognition focus, just like a Sony! I did not know that.

(I also did not know that I could set my Mac to Dvorak with the simple click of a toolbar button. But that's a different subject. Off to typing practice now. Enjoy your Saturday! All best from the beautiful Finger Lakes region of Upstate New York, where it will be 55°F today and I will be walking on the hill come evening).

Mike

Book o' the Week:

ArtCurious: Stories of the Unexpected, Slightly Odd, and Strangely Wonderful in Art History by Jennifer Dasal (Penguin, 2020. Adapted from the popular podcast.)

The link above is a portal to Amazon, through which most anything you purchase will be credited to TOP. As an Amazon Associate I earn from qualifying purchases.

Original contents copyright 2021 by Michael C. Johnston and/or the bylined author. All Rights Reserved. Links in this post may be to our affiliates; sales through affiliate links may benefit this site.

(To see all the comments, click on the "Comments" link below.)
Featured Comments from:

James Bullard: "The f-numbers are a ratio of the diameter of the opening relative to the focal length of the lens. An ƒ/2 opening is ½ the focal length, ƒ/4 is ¼ the focal length, etc. The actual amount of light varies because the focal length is measured from the optical center of the lens, the point where the light crosses and inverts the image landing on the focal plane, and that point moves outward as the lens is focused on anything closer than infinity. So yes, it is approximate.

"FWIW, the optical part of photography is entirely based on math. It is not something to squabble about; math is math. You understand it or you don't. Since the advent of TTL metering it is less necessary to understand that math of optics/light but IMO still helpful.

"The perspective of an old guy who learned on all-manual cameras when you had to understand how it worked.  :-)  "

Steve Renwick: "Re 'But one thing I will toss out there is not to forget about the degree of precision in the device, which is a wild card we don't know.' Well sure. I have here two identical camera bodies with two identical lenses which when set to the same identical apertures and shutter speeds yield exposures that differ by about 2/3 of a stop. Or maybe it's 5/6 of a stop. So, y'know, stuff happens."

Andrew J.: "Anybody interested in a thorough explanation of relative apertures should read the excellent article 'f-number' by Wikipedia. Is ƒ/1.2 half a stop or a third of a stop faster than ƒ/1.4 is purely academic for me as I don't own such a lens and thus can't comment on the true subject of this post."

Bruce Alan Greene: "Motion picture lenses are marked in T stops, for the light actually transmitted rather than the mathematical stop. Even then...they are generally only accurate to ± 1/3 stop, except...at the widest aperture."

[Bruce is a cinematographer. —Ed.]

Comments

[Robert Pillow]

If you shot Tri-X extensively at F/1.2 when you were young, switched to digital in middle age, and still practice photography in your sixties, you may buy today’s mirrorless equivalents just because you miss the way the world looked in the old days.

[Jeff]

Isn't talking about the f1.2 lenses really a discussion of wanting vs needing? Thats something that permeates all of photography now.
If people only bought what they "needed" vs what the photography and internet industries works so hard to convince them that they "want" all of the camera companies would be out of business now.

[Michael J. Perini]

Mike, I completely agree with this post as a good benchmark attitude of how to think about lenses. It is good to understand the math behind the system, and just as important to understand the imprecision of camera systems and the range of judgement as to what constitutes "Correct" exposure for any given situation. Understanding it all helps you to manipulate the variables to get the results YOU want.
It is pointless to argue about it, because we can all do what pleases us.

If you do work for clients, where you are expected to bring back good results weather you find yourself in dungeon like darkness with no flash allowed to high noon sun at the beach, you have to have the tools that allow you to do that. So fast lenses still have their place. If you shoot for yourself, then it is your choice.
High Iso capability has mitigated those extremes to some degree, but not eliminated the need. The other point that sometimes gets overlooked is that sometimes manufacturers put their best technology in the fast premium priced lenses which can be use at all f/stops.

Having said that, I will also be the first to admit that another benefit of advancing lens technology and manufacturing has been the absolutely superb quality now available in f/2.8 so called 'Trinity' Zooms. Because of that I actually use really fast lenses less --more as special situation lenses.
There has never before been a time with SO many incredibly good lens choices available AND it is occurring at the same time as an unprecedented ability to adapt old lenses to our cameras.
True abundance if you ask me......

[SteveW]

I'm no expert, not remotely. But I owned a Nikkor 58mm f/1.4G for several years. I also owned the 50mm f/1.8G during the same time. The 58mm had a look or rendering that was distinctly better to my eyes than the 50mm. The difference in low light performance was not significant, but the 58mm made better photos in low light due to its lens design, and I pretty much never used the 50mm in any light, the 58 just made better photos. (But that 58 was expensive too, the most expensive lens I've ever owned).

[Steven Willard]

I thought you might mention “T” stops in your discussion.

[c.d.embrey]

I did not know that.

Since time immemorial, the answer has always been RTFM—simple as that 8-)

BTW without picayune differences. the internet as we know it would not exist. Before the World Wide Web there were Usenet News Groups. News Groups always had flame-wars raging over picayune differences.

The high in coastal Orange County, California is low to mid 60s Fahrenheit.

[Malcolm Myers]

I had typing lessons about 20 years ago and I can sort of touch type. However, I find I look at the keyboard too much. As a result of your posts about typing I have been staring at the screen as I type and I find that my touch typing is becoming much faster. I'm still using the backspace key more than I would like, but it's definitely an improvement. So don't look at the keyboard!

[Dan Khong]

Your scale shows the currently displayed f-stops in red. I have some ol' Leitz glass that uses the Stolze aperture scale and I would set my favourite mid-aperture to f6.3.

[Jeroen Pulles]

And then there's t-stop! If you're buying the lens for its fastness, you may want to verify the t-stop.

T-stop is the measurement of how much light passes through the lens. Since no lens has 100% transmittance, these numbers are guaranteed to be lower than the f-stop.

DXOMARK publishes t-stop for the lenses it tests: The Olympus 25mm f/1.2 Pro has a t-stop of 1.8; And that's with 19 elements. The nine elements Pentax 55mm f/1.4 da* has a t-stop of 1.7.

That Fuji 56mm f/1.2 you mentioned has a "APD" variant with an apodization filter ("STF" in Sony/Minolta-speak). That darkens the outer part of the lens. From what I can find on DPReview its maximum t-stop is 1.7.

[David Dyer-Bennet]

My habits are certainly partly left-over from when I shot film, and hence had to worry about ISO speeds of 1200, 1600, maybe sometimes 4000.

But, for example, since my 150mm lens is only f/2.8 (field-of-view equivalent for full-frame is 300mm), and since I need an absolute minimum of 1/350 second shutter speed and would prefer 1/500 to freeze faces reliably (letting moving hands and feet blur some adds to the sense of motion so that's okay), and since ISO 6400 looks significantly better than ISO 12800 on my camera...I'm glad I've got the 40-150/2.8 lens ($1400) instead of the 40-150/4-5.6 lens ($149). Call it 10x for 2 stops better at the long end. (Actually I've got both, but the slow version is basically deadwood in my cabinet.)

But f/2.8 is annoyingly slow for a 150mm lens. I've got a Canon 135/2 on a fancy enough adapter that AF works usefully...but not usefully enough to track roller derby skaters moving at up to 35 MPH (plus they get close and far and close again fast enough I really want the zoom). (Basic minimum skills to play requires about 12MPH for extended periods, but lots of teams have people who burst up to 35 when they want to.)

[Marc Lawrence-Howe]

I feel like 90% of photo forum opinion is just subjectivity being defined as objectivity and being rigorously defended by every individual from their own piece of flat earth... is what my frustrated, internet self wants to say.

But the more I do my thing in photography (or writing) I realise that the thing that matters most to me is to strive to get to the point where all aesthetic advice is “wrong”, all technical advice is irrelevant, and all “rules” are a pox on my pleasure. My choices, desires, practice in this area and others are as inseparable from my self as my neurology, and as I fully try to relieve myself of my dissociative tendencies of depersonalisation and derealisation (something for which my camera is a medicinal tool of realisation) I wish every others’ words to become a low hum that I hear in the gaps of my attention and which humbly remind me of my own virtualised solipsism and that this gentle tinnitus is simply a social lubricant to shared joy.

One day I will own an F1.2 (or f1.0) lens and it will be exactly, perfectly, uniquely “right” in a way no other lens can be. Anyone who argues differently... it’s okay, chill, mate, no one can statistically prove that anchovies are good or bad, and if you think they are bad, well, there are plenty of other fish in the sea.

[Ilkka]

Can you also show the ‘accepted’ list of half stops? Especially between f/1 and f/2. Comparing that with the above third stop list will be interesting indeed.
There is nothing sacred about thirds of a stop. Many large format lenses had thirds of a stop markings while small and medium format lenses were more likely to have half stop clicks. Hasselblad lenses had the EV scale with interlocking aperture and shutter speed scales, at half stops for aperture. Only lately some lenses started to have 1/3 stop markings or click stops. Mechanical shutter speeds were almost always only available at full stop intervals while film speeds were traditionally offered at 1/3 stops (for example ASA/ISO 32, 125, 160).

[hugh crawford]

Funny, I’ve never seen a 1.6 lens except maybe on some 8mm cameras and some projectors where I got the impression that calculating the f/stop was something along the lines of “the guy setting up the engraving machine wants th know what the f/stop is” and figuring it out.

[Clay Olmstead]

Acknowledging that it doesn't matter from a photographic point of view, but it brings out the engineer in me: is a lens with maximum aperture labeled '3.5' actually 1/2 stop slower than a 2.8 lens, (actual 1/f = 3.414) or 2/3 stop slower (actual 1/f = 3.609)?

[Daniel]

On how precise so much is when we make photographs.
As one wag put it"Ansel and Fred Archer came up with the Zone System, not the pinpoint system".

[Roger Bradbury]

I set all my cameras up to change apertures (and shutter speeds and ISO) in half stop increments. It's quicker to change the exposure compared to if it was set to a third of a stop increments, and the furthest out I can be from the ideal exposure is a quarter of a stop. More than close enough.

[james wilson]

I recall that in the early days of super fast lenses that they had better specs than the slower versions. Wide open maybe not so great but the 1.4 would be sharper than the f2 @ f4 or f5.6 and the 1.4 would always be sharper @ f2 because the f2 lens was wide open so you did get something for your money even if you never shot that f1.4 wide open

[Nikon, the leading professional brand in the second half of the 20th century, did make its fast lenses its best lenses. The rationale was that it was mostly pros who bought fast lenses, and pros would spend more. As to your last point, it was true in some cases, and not in others. --Mike]

[Andrew J.]

The problem is that by convention, apertures are noted with one decimal. If it were two decimals, the whole discussion would be moot.
By the way, I would greatly appreciate seeing what you snapped with your 56mm f/1.2 on your last evening stroll.

[xf mj]

Ya know, despite being a math person (or maybe because of it) I made the same mistake several other readers did ("1.1 squared is just above 1.2, 1.2 squared is just above 1.4, and 1.4 is nearly the square root of 2.0"). Thanks to your listing above and your chastising (and trusting in your authority) I'm now finally understanding why f numbers can go below 1.0. Your proverbial wack on our collective knuckles with a ruler made me rethink how those numbers must work and what they mean, and hopefully correctly now.

Meanwhile, because I normally like math and think I'm often good at it, I am so ashamed! How internet of me.

[Kirk Decker]

Back when I was in the PPA, I took a week long portrait class from a guy who would create environments which usually entailed what seemed like several hundred yards of unbleached muslin wrapped around whatever was handy, including the models. He would use an analog meter (like everyone else) to set the camera and shoot a Polaroid that we could pass around while he told us how his work was all about PASSION and FEELING and not f/stops.
A few years later he was a speaker at an event. Digital meters had just come out. His meter displayed the results in 1/10 stop increments. Well that changed everything. When he metered his lights it had to be f/8 and 0/10ths. To be off by more than a 1/10 of a stop meant your photo was a failure. We’re working with people on color negative film here. Flash - Check the Meter - Scoot the big softbox back 3 inches - Flash - Check the Meter - Scoot it forward an inch - Flash -awww PERFECT!
After lunch, I decided I’d rather go on a one person photo walk than go back to the seminar. The battery in my meter was dead, but f/8 at 250 seemed reasonable. My photos came out pretty good.

[David Lee]

An f2.0 aperture is an f 2.0 aperture only when the lens it’s in infiniy setting.

[Animesh Ray]

I own a f/1.2 50mm AIS lens, which I used extensively on my Nikon F3. I liked the bright image through the finder and the images rendered well on Tri-X when I used the maximum aperture. At above f4 the images are very sharp. When I switched to dslr, I kept using the same lens because who wants to buy another lens when one has a perfectly good one? Yes, it might be a little difficult to focus relative to modern autofocus lenses but I never know it because it is the only 50mm lens I have on a "through-the-lens" body. I now use this f/1.2 lens at the maximum aperture for making dreamy portraits, which allows me to resist the temptation of buying a shallow dof 80mm portrait lens.