Geek Articles

Why Manufacturers Make a Specific Camera Lens

Published February 2, 2022

A Camel is a Horse Designed by Committee

There is a lot of online discussion about why a manufacturer made this lens when they so obviously needed to make that other lens. Or why this manufacturer’s design is better than that manufacturer’s magical solutions. So, I thought I’d share the second-hand information I have about the process because it’s probably better than the made-up speculation I read online.

Let’s get my limitations out of the way first. I am not a lens designer. I don’t have any degree in optics. I don’t even own a copy of Code-V or Zemax. No manufacturer has ever let me sit in on the complete design and development process, nor are they ever going to. Nor would I want to; it takes an eternity and I have the attention span of a squirrel.

But I am somewhat educated in lens making. I hang out with manufacturer’s engineers and listen to them gossip over too much sake. I own half of a company (C-4 optics) that has designed a few lenses and manufactured one. Occasionally, I participate in the process for both major and third-party manufacturers, testing prototypes during the later development process.

Lens Designers are Fighter Pilots

Can you imagine telling a fighter pilot “I bet you can’t fly that?” Lens designers are like that. Sure, they’re geeks, but the top lens designers are certain they can build anything, and they want to prove it.

Haven’t been able to source this. . . . .

For example, when Dr. Bryan Caldwell and I formed C-4 Optics years ago, the idea was it would be a hobby business for both of us. Bryan, a superb lens designer, had already designed lenses he thought were great ideas. A 67mm f/0.666 because he thought The Aperture of the Beast would be a cool name. A 150mm f/1.0 lens just to prove it could be done. The most practical of the bunch is the one we made: a 12-pound, 270-degree field-of-view fisheye lens. Narrator: but it really wasn’t really practical.

This is what happens when you leave lens designers unsupervised. (No one is going to accuse me of being adult supervision). I literally wrote them all up in a blog post, which I released on March 31, 2015. People thought it was a hysterically funny April Fool’s joke. Later, when we made the Hyperfisheye, people were all ‘wait, you idiots were serious’?

But Companies Aren’t Run by Fighter Pilots

Here’s one thing I always check that almost none of you do: what is the background of the CEO? That will tell you a lot about where the company’s emphasis is. They are usually run by cost-conscious business people or, even worse, marketing people. Very occasionally, a lens company is run by an engineer.

No matter where the boss comes from, every company has a marketing department, an accounting department, and an engineering department. One of those may drive the bus, but the others all make suggestions from the back seats. In most cases, it’s a big bus, and kids in the back get rowdy. Let’s consider what they all want.

Some non-optical engineers will make the mechanical and electrical parts, sort of the ground crew. They don’t get the publicity, but their job might be more difficult. For example, when we designed the Hyperfisheye, Brian’s optical design covered 7 pages. The mechanical designs took 142 pages. And this was a very simple manual focus lens with no electronics.

There are, at most, 2 dozen elements in the optical design of a lens, but there are thousands (literally) of mechanical and electrical parts.

The mechanical design engineers don’t want to redesign 5,409 parts from scratch; they want to use existing parts they have in bins in the back. Failing that, they want to outsource the parts to someone who already makes them. Other engineers specialize in designing the assembly process for the lens, either on existing lines or by designing new lines and methods. (Hint, these guys would like to use the existing lines and methods).

There are marketing people who want to have theoretical specifications they can brag about, whether they can really be manufactured or not. Here’s a thought experiment for you. Let’s pretend the design team has designed this lens that has an AMAZING computer-generated MTF chart. The mechanical team says, “we can’t possibly make it to those specifications in the real world”. Which way do you think the marketing team votes?

The marketing team also evaluates what the market wants (or what they think it wants), what the competition has, and is always hoping for cool special features (real or imaginary) that they can use to differentiate their lens. Otherwise, they have to write copy about “seeing the world differently”, “preserving visual integrity”, and “bringing your image to life”. Again.

So, Why Do They Decide What They Decide?

This varies a bit depending upon if it’s a camera manufacturer, a large third-party lens maker, or a small boutique lens maker. There is flow up and down the food chain. Mostly the higher-ups say ‘make this lens’ but sometimes the design team floats up some designs for consideration or the marketing team identifies opportunities.

Wherever the idea originates it goes up and down and back and forth for a while as the various groups balance out the ‘lens design triangle’: cost vs size vs quality. Every company emphasizes different things. It might be of great quality and low cost but massive. Others make their living with small inexpensive lenses with adequate image quality. Some make small, superb lenses that cost as much as a decent car. You all know who is who. The camera brands generally try to balance the design triangle pretty evenly.

Major Manufacturers

Camera manufacturers first and foremost need to make sure the basics are covered: a full set of zooms from wide to telephoto, at least a few prime lenses, often some not-quite-so-wide-aperture primes and zooms and superzooms. Manufacturers usually want at least one macro lens in their lineup and those who want to be perceived as ‘professional’ for sports photographers want an expensive super-telephoto prime or two. Tilt-shift lenses used to be part of the ‘we are professional grade’ merit badge, but that seems to not be the case as much now. 

Manufacturers will usually prioritize a show-off lens or two in their lineup, a loss leader that says ‘we did this thing no one else does so you can see how good we are’. Think Canon RF 28-70mm f/2 or Nikon Z 58mm f/0.95. Sometimes they hold down the losses on that loss leader by making it in such small quantities that it’s really difficult to buy one.

Primary manufacturers (Canon, Sony, etc.) tend to be more top-down; the higher-ups give a list of lenses to be made and let the various groups fight about how best to do that. This is not a fast process. The directive to make a 24-70mm f/2.8 lens, may spend a year in optical design, often with a few teams developing separately. Then several designs do the back and forth amongst the various committees. Sometimes two or three are actually prototyped and tried out. Only after that do they get into sourcing elements and parts and designing the assembly process. Even a prioritized design takes several years, minimum, to reach the market.

Third-Party Lensmakers

The big third-party manufacturers have resources equal to the primary manufacturers. (Some primary manufacturers’ lenses are actually designed and to some degree manufactured by third-party manufacturers, then rebadged. But that’s another story that most of y’all don’t want to know about). At least one third-party lens I know of was an optical design made to sell to a major manufacturer; when it was turned down the third-party company made it themselves.

The usual third-party formula is nearly-as-good-but-significantly-cheaper. Or it might be better optics but big and heavy (think Sigma primes).  They also are like predators, swooping in where they find a weakness in the manufacturer’s herd. But while third-party manufacturers may be more nimble than the primary manufacturers, lens creation is still a slow process. “Swooping” takes a couple of years, at least.

Third-party lens makers often allow designers a bit more freedom. For example, I sat down with a large group of engineers from one company and asked why they had chosen to make a very excellent lens in a really odd focal length. They all started laughing and introduced me to the designer – he had come up with the design on his own time and presented it for consideration. It got accepted and he got promoted.

So, How Does This All Work Out?

Rather randomly, actually. Let’s say, for example, the lens designers play around with their software and come up with an amazing 50mm f/1.2 lens that weighs 8 ounces and is sharp as a tack. Then the manufacturing engineers say “we can’t get those parts”, the financial guys say “it’s going to cost $12,000”, marketing chimes in with “nobody wants that”, and the assembly engineers add “we’ll need a new assembly line.” The project is dead.

In many companies, the commands are more top-down and less flexible. The CEO says “Thou Shalt Make a 50mm f/1.2 costing under $1,200.00 and weighing less than a kilogram.” Then the committees go back and forth figuring out how to do that. Eventually, they arrive at a conclusion like “We can make a 55mm f/1.2 costing $1,600 and weighing 1.1 kilos.” Then the higher-ups decide that’s either close enough or it isn’t.

Larger companies often assign a specific lens to two or more competing design teams. This may be ‘each group design a 24-70 f/2.8 zoom’. It may be ‘this new mirrorless system is coming out, we need one spectacular special lens for it’.

And remember this is NOT a quick process; it takes years. One company I worked for told me they had the replacement for a certain lens as their absolutely top priority. The lens was actually released about 4 years later. That’s what a rush job looks like in this business. As you’ve noticed, standard ‘upgrades’ to mainstream lenses are released a lot less frequently, often a decade or more apart.

Remember when Canon and Nikon went mirrorless, and early on they both came out with unique, show-off kinds of lenses? Lots of people asked ‘why this and not that’. Truth is there were probably 4 or 5 design teams competing to make the show of lens, and committees decided which of those camels would be the most unique.

So when you’re online screaming ‘they need to come out with a new version of this lens’, well, they’re probably already doing that. But you won’t see it released until you’re shooting your next camera. And when you’re wondering ‘why in the world did they make that’? Well, usually either because somebody had a bright idea, or compromises were made.

Roger Cicala

Lensrentals.com

February 2022

Author: Roger Cicala

I’m Roger and I am the founder of Lensrentals.com. Hailed as one of the optic nerds here, I enjoy shooting collimated light through 30X microscope objectives in my spare time. When I do take real pictures I like using something different: a Medium format, or Pentax K1, or a Sony RX1R.
Posted in Geek Articles
  • Roger Cicala

    I kind of hint here and there, but that gets right against some NDAs. And it’s becoming moot; the name on the outside contains 4-5 other names on the inside, with a few exceptions.

  • Roger Cicala

    You’ll like my next article on early lenses; basically lens designs from pre -1900 were refined and refined and still being used in 2000. Only in the last decade have they been dropped for new technology.

  • Roger Cicala

    As well they should have.

  • Zé De Boni

    Edit: I should have said “… from creative and visionary USERS”, as I must also acknowledge the ideas you share with your audience and recognize how much the industry misses by not hiring consultants like you.

  • RC Jenkins

    One theory I’ve had is that the Nikon Z50–which was released in late 2019–is actually Nikon’s oldest Z camera by design. The clue is that unlike any other Z camera, instead of USB-C, it uses micro-USB (which got out of fashion in electronics / phones by around 2015-2016). So I would suspect the Z mount specs were already well defined by that time period.

    So 2014-ish sounds plausible to me. This would have been in the year following the first Sony a7 (announced late 2013), where Sony’s head start came from reusing E-mount from their earlier NEX cameras.

    And also for Nikon specifically, 2014-ish also would have been at the tail end of the Nikon 1 system releases (remember those??)–their last Nikon 1 product was released in early 2015. I don’t think it’s a coincidence they stopped after that–I’m guessing these teams would have been reassigned to work on Z’s instead from that point onward.

    I’d guess Canon would have a similar timeline.

  • Franz Graphstill

    So good to see a new Roger C blog entry.

    I had not realised it took so long for a lens to get to market.

    It was five years from the announcement of the Sony 70-200mm f/2.8 GM to its replacement by the mark II late last year. So Sony released it, and almost immediately began work on the replacement – hmmmmmm 🙂

  • Henry Winokur

    I must say it’s been horrific for almost everyone but that little monster from China. I’m definitely looking forward to seeing your musing, Roger. Let us all hope that we’ve seen the worst of it and things will get better for the entire world!

  • Zé De Boni

    Thank you too, Roger. Technical articles like the ones you write are always enlightening. We learn a lot. It is a pleasure to add ideas collected in the long term of practical experience. Commenting on the subject of this text and inspired by your historian side, I see the evolution of equipment, not only photographic, always too tied to past models and paradigms. Consumers and professionals are so slow to accept evolutionary leaps and explore new possibilities, which reflects on the industry’s restrained creativity. The automobile analogy is again appropriate: how long did it take to abandon the wagon model and adopt the more appropriate streamlined forms? It’s been over two decades of digital photography and cameras still keep features that are conditioned by the limitations of film. The resistance to abandoning the optical viewfinder is just one of many examples that can be mentioned. Meanwhile, evolution is contained and many of the facilities that digital technology could already be providing are still reserved for the future. Engineers work wonders with electronics and computing, but much more can be gained through simple physical, mechanical and design improvements. In this constant evolution, I believe that manufacturers (all them) need the active collaboration (not just feedback) from creative and visionary photographers.

  • Roger Cicala

    What Ryan said, and we’re probably overdo for one of those. We did a couple of FB AMAs but nobody came so I doubt we do any more of those.

  • Roger Cicala

    For what it’s worth, 2022 was a really difficult year for business, life, and general chaos. I plan on getting back to writing regularly now.

  • Roger Cicala

    I don’t know either, but I’d bet 2014-2015 at the discussion level. Prototypes would have been a bit later.

  • Roger Cicala

    Thank you! That is well thought out information that I hadn’t really considered.

  • Ze Dong

    post first, read later

  • Ryan Hill

    Ah, I should’ve mentioned that in the first place. Just fill out this form!

    https://docs.google.com/forms/d/e/1FAIpQLScSnDpnml8hMFQPQbWClnG7Pj5O2vgbFvM0KkQ0KN_aOB4–g/viewform

  • Hubert Baierl

    Love this article … Roger’s blogs are always worth reading.

    Pls allow me to comment: A joke only works if it contains a decent share of truth. BTW: I am a marketing guy, but not in the photo industry.

  • H.A.

    Gotta say I love Sigma’s oddball focus lengths. 120-300 (Basically what you get if you are used to 70-200 on crop body), 300-800 (Even if I don’t own it), bigma, etc.

    Thanks for the insights in the process!

  • Karl Fiegenschuh

    Thanks, Ryan. How do I submit questions for an “Ask Lensrentals” episode?

  • Impulse_Vigil

    I’d love to know how the marketing people really try and read the pulse of the market tbh, I imagine it varies by manufacturer but are they in there in enthusiast boards taking notes or are they simply going by what’s always been classic etc.?

    I for one actually like seeing the oddball focal lengths and ranges, give me all the 35-150s, 17-28s, 65mm and 150mm, heh.

  • Impulse_Vigil

    Huh, hadn’t really thought about all the ramifications or ways in which T-S lenses could evolve on mirrorless… Meanwhile it’s kinda curious that there’s at least one 3rd party “swooping” in as Roger put it and making some native (but still manual) tilt lenses for mirrorless mounts.

  • A Roger article! I thought you had left us! Please visit again sooner next time.

    I read your DPR articles, but the more technical content fits better here.

  • Zé De Boni

    “Tilt-shift lenses… seems to not be the case as much now”
    Amazingly, it is inappropriate to use TS lenses with SLR’s as the optical axis gets misaligned to the mirror. It is really a mess. One must switch to life view, therefore a MILC is the perfect solution, with the added help of magnifier focus checking, for when we tilt the plane of focus to get things in focus, rather than that weird miniature model effect. We should expect TS to proliferate in the mirrorless environment.
    I’ve been dreaming of a future generation of tilt lenses (no shift ) that could automatically put in focus two points chosen on the touch screen, resembling the advancement that the Sinar P brought long ago. Surely an illusion, as it would need AI processing, some servo motors end else. But maybe I can bargain my dreams with tilt lenses that would transmit the settings info to get the proper software optical correction. Ok, this will need increased processing too, but it is fundamental for the use of such lenses with high demanding sensors. That may answer Roger’s original comment above, why they are forgoten. Today’s optics (I stress, optics not just lenses) rely a lot in post processing to achieve results that should be impossible just by means of glasses. It seems unfeasible to get automatic distortion, vignetting and LaCA software correction for a shifted image without inputting those extra settings. It is more effective to correct perspective in post and here again I dream of cameras registering pitch and roll to apply in camera (and in post) corrections. OTOH, the selective focus of a tilted lens is unique and its aberration corrections are much close to that for its standard (no tilt) position. We (I) need a new generation of tilt lenses. Please rush, not 5 years from now!

  • Ryan Hill

    Hi Karl,

    Thanks for listening! We actually do Q&A episodes over on the podcast. Look for any episode titled “Ask Lensrentals.” We’ll be doing even more of these after we come back from our winter break. Though I do also like the idea of making it an occasional column if we get enough questions, so we’ll see if we can make that happen. I appreciate the suggestion!

  • Athanasius Kirchner

    What a great article! It has a ton of valuable information, and actually made me realize that lens production is a more random process than I’d have thought.

    (Some primary manufacturers’ lenses are actually designed and to some degree manufactured by third-party manufacturers, then rebadged. But that’s another story that most of y’all don’t want to know about).

    But I do wanna know! Please write about that at someday, pretty please.

  • Not THAT Ross Cameron

    Hey, I don’t see any mention of the polishing of optics with ground up unicorn horn, sprinkling lenses with pixie dust or imbuing them with the wand of steady hands.
    Please tell me it’s not the same old politics as any other organisation.

    ;~)

    Awesome to see another article, and hoping for more in 2022.
    +1 to Supreme Dalek

  • Tuolumne

    So development of the Canon and Nikon full-frame mirrorless systems must have been started by, IDK, maybe 2014 at the latest? As in decide on basic specs (optical constraints) for the new mount, get teams working on the optics for a couple of introductory lenses, and start work on a new sensor, then fill in the mechanicals and electronics. I would have guessed 2016 or maybe 2015 before reading this article, but what do I know?

    I get a kick out of articles that state that one or both of these companies decided in 2018 to come out with a full-frame mirrorless system. There is no way.

  • Supreme Dalek

    Man, it’s great to see Roger’s byline again. I was afraid he’d given up writing except for DPReview. I hope this means there’ll be more geek articles and weird history stuff on the blog in the future!

  • Henry Winokur

    I always love to read (and as Karl Figenschuh said, hear) your stuff, Roger. I always learn stuff and I especially appreciate the humor with which your “musings” (can I call them that? ? ) are written. I assume you’re a busy guy, but I’d love to see more…and I’m gonna read some of the “old” stuff I missed. Many thanks for taking the time to let us know. It’s fascinating and explains why certain lenses haven’t been redesigned over the years.

  • Karl Fiegenschuh

    Very interesting article. I always enjoy your insights here and in the podcasts. Thinking of the podcasts, I’ve been listening to some of the recent rebroadcasts and wanted to ask a couple of questions, but the comments were closed. So I’ve been wondering: would you consider doing an infrequent Q&A column to answer photographers’ questions raised in your podcasts and blogs?

  • Eric Calabros

    I don’t know man..I want $1500 35-85mm f/2.8.

  • Speeding

    A great read in both content and style Roger. Appreciate getting a brief glimpse behind the curtain!

Follow on Feedly