[Not a Number]

The way he describes the limited series as having qualities that can't be quantified, only qualified, is most startling, coming from someone like him. Yet it closely reflects what others have stated.

That's exactly what Ricoh has said about the much maligned accelerator chip yet people insist on dragging out charts and graphs in trying to shoot down the AC.

[house]

The general design goals of the limited series are outlined. I think it's fascinating that he contrasts those goals explicitly against the * series lenses, which are designed to bench standards with numerical goals. The way he describes the limited series as having qualities that can't be quantified, only qualified, is most startling, coming from someone like him. Yet it closely reflects what others have stated.

It's quite a strange thing to say because many world argue that the * glass has those qualities as well but in addition to excellent numerical performance in most areas. When the conditions are right the DFA50 images are quite special.

We are dealing with a corporation flogging gear so what they say must be taken with a pinch of salt. Their statement that Limited lenses are designed by judging prints sounds almost unbelievable in this day and age. I buy it for the FA77 but can't imagine them doing it now as a *major* part of the design process. What do folks think, is it possible these days?

[c.a.m]

We are dealing with a corporation flogging gear so what they say must be taken with a pinch of salt. Their statement that Limited lenses are designed by judging prints sounds almost unbelievable in this day and age. I buy it for the FA77 but can't imagine them doing it now as a *major* part of the design process. What do folks think, is it possible these days?

Forum member @JPT reported on a 100th Anniversary talk he attended in November 2019. The presenter described some of the aspects of the development process for the FA Limited lenses: The FA Limited Development Story (also a little DA Limited) - PentaxForums.com

If I may quote JPT:


"Unique approach to design and evaluation
As they developed the lenses, they were advised by a famous photographer, the late Shoji Otake. The common practice at the time was to use to use computer aided design and evaluate using test charts. They rejected that methodology for the Limited lenses and instead used a painstaking method of human evaluation of prints. This is understandably a very slow way of developing lenses.

They invested in an expensive, state-of-the-art printing system from Fujifilm. Using the images produced, prototypes with different designs and coatings were evaluated. When they started this practice, the designers were not always able to see the differences, but Otake-sensei could immediately see the significant areas and told them where to look. Once they developed the ability to see the differences, they became skilled in the art of evaluating images themselves. Ikenaga-san commented how much they felt their own eyes improving by being part of this process.

For each lens they made a description of what it should be able to do. For example, with the FA 77mm, they wanted to create a lens that suppressed flare to to avoid washed out photographs. They defined this as being able to depict the glint of a white button on a white shirt. Then they went about the process until they achieved it, and this is why the ghostless coating was applied."

I would be surprised if the designers of the FA Limited lenses did not incorporate some numerical design techniques or never measured fundamental optical properties. Such steps would have allowed them to get an early confirmation of their key design choices and to arrive at their first set of candidate prototypes. Evaluating prints would have provided important qualitative insights, certainly, but it would have been difficult to specify all of the many real-world test markers (such as the glint of a white button) that would be necessary to qualify a lens completely and to conclude that it's ready for the real world. Furthermore, as suggested in the November report, evaluation or interpretation of printed images by humans depends on a degree of objective skill, but different humans can have different subjective interpretations. Consider the extended discussions that happen here in these forums concerning a tinge of purple fringing, bokeh, and corner sharpness.

I think the designers probably depended on the qualitative assessments, while also using numerical design and measurement techniques.

Maybe @bdery can offer his perspective? What do you think?

- Craig

[texandrews]

It's quite a strange thing to say because many world argue that the * glass has those qualities as well but in addition to excellent numerical performance in most areas. When the conditions are right the DFA50 images are quite special.

It's one reason I found the comments unusual. It's not the first time Pentax has made noise about the limited series (just the fact that it exists is curious...)---they did a whole video about them.

We are dealing with a corporation flogging gear so what they say must be taken with a pinch of salt.

Always true, but this isn't the marketing team doing this. Having worked in a company split between marketing and sales and production, I know what pure marketing sounds like. This was something else...

Their statement that Limited lenses are designed by judging prints sounds almost unbelievable in this day and age. I buy it for the FA77 but can't imagine them doing it now as a *major* part of the design process. What do folks think, is it possible these days?

Well, I think for a small company like Pentax, a niche product like the limited series, and a product the qualities of which resist straightforward bench testing for all its "goodness", prints sound like a great idea! What better way to see if the lens was delivering this "character"? And if you could do them for the 77, why not the others?

[monochrome]

The general design goals of the limited series are outlined. I think it's fascinating that he contrasts those goals explicitly against the * series lenses, which are designed to bench standards with numerical goals. The way he describes the limited series as having qualities that can't be quantified, only qualified, is most startling, coming from someone like him. Yet it closely reflects what others have stated.


This mirrors all sorts of discussions here and elsewhere about lenses with "character" , and also mirrors discussions about the differences, especially now, of the best FF cameras and digital medium format---differences that some very solid people struggle to quantify, yet others can see, at least in certain circumstances, differences that are beyond DOF considerations.

It's quite a strange thing to say because many world argue that the * glass has those qualities as well but in addition to excellent numerical performance in most areas. When the conditions are right the DFA50 images are quite special.

We are dealing with a corporation flogging gear so what they say must be taken with a pinch of salt. Their statement that Limited lenses are designed by judging prints sounds almost unbelievable in this day and age. I buy it for the FA77 but can't imagine them doing it now as a *major* part of the design process. What do folks think, is it possible these days?

We call lens character pixie dust and frequently apply that quality (the lens has pixie dust) to the FA Limiteds, especially the FA 31. I think we say the DFA50/1.4 and 70-210/2.8 are technically superlative lenses but measure their characteristics quantitatively more than qualitatively.

We know pixie dust is there. We can’t measure it, but we can see it - especially in prints. It would not surprise me at all to learn Limited lens design Intentionally aims at a different output than * lenses, that proprietary Pentax design “know how” is used for Limiteds with pixie dust as a design goal, and final evaluation of Limited lens rendering is done by viewing prints.

I have to hope the 21 Limited has pixie dust (I think it will).

[MJKoski]

Well it better have quantitative pixel dust. 8K monitor on the wall in 100" size is the future of displaying large photos. 8K is pretty ruthless here - it is K-1's native resolution and lenses must draw corner to corner to be usable. Pixel-Shift is preferred. S1R 160MP or A7R4's 240MP (competitive superresolution cameras) is the new norm from which the 8K material is to be downsampled from.

[Parallax]

Yay...!.
Much needed wide angle prime for the K-1. And its a limited series. So we all know what to expect.

Let's add it has ED element, is weather resistant and has built in DC motor.

I've had little interest going going FF other than to make the best use of my 31 Ltd.
Little interest until reading about this lens, that is.

[texandrews]

We know pixie dust is there. We can’t measure it, but we can see it - especially in prints.

We can't measure it yet, or we aren't measuring it yet.

[texandrews]

We know pixie dust is there. We can’t measure it, but we can see it - especially in prints.

We can't----or maybe aren't---measuring it yet. As was hinted in the above passage, maybe Pentax has found some rough way of measuring this? Anyway, if it's a definable quality, there's going to be some way of measuring it. If we are detecting difference---and the difference is there and not imagined---then there has to be a way of measuring it.

Well it better have quantitative pixel dust.

We'll see before we'll quantify, I'll wager. We should not be surprised that something is not yet quantified----science and engineering don't measure things all the time! They measure things the measurers decide to measure.

8K monitor on the wall in 100" size is the future of displaying large photos.

Well, it's certainly part of the future. But here, I think I have more knowledge than you: I work at a major museum of contemporary art, my wife works in another significant museum in another city. Both of our museums would classify as "second tier", where the British Museum, the Louvre, the National Gallery of Art, MOMA, & etc are in the first tier----so second tier museums are a big deal. I work in the Exhibits Department, and my wife splits her duties in the Registrars' Department and Curatorial (Asian art, and her last exhibition was contemporary Asian photography...). Yes, we use monitors to display photography....sometimes, and when the artist suggests it. Don't be holding your breath for museums to be switching over to 8K monitors anytime soon for photography exhibitions---and I mean major ones. Galleries? only the very top ones, names like Gagosian, Hauser and Wirth, Pace, Matthew Marks, Zwirner, & etc in the U.S., comparable ones in Europe and East Asia----a very rarified group. And believe me, they will still be handling prints...Very few museums or galleries have pockets deep enough to mount an exhibition with only 100" 8K monitors, even the biggest ones. Furthermore, few places want to invest in tech that will be quickly obsolete. We, for instance , have a whole painting storage screen (they are huge, 14' tall x 30' long) covered in obsolete monitors. Storage problem? Uh......ya think? Remember, stuff like that's not art, and storage is prioritized for art. Call any museum in the world, ask for the registrar's office, and ask them if they have enough storage....We can't even surplus them. We use them for some time based works still, but more than that we project. I know we won't be buying a bunch more, especially now. Do you know what a deep financial hole museums are in now? That's not magically going away soon, nevermind what a certain world "leader" says (or his son....). The major galleries could rent, of course...but I'm not holding my breath. Eventually of course you will be correct, but that eventually is some ways off, many years. And there will still be prints.

8K is pretty ruthless here - it is K-1's native resolution and lenses must draw corner to corner to be usable.

That is biased towards a certain kind of photography that you (and I, sometimes), plus the Dusseldorf School, photographers I like such as Toshio Shibata, and many others, it's true, favor. But art photography is a mighty big world---plenty of room for stuff that's not so sharp in the corners...even an artist like Sugimoto has work in which corner sharpness means zero. And there's room for things that aren't 100" on the diagonal as well.

[Serkevan]

I'm pretty sure that the first thing on anyone's mind when they look at Don McCullin's Vietnam photos is "man, those corners are not as sharp as they could be".

[BruceBanner]

'clinical' or 'character'... what is 'character' anyway?
I have the 'clinical' D FA* 50mm f/1.4 and I can't see any downsize in the images it can render. Except for me using it, which is quite a waste of an excellent lens

While I've seen a few cases where 'characterful' bokeh made for a more interesting picture, I'm on the "it shouldn't distract from your subject" camp, like Norm.

I'm sure that's not true.

We must remember that photography is a 2D image, it's art and we all like different things and that is totally fine. It's ok to like a portrait where strong emphasis is purely on the subject and the background context not competing. I like this kind of image too (studio work for example), just perhaps not always, perhaps sometimes I'm taking a portrait shot in a pretty location and I want the entire frame to be interesting and lovely to look at, not just the portrait, we're crossing over here from portrait to environmental portrait. And its not that the DFA 50 can't do fantastic environmental portraits, of course it can! It's just there's a different feel and look to the bokeh. There's really no right and wrongs. I would love to own the DFA50.

I'll try to give some examples of 'character' images;



























It's a different kind of portrait, some of the images are trying to have 3D subject isolation but also context to where they are, it's unlike the kinda work you might produce shooting with a 70-200 (@200 with 2.8). Where possible we're enjoying natural producing vignette, the bokeh should hopefully feel layered and textured, detailed (but not overly) to compliment the image. Dunno if that helps..

Well it better have quantitative pixel dust. 8K monitor on the wall in 100" size is the future of displaying large photos. 8K is pretty ruthless here - it is K-1's native resolution and lenses must draw corner to corner to be usable. Pixel-Shift is preferred. S1R 160MP or A7R4's 240MP (competitive superresolution cameras) is the new norm from which the 8K material is to be downsampled from.

Well I would respectfully disagree. You're judging all photographs to meaning they must hold up well at those sizes, which I don't think I agree with. I haven't met a bride and groom that wanted their kiss shot in 8k 100" on the wall, sometimes these images are best left in a format more tactile such as a wedding book or album. Not every image needs to hold up in large to be judged successful or not.
I get what you're saying, the future is here, some landscape shots taken with massive megapixels is really cool fun and a joy to view large, but this isn't what I consider 'a standard'. I don't think the Ltds are designed with the intention of ending up on 8k 100+ screens (and that's ok).

[Serkevan]

I mean if super high resolution for ultra-sharp-corner-to-corner-football-stadium-sized display is what's needed, might as well do stitching. Or use a good old 8x10, set f/64 and be done with it .

[Parallax]

Or use a good old 8x10, set f/64 and be done with it .

Perfect, and an 8x10 contact print from Tri-X has remarkably fine grain.
(Grain, for you yunguns who've grown up in the digital age, is roughly equivalent to "noise" )

[yorik]

They should have made it a prime number - a 19mm!

Or 23, then it would be prime, wide (on K-1), and ~35mm equiv on K-new.... Win win win.

[bertwert]

(Grain, for you yunguns who've grown up in the digital age, is roughly equivalent to "noise" )

I thought grain is what they made flour from...