Olympus PEN EE-2 Review – Fixed Perfection? – by Alan Duncan

Alan Duncan replied:

Comment posted: 16/05/2016

Interesting thoughts and in honesty without interviewing the design team from Olympus we'll never know and yes the main point is to cover as wide a range of EV settings as possible - but diffraction was an understood phenomenon well before the 1950's and arguably well before the invention of the camera (just google Rayleigh criterion and/or Airy Patterns for starters). Charles Johnson on his blog can recall reading about diffraction in Kingslakes seminal 1951 Lenses in Photography: The Practical Guide to Optics for Photographers although it's only his later 1989 A History of the Photographic Lens that can be viewed on line (and heck it mentions diffraction too). The development of diffraction limiting lens predate the rise of digital cameras too ((Leica for example are keen to point out their 1993 APO-TELYT-R 280 mm f/4 is one of the most diffraction limited lenses on the planet). The digital issue is intensified not directly due to sensor size but more the grid arrangement on the sensor being vulnerable to Airy Circle overlap (okay that's worse with higher density sensors so resolution does matter but in an indirect way)

ErikP replied:

Comment posted: 16/05/2016

Since this will inevitable turn into a pissing contest: All you writeand all your argument is an after-the-fact theory. These cameras was designed first and foremost to squeese out 72 half-frame slides or id lucky 76 frames, out of a common 36 exposure leica film. Even if slide films was way ahead of negative films and their fast fading copies in 1959, have you ever checked the quality of the films available around 1960? How old are you, was you THERE at the time? I have slide films from ca 1963, and kodacolor negative films still in my posession from the same time era. There can be no question at all thet resolution was "diffraction limited" with the films avaialable at the time. Try going to 12 - 15 ASA high resolution films you say? Well the inherent high contrast does not work well with the simplistic exposure system in this camera.... The film was the limiting factor. The film is still the limiting factor, except the type of modern high resolution films you seem to have chosen to show here. I have slides exposed by a similar, simple half-frame, a Ricoh half, exposed in the north sea, onboard platforms from 1979, Ektachrome. Also an excellent little camera, within its limitations. In failing light, difficult lightning conditions and extreme, against the light situations against massive electric lights, neither exposure nor resolution (in the film) was much to brag of... Your theory cannot be proven, nor easily disproven, that is the only thing you are correct of here. But your pictures are impressive, not much difrraction there!

Erik Prestmo replied:

Comment posted: 16/05/2016

Alan have you ever tried to scan 1960's Kodacolor films at full resolution? Or Agfacolor, Orwo or Peruchrome? I have, just recentli over tha past month. Enough said. Dysmal graininess, dysmal sharpness og first and foremost, color in the films are fading in spots, since the film is not stable. So no, even a simple achromat lens was limited by the FILM, not optical laws, that is for 4x4 or 6x6 negatives, 35mm demanded anastigmats a fact quickly recognisedby camera manufacturers in the 1930's, but was stilllimited by the dysmal films available.

Alan Duncan replied:

Comment posted: 16/05/2016

Not looking for an argument - as I've said I think your comment about the EV range is spot on and indeed your comments about 50's and early 60's film tech are valid (it's not my area really but my sense is the next big jump was the around the start of the 70's and given I don't date before then I'm not really in a place to comment). I just get rankled about the diffraction thing being only a digital issue (probably too as it was part of my CSYS Physics project back in the 1980's at school). Whether the PEN and Trip 35 metering system 1/200 to 1/40 step is geared to avoid diffraction is conjecture abet a mentioned one but I agree no one knows for certain

Erik Prestmo replied:

Comment posted: 16/05/2016

The point here is hyperfocal distance. Because in a fixed focus lens, only one distance is sharp, will focus to a point, every other distance "focus" is a circle of confusion, and more confused the farther away from the point of focus. Now all arguments about diffraction only applies to the point of focus set by the factory, everything else is hidden in the circle of confusion, and the factory operates with what the calls an acceptable circle of confusion. I've spent a few minutes looking this up on wikipedia and plugged in the numbers. Given the films available back in 1959, I think its optimistic to think Olympus (or any other manufacturer of half-format cameras) took the viewing distance fully into account and applied a circle of confusion of 0.017mm as suggested in the wiki article, I think they set the circle of confusion at the more normal 0.029mm which was used to calculate depth of field in hi quality cameras (SLRs) in later years. It might even be that they used a circle of confusion more like that was used in the 1930's by simple camera manufacturers, 0,050mm. Obviously what you use as a base defines the calculation. For a 0,029mm circle of confusion the hyperfocal distance (the only distance the lens will focus to a point) varies with the aperture chosen and is : f:5,6 = 4,86m that means everything looks reasonably sharp from 2,43m to infinity, if one accept unsharpness to 0,029mm. f:8 = 3,41m that means everything looks reasonably sharp from 1,70m to infinity, if one accept unsharpness to 0,029mm. f:11 = 2,49m that means everything looks reasonably sharp from 1,25m to infinity, if one accept unsharpness to 0,029mm. I dont know at which point Olympus fixed the focus for these simple little cameras, but what governs "sharpness" is acceptable circle of confusion, and at some point when the lens opens up, there will be unsharpness at the closer ranges, and that unsharpness has nothing whatsoever to do with diffraction, au contrarire, but of lack of means to adjust focus distance. Now, since we suspect focus distance to be set somewhere between 3 meter and 2,5 meter, the entire SYSTEM is interesting. IT will be sharp at one point only, but this is glossed over, to appear reasonably sharp by two more factors in the system, the film, which has a very limited ability to resolve detail in such a small negative, and the variance in the aperture, according to the lightning conditions. Films back then was able to resolve ca 50 - 60 lines pr mm, that is 0.017mm circle of confusion. If the lens is set to accept a circle of confusion of 0.029mm from infinity to half the hyperfocal distance, this will beseen as "sharp" with the film, since it can resolve that. What does this translate to in this days world? 0,017mm translates to 59 lines pr mm, or a resolution of 1062x1416 for a 18x24mm negative, or ca 1 503 792 image dots, or 1,4 "mpx" if you like. This was what the film was able to "see". 0,029mm translates to ca 35 lines pr mm, or a resolution of 630x840 for a 18x24mm negative, or ca 529 200 image dots, or ca 0,5 mpx if you like. This is what the lens was set to see, since the designers gave up the ability to focus the lens prescisely. If you think about it this translates really well with everyday experience with half-frame cameras, their image detail is usually limited, far less than with fill frame 24x36mm images. And even if diffraction is part of the rather involved mathematics that is at work, it has no bearing here, the engineers gave that up, when the went for fixed focus, or even scale focusing as in the Trip 35, and ditched a proper Leica rangefinder. Hyperfocal distance is about confusion, acceptable circle of confusion, and represents a shortcut, which frees any argument from advanced stuff like diffraction, which only was brought to the attention to ordinary amateurs like me, with the advent of digital cameras and super-resulution in modern digital cameras, which far outshines anything that was attainable with 35mmfilm and simple cameras.

Alan Duncan replied:

Comment posted: 16/05/2016

Erik that's a good summary of hyperfocal distance but diffraction still mattered even then. In my earlier response I mentioned Rudolph Kingslake 1951 book Optics in Photography. Kingslake was a pretty big cheese in the world of optics in the 20th century - both academically and in the camera industry (he designed lens for Kodak). The book was intended for the likes of you and I. You can find the book in part on Google books and luckily that part includes his discussion on "Diffraction blurring at very low apertures" (pg 62-63). In it he gives an example of the effect on a 2 inch lens (near enough 50mm in new money) taking a shot of brickwork at 50ft (~15m) and how at a very small aperture (f/64) the cement between the bricks would be obliterated by diffraction spot of 19mm. Okay that not a realistic aperture here but he makes the point that even at f/16 the diffraction spot will be 0.2mm much bigger than the circle of confusion effect you give in your example. The effect would be worse on the PEN with a smaller 28mm lens. I'm not really looking for an argument here and I'm in no way knowledgeable about 1950s film emulsions. Equally you're right my & others views on the f/11 step down on the Olympus are pure conjecture (and arguably more easily criticised on the fact that the PEN will stop down to f/22 in very bright conditions with 400 ISO film) but I guess I'm trying to make the point that some years before the PEN arrived Kingslake was talking about diffraction issues in a book aimed at the likes of you and and I and as he describes it was influencing lens design well before PEN EE arrived

Blinx replied:

Comment posted: 16/05/2016

I think you may be mistaken. Films in the 1950s and 1960s were fully capable of high resolution that is indistinguishable from modern emulsions. That is not to say there hasn't been progress, especially in colour print film and grain size in faster films, but a look at any black and white photograph of the period will illustrate that the optical quality of the lens is infinitely more important than any technological advances in film in the intervening years. On the subject of sharpness, an image is dependent on the lens formula, and the aperture used. All lenses exhibit image circles of increasing size the further they are from the focal node, but those limitations have been understood since photography's earliest days. What denotes acceptable depends on a number of variables, but primarily image size and viewing distance. The EE-2 is a middling kind of camera optically, with an aperture chosen to render an image sharply at the kind of print sizes then popular, which was usually 7 x 5" or less, and maxed out at 10 x 8". At those sizes the circles of confusion are not readily apparent and diffaction is not intrusive. Apertures of around f9 to f11 were standard in fixed focal length point and shoot cameras, but are also the apertures of choice for most street photographers using varifocal lenses at hyperfocal distances, who aim for as large a depth of field as possible before optical aberrations become obvious. I believe Olympus were adopting standard and well-worn optical practices in their half frame cameras, and the main innovation was in overall body size. Printed or projected at half the size of a 35mm negative or slide, the EE-2 will be as sharp as most 35mm cameras in its class. Similarly, a Pen F SLR will be the equal of 35mm SLRs at 50% magnification.

Alan Duncan replied:

Comment posted: 16/05/2016

Thank you. The speed point is worth highlighting - there is no quicker camera than a fixed focus and the PEN E series represents a key step for true point and shoot cameras that wouldn't be advanced until the Konica C35 AF arrived in the late seventies.

Blinx replied:

Comment posted: 17/05/2016

My position is stated above, Erik. Enprint sizes varied, but 5 x 7" - or supersize as they were often called - was a machine print option. Enthusiasts with access to a darkroom could print any size they cared for, and with suitably slow film and scrupulous development technique large prints were possible, but I don't think Olympus intended half frame as a large print format, and as you say slides were a popular option at the time. Half frame is what it says, half the size of a 35mm negative, so a half frame 8 x 5" print will resemble exactly half a 10 x 8" 35mm photograph in grain structure and sharpness. Focal length standards vary with format, obviously, and I'm assuming an equivalent lens design for comparison. Remember, 35mm is a miniature format, so half frame is sub-miniature. However if you look at Flickr's half frame sites, the quality of the best images shows it to be a useable format, so long as the highest quality equipment is used.

Carl replied:

Comment posted: 17/05/2016

I just read your posts andwhatever you state about print sizes and old film stocks, you are completely on the wrong track when you assume that in the 50s, 60s or even later they did not understand or know about diffraction. There was no "need to understand" diffraction when high resolution films/sensors came up because diffraction is NOT purely limited to photography and image taking but strictly a physical phenomenon that appears with all kinds of waves in all kinds of applications. Applied to EM waves such as visible light it is described and explained by quantum mechanics which has been put together (by probably the greatest minds of the history of humankind) leading to the Solvay Conference in Copenhagen in 1927 (!). Lord Rayleigh who defined the Rayleight criterion for image resoltion limited by diffraction (mentioned in on of the replies here) died in 1919. Now all of that dealt with and led to understanding, even hundreds of years before these events took place people observed diffraction and while they could not explain it yet it was something they were well aware of.

Erik Prestmo replied:

Comment posted: 17/05/2016

"blinx" how old are you? I don't think you was around when these cameras was designed, nor when they sold. Supersize prints was an option - towards the end of film sales and the print finishing business. But not in Japan in the late 1950's. In fact print finishing business as we once knew it (it is as good as dead in this country) hardly existed, especially as far as color prints goes. And standard sizes? Well, when I started in the darkroom, the common print sizes with the local photographer on the corner was 6x9cm, contact copies from 120 film 8 on, 6x9cm, you had to order special for "enlargements". The stadard enlargement sise, blackandwhite was 9x12, which was less than 4x6", around 3,5x4,75. But this was not som important, these cameras was designed for half-frame slides, and when shown on a screen people just accepted grain, unsharpness and error of exposure, as long as some "likeness" was to be seen on the screen. Now onto your halfframe misunderstandings. Halfframe means with the lenses that we operate by square of 2, when comparing lenses to the international standard,. A 35mm lens compares to a full frame 50mm lens, roughly A 28mm lens compares to a full frame 40mm lens roughly If they compared to what you stated originally you'd have a quarter-frame, or a pocket instamatic frame, roughly Interestingly ypu can compare this to todays digital cameras: APS-C is roughly a half-frame format and the lense lengths compares to Olympus Pen FT lenses.... m4/3 is roughly a quarter.frame and the lense lengths compares to Pentax Auto 110 lenses roughly I have both and uses them from time to time, and last summer's outing with the Pen FT made it painfully clear that half-frame negatives leaves a lot to de desired in the sharpness department, compared to pictures made at the same time with a bog standard Canon EOS camera, even when usinf the famous Olympus Pen Lenses.....

Alan Duncan replied:

Comment posted: 18/05/2016

Erik At the end of the day whether you think it was intended or not it is fortuitous that the PEN steps down at f/11 not smaller apertures for diffraction reasons. Irrespective of how you feel the science has evolved since the 50s and 60s small aperture diffraction was known about and taken into account by lens designers and is not just an issue of the last decade which is the point that I object to. As Kingslake described in the book I referenced before from 1951 a whole decade before the PEN EE, the preference for 2inch (~50mm) lenses was a minimum aperture of f/16 because of this (this book was intended for the public not lens designers so would not be cutting edge) and he highlights the problem would be worse for shorter lenses so it is a bit disingenuous to say they really weren't too knowledgeable about it (a hell of a lot of the science like Rayleigh criterion come from much earlier). Of course the science has moved on but a lot is to do with issues related to sensors.

Alan Duncan replied:

Comment posted: 18/05/2016

Know exactly what you mean with the 72 shots taking for ever. Half frame cameras are a double edged sword that way - you start out appreciating the twice as many shots for the same cost but by no. 46 you're just itching the to get the roll out. I literally found myself starting at my FED Mikron the other day muttering please let it stop ! The EE to Trip comparison fall mainly around the exposure system but you are right the EES is a more correct comparison (even Olympus admit the Trip was based on the EES)

Alan Duncan replied:

Comment posted: 08/11/2017

Goodness this is the post that keeps on giving !! The joy of the internet is folk get to express their views Jaco hopefully in constructive discussion - much of the comments here disappeared off into the history of optics science. I’m not quite sure what your critique is re films are. The original post didn’t really discuss and whilst there is a discussion of vintage Film stock most of it centred on colour films not B&W, where the development of better layer technologies in the last quarter of the 20th century did have an impact and certainly Film speeds dramatically altered between the early 1960s and 70s for colour. Degradation is more marked with colour stock which probably explains some comments about poor scanning of old images. Not sure anyone mentioned contrast etc here but your comments are valid. I like shooting C-41 B&W as it is virtually grain free to the naked eye if shot right (esp BW400CN whom Kodak marketed on this)

Hamish Gill replied:

Comment posted: 20/08/2019

Meh

Hamish Gill replied:

Comment posted: 24/09/2019

*yawn