Related Links:
How Critical Is Focusing?
Tips to improve resolution

Fifty Lines Per Millimeter

Film

Handheld

Autofocus Sensors

If you are using an autofocus camera in autofocused modes, chances are very good that you can't beat the 50 lpmm limit either.

Focusing Consistency and Accuracy

If you have a popup 2X magnifier eyepiece, you can repeat this test. In my case, the extra magnification made it significantly easier to get a more precise focusing point. But most 35mm SLR viewfinders have low magnifications (e.g., 0.72X), especially compared to medium format magnifying hoods and other vision improving techniques. A simple accessory like an eyecup to block extraneous light and the popup 2X magnifier (Nikon, Spiratone..) can be good investments in more accurate focusing and higher resolution results!

My point here is that you may have to practice focusing to develop a crisp and accurate focusing technique. Careful focusing with higher magnification aids can significantly improve your focusing accuracy. When in doubt, you can still "bracket focusing" just like you bracket exposures, especially for macrophotography.

Purists may also want to take focus shift into account by focusing at the taking aperture rather than wide open and then stopping down. Focus shift occurs when you stop down with some lenses, their focus shifts to throw a subject slightly out of focus during exposure. Large format enthusiasts routinely use a high powered loupe on a 4x5" ground glass of the image at the taking aperture, often 10X power(!), to ensure getting a truly precise focusing point. Now you know at least one of their secrets!

Zooms vs. primes

If we all used primes, we would have higher resolution performance optics than if we use zooms. But the market reality is that many primes have been dropped from sales by many manufacturers. Moreover, zooms have taken the place of primes on most new cameras being sold. So for many users, their consumer quality zoom lenses are limiting them again to the 50 lpmm range.

Processing

Similarly, amateur films are more likely to range around the optimal color points, especially if stored improperly or past optimal use dates. Pro films are usually "aged" and then frozen to preserve them at the optimal color balance points. Process quickly after use for optimal results too.

Finally, there is a problem with x-raying of film for travelers. Don't leave your film in your checked luggage, unless you don't mind a higher degree of film fogging impacting your results. Recent Popular Photography tests have shown serious impact, especially in the faster films.

Scanners

Even if you have a true 2,700 dpi film scanner, that equates to about 100 dots per millimeter (since 25.4 mm = 1 inch). It takes a line of dots and a line of clear pixels to make a line, so we are really looking at circa 50 lpmm again.

Conclusions

Autofocus cameras are inherently limited to around 50-60 lpmm, based on the discrete nature of their sensors and other technical issues. But most users will be using fast color print film with a resolution limit around 50 or 60 lpmm too (e.g., 94% of film use, see death spiral of serious amateur photography). Similarly, the lack of high quality prime lenses and substitution of zoom lenses is another loss that doesn't matter much in this scenario.

System engineers call this local optimization, but overall system sub-optimization. Spending a lot of money for high quality lenses won't greatly improve the system resolution (e.g., perhaps 3-5 lpmm improvement from going from 50 lpmm to 100 lpmm on the lens element).

On the other hand, you can see why so many folks like to use Fuji Velvia, a slow 50 ASA color slide film with an astonishing 80 lpmm resolution (for average contrast scenes of 1:1.6). It is much, much cheaper to use a higher resolution film to improve your results than to improve a good lens enough to make a similar improvement (probably impossible or prohibitively expensive with standard 50 lpmm color print film).

In other words, instead of worrying about the quality of your lenses, you should probably be worried more about what film you are using. I also recommend monopod use when you can't use a tripod over handholding. Add synchro-flash or other tricks to reduce contrast range, and a lens hood to reduce flare, and you may be surprised how much such low cost improvements can impact your results.

My point here is simply that you have to make a serious effort to beat all these factors trying to limit your results to the 50 lpmm limit range. See our page of tips to improve resolution based on similar and related points. You can beat 50 lpmm, but you have to use high resolution (low ASA) films, a tripod, good lenses, and good technique and processing!

Good shooting!

Beating the 100 lpmm Barrier

The 100 lpmm barrier is similar to the sound barrier; most of us will probably never get past this quality limit. However, see the article in October 1978 Modern Photography by Bennett Sherman and Al Gordon titled "How Sharp Can You Get" [p.112-3, table from p. 178, 182, 186]. They noted:

"The first thing which can be inferred from the tables is that the 100 lines per mm level of image sharpness can be reached by many normal focal length lenses when the aperture is set at f/4 or f/5.6, and the "best" film is used." [Ibid., p. 174]

My first comment would be that I have got to get some of that micro-Ektrachrome slide film, Kodak Photomicrography color film #2483 (PCF 135-36). Wow!  Only black and white high contrast copy film (HC) beat micro-Ektachrome #2483.  Our studies of film resolution limits suggests that color films are often the limiting factor in system resolution.  So if you are using either Panatomic X or even Tech Pan you are probably going to miss breaking the 100 lpmm barrier according to Mr. Sherman's tests. 

Similarly, you have to use a very good lens. While many of the macro lenses are highly regarded, they can't deliver the highest resolution levels needed to break the 100 lpmm barrier, according to these tests. Similarly, the faster 50mm f/1.4 normal lenses of this mid-1970s vintage could not out-perform their slower cousins. In both cases, I think we can attribute the need to correct various aberrations in both the macro lenses and the faster lenses to slightly reducing their performance against the slower and easier to correct 50mm f/1.8 and f/2 lenses. 

The top performing lens is, not surprisingly, a Leitz 50mm f/2 Summicron. But is it top performing because it is a superbly designed and optimized build quality lens (as it surely is)? Or is it the top performing lens because it is an f/2 rather than a good bit faster f/1.8 or f/1.7 lens (which are 1/3rd and 1/2 stops faster)?  While the system resolution differences are small (e.g., 86 vs. 82 lpmm on Kodachrome 25), the implied aerial resolution differences for these lenses are much higher (e.g., roughly 600+ lpmm on the Leitz aerial resolution value vs 450 lpmm on the Nikkor or Minolta optics, or about a 35% difference).  In other words, it takes a very much better aerial resolution lens value to raise overall system resolution even a few lpmm when close to the film resolution limit.

The best technique is also needed. This technique includes heavy tripods and camera dampening, mirror lockup or self-timers to reduce shutter shake, and optimal lighting of the test target.  Did we mention that they also took an entire series of photos at 1/16th of an inch lens focusing steps to ensure getting a truly sharp image? If you are just shooting one shot for your lens resolution tests, you are probably not going to hit the ideal focusing point, and so derive a rather lower than optimal value.

Actually, you can get even higher lens resolution values if you use Dr. Eugen J. Skudrzyk's techniques. His lens resolution values are roughly a third higher than similar values reported by other testers. Why? Besides using the above style techniques, Dr. Skudrzyk uses strobe lighting to help freeze camera motion effects. Understand that very, very small camera and lens movements are needed to mess up the resolution of lines only 0.01mm apart! 

"It turns out that almost every lens has a "critical aperture" at which the lack of sharpness due to aberrations is almost gone, and the image degrading effects of diffraction (scattering of light rays around the iris diaphragm) have not yet taken over completely. For most normal focal length lenses on 35mm SLRs (50 to 55mm), this critical aperture is somewhere between f/4 and f/5.6." [Ibid., p. 168]

This "critical aperture" is the beginning of the "sweet spot" of the lens where optimal performance will be achieved. 

"If you allow a pinpoint of light to be recorded as a blur of .01mm (1/100mm) diameter, and work at an aperture of f/4, then the allowable depth of focus (at the film plane is only .04mm (You may take only half of the full allowance when working at infinity). This tolerance includes all factors which can influence the final image focus, such as the natural bend of the film, tightness of the lens mounting flange or thread, accuracy of the infinity stop on the barrel, and accuracy of the distance of the lens flange and the film plane as determined by the camera body." [Ibid., p. 170]

We have an extensive discussion on the critical focus problems and related depth of focus issues which highlights these issues and their impact on achieving maximum resolution limits. A two millimeter error in focusing accuracy on the lens mount corresponds to a 0.1mm displacement from the film plane, which can cost you up to 50% of your potential lens resolution. Wow! 

"Finally, we found that among the better lens brands, the f/2 or f/1.8 lenses generally seemed to give slightly sharper images at f/4 or f/5.6 than did the faster f/1.4s offered for these cameras." [Ibid., p. 174]

In an article on "Super Speed Lenses" by Bennett Sherman and Jason Schneider in November 1978 Modern Photography (p. 104), they noted (on p. 168) that the wide open image quality (a combination of resolving power measured in lpmm and contrast measured as percentages) for fast lenses did not merit any excellent ratings, and only the aspheric Canon 55mm f/1.2 rated a very good in the corners. So we should not be surprised that the tradeoffs in aberration and other optical design issues for these faster lenses results in slightly poorer performance even stopped down to f/4.  On the other hand, this is good news if you like to use the 50mm normal lens of average speeds often disdained by other photographers. 

 

Lens:at f/4, infinity	Pan X	Tech Pan	High Contrast Copy Film	Kodachrome 25	#2483 Micro Ektachrome (PCF)
Summicron  Leitz 50mm f/2	88 lpmm	96 lpmm	105 lpmm	86 lpmm	102 lpmm
Nikon 50mm f/1.8	88	96	105	82	100
Canon 50mm f/1.8	86	92	102	80	100
Minolta 50mm f/1.7	86	92	102	82	100
Pentax 50mm f/1.7	86	92	102	80	100
Olympus 50mm f/1.8	86	92	102	80	100
Macrolenses at f/5.6 & 1/8X magnification
MicroNikkor  55mm f/3.5	70	88	92	76	88
S-Planar 60mm f/2.8	72	92	96	82	92
Vivitar 55mm f/2.8	70	86	90	78	88
Minolta 55mm f/3.5	70	86	92	78	88
Olympus 55mm f/3.5	70	86	90	78	88

Related Postings

Date: 27 Dec 1999
From: kartboad@pc.highway.ne.jp (Hiroaki YOSHIFUJI)
Newsgroups: rec.photo.equipment.35mm
Subject: Re: Leica story

....

  | > Only if you are talking about 400 speed or
  | > higher film, or handheld lower film.
  |
  | Handheld shots includes an extraordinarily large chunk of photography.  For
  | every person I see with a tripod, I see at least 100-200 without.

I think a person using 2700dpi scanner cannot see more than 100. According to , 50 line/mm is the max of 2700dpi.

Hiroaki

[Ed. note: Claude Shannon's information theory work is meant, I believe. Nyquist determined that you have to sample at least twice as often (cf. frequency) as the signal frequency (cf. lpmm) to ensure accurate sampling... ]

Date: Tue, 12 Oct 1999
From: RedDrake@aol.com
Newsgroups: rec.photo.equipment.35mm
Subject: Re: Leica, Contax or Nikon???

shadcat11@aol.com (ShadCat11) wrote:

> Day before yesterday I took a borrowed Contax G2 for a walk around the block
> with my Nikon FM2 with equivalent 28, 45/50 and 90/85 mm lenses for a slide by
> slide comparison.  ...

Unless you wandered around with a tripod and used flawless technique, handholding (and maybe the film) could act as a bottleneck for sharpness. Even an absolutely perfect lens (and of course there aren't any) is only as good as the steadiness of the operator, the quality of the film, and the operator's technique. Maybe you can't tell the difference under these conditions?

Which is not to say that Nikon lenses are inferior, just a comment on the quality of the experiment. Frankly, I have a hard time telling what camera took what slide if they're not marked and I use lots of different cameras (for fun, I'm not a professional photographer).

Date: Sun, 10 Dec 2000
From: John Owlett owl@postmaster.co.uk
To: nikon@photo.cis.to
Cc: Robert Monaghan rmonagha@post.cis.smu.edu
Subject: Rabbit and Owl counts megapixels

"... which clearly shows," said Rabbit, "exactly how many angels can dance on the head of a pin, AND how many of them will be in focus."

"Oh, well done, Rabbit," said Kanga, perhaps a little wearily.

It was a languid early evening, and the animals (even Owl, who had only just woken up) had gathered at the sunny end of a clearing in the Hundred Acre Wood. Each was engaged in a favourite pastime. Rabbit was talking, Pooh was humming, Piglet was lying back with his eyes closed -- so that nothing would distract his listening to Rabbit -- and Tigger was bouncing around in the hope that nobody would realize that it was bedtime for him and for Roo.

Eeyore, who had been having a wonderful time feeling gloomy, added, "Well done, indeed, Rabbit. Nobody could have gone into more detail on the subject. Even if they had wanted to. Which I doubt. But I thought that Owl and you were supposed to have been counting megapixels."

"Oh, we did that too," said Rabbit. "Ever since the Great Alexander posted the estimate that a digital camera would need 54 megapixels to match a medium speed 35mm film; and the Remarkable Rob quoted a source as saying that 1200 pixels per inch -- 1.9 megapixels in all -- would be enough."

Piglet sat up suddenly and said, "But surely that shows how irrelevant these numbers are! If two reputable sources can come to such different answers what's the use in counting megapixels?" He felt very daring, disagreeing with Rabbit like this.

"Well, yes, we knew that there are many who share your opinion, Piglet," responded Rabbit, "which was why I gave you all the pinhead results first. But we do believe we have some useful megapixel results too."

"Oh, that's all right then," said Piglet. "I'll just lie back and close my eyes again, so that I can concentrate better on what you are saying."

"But I thought you'd reached a rather simple conclusion," said Kanga. "Didn't you, Owl?"

"A quarter past six," said Owl. "Coffee."

So Kanga decided she'd summarize the results quickly, before Owl had a chance to wake up fully:

*************************************************************************
*
* THE SHORT ANSWER ...
*
* ... is that ordinary careful amateur photography can record about
* 6 megapixels on a frame of 35mm film.  Once 6 megapixel SLRs are
* available with a reasonable speed -- oh, say ISO 400 -- and at a price
* that's hundreds of pounds rather than tens of thousands of pounds, we
* can reasonably expect amateur photographers to go digital in droves.
*
*************************************************************************

Piglet sat up again. He felt Very Strongly Indeed about this. "Not droves," he said. "You have cattle drives, so it's reasonable for cattle to do things in droves. But who ever heard of a pig drive, or a drove of owls? You have a parliament of owls, not a drove."

"Mum," said Roo. "Why do we call a groups of owls a parliament?"

"Well," said Kanga. "A parliament is where politicians get together to talk about laws and government policies."

"Yes," replied Roo.

"Do you know anybody who does more talking than Owl?"

Pooh stopped humming and thought for a bit. Six million pixels was an awful lot. How could anybody count six million of ANYthing without making a mistake? He wondered whether he should ask for a hand recount, but then decided that That Would Not Help. Some of the American gray squirrels in the Hundred Acre Wood were very sensitive about that sort of thing.

Anyway, Rabbit had started again.

"Well, yes, that's the short answer," he said, "but it misses out a lot." So he carried on:

*************************************************************************
*
* THE MEDIUM ANSWER
*
* How many pixels you can fit into a frame of 35mm film depends on how
* many line pairs per millimetre you can achieve with your photography.
*
* The Great Alexander gave, as a reference for his 54 megapixel number,
* a paper by William Oliver.  Now some parts of the paper are hard to
* to understand, but the 54 megapixel number isn't.  On a high contrast
* subject, a medium speed film like Ektachrome can resolve 125 lp/mm.
* Now, it doesn't matter whether your test chart has 125 black lines
* on a white background, or 125 white lines on a black background:
* you need 250 pixels per millimetre to record 125 line pairs.
*
* So a frame of Ektachrome film can hold 36mm x 250 pixels/mm one way,
* and 24mm x 250 pixels/mm the other way.  That's 9000 pixels by 6000
* pixels, or 54 megapixels in all.
*
* ("Hmmm," said Pooh quietly to Piglet, "and that's the EASY part of
* the paper?"
*
* "Zzzz," said Piglet.)
*
* There's nothing wrong with the mathematics.  The only problem is that
* so few of us specialize in photographs of test charts.  Dr Oliver may
* have high contrast subjects in his pathology lab, but the flowers and
* fungi in the woods are all low in contrast.
*
* Fortunately, we can turn to Monaghan's Magnificent Megasite at
*
* http://www.smu.edu/~rmonagha/mf/limits.html
*
* to find out what can be achieved.  Monaghan the Magnificent points out
* how very hard it is to exceed 50 lp/mm: possible but very hard.
* Ordinary careful amateur photography tends to be more like 40 lp/mm:
* by a similar calculation to the one above a frame of 35mm film can then
* hold 36 x 2 x 40 x 24 x 2 x 40 pixels, or 6 megapixels.
*
* ("Ouch!" said Pooh.)
*
* There are some films that allow higher resolutions on ordinary subjects
* ... provided you have the tripod, the technique, and the lenses to take
* advantage of them.  They are the Kodachromes (which were way ahead of
* their time) at 65 lp/mm, Velvia at 80 lp/mm, and Technical Pan at 100
* lp/mm.
*
* So, instead of the simple 6 megapixel figure of the Short Answer, we
* have different figures for photography at 40, 60, 80 and 100 lp/mm:
*
*    to match normal amateur use of 35mm,         you need  6 megapixels
*    to match meticulous use of Kodachrome,       you need 12 megapixels
*    to match perfectionist use of Velvia,        you need 22 megapixels
*    to match perfectionist use of Technical Pan, you need 35 megapixels
*
* Translating this into 35mm film scanner terms (1 lp/mm = 50.8 DPI):
*
*    to match normal amateur use of 35mm,         you need 2000 DPI
*    to match meticulous use of Kodachrome,       you need 3000 DPI
*    to match perfectionist use of Velvia,        you need 4000 DPI
*    to match perfectionist use of Technical Pan, you need 5000 DPI
*
*************************************************************************

"I have a bad feeling about this," said Eeyore. "We've had the Short Answer and the Medium Answer. I have a sense of dread that there is a Answer and the Medium Answer. I have a sense of dread that there is a Long Answer -- possibly a Very Long Answer -- as well."

"Of course," said Rabbit. "Now that Owl is fully awake, I'm sure he'd like to take over."

Owl fluttered up to his favourite lecturing branch, but before he could begin ...

"Goodness, is that the time?" said Kanga hurriedly. "The smaller animals have to go to bed, and they mustn't miss Owl's interesting talk, so we'll have to stop for this evening. Don't you agree?"

"Zzzz," said Piglet.

--------------------------
John Owlett, Southampton, UK

[Ed. note: see Erwin's Posting on gigabit film tests...]
From Leica Mailing List:
Date: Tue, 2 Jan 2001
From: "Anderson, Ferrel E" AndersonF@ria.army.mil
Subject: [Leica] Exploring the limits

I explored the limits with my new Leicaflex (Mark1) and 50 Summicron and 90 Elmarit lenses in 1968. Like the testers in Camera 35 magazine, I used the United States Bureau of Standard lens testing charts and directions, but instead of using Pan-X film, I used Kodak High contrast Copy film and H&W; Control developer. I tested the lenses at 76 focal lengths, which provided a maximum possible resolution of 240l/mm, or 120lp/mm. Results:

The highest resolution was achieved with the Summicron at f4 in the center of the image, where the resolution was an astounding 240 l/mm (120lp/mm)! I could only see this resolution with a microscope. When I put the negative in the focomat 1C and checked resolution at 8X and at f4.5 with my Focotar, I could not see this level of resolution, but not because of the quality of the lens, but because of the low level of magnification. At 16x, I could just barely make out the resolution lines. My conclusion: just like Erwin, there is a point of diminishing returns, and for normal superb quality, 80l/mm (40lp/mm) is certainly it. Ctein states that about 5 lp/mm in the print is needed to get satisfactory sharpness, but one needs 30 lp/mm on the paper to achieve optimum sharpness - read resolution of the image structure = grain. This can be achieved with the best enlarging lenses.

The best resolution across the field was at f5.6 with both lenses, and at this aperture resolution across the filed was from 160 to 200+ l/mm (80-100 lp/mm). I tested some other normal lenses, and the best was the 50mm f1.4 Nikor, which achieved 160 l/mm (80lp/mm) across the field at f5.6. To achieve uniform performance across the field. all of the lenses I tested should be stopped down to f8- f9.5. At this aperture diffraction evens out optical performance, and at f11 performance levels out at about 68-80 l/mm (34-40 lp/mm) but with reduction in contrast.

More recently, I checked my lenses with the USBS charts at 26 focal lengths which results in a maximum possible resolution of 80 l/mm (40 lp/mm). the best lenses in my current stable are the ancient 90 Elmarit, the 60mm Macro Elmarit - -R and the100 APO Macro Elmarit-R. the APO lens is the best lens I have tested. The resolution lines are etched on the negative right across the field from f2.8 until diffraction takes its toll. I plan on testing all of my lenses later this winter with Tech Pan film, Technidol developer, and at 76 focal lengths. I'll share my results with you.

Ferrel Anderson

From Leica Mailing List;
Date: Fri, 02 Mar 2001
From: imx imxputs@knoware.nl
Subject: [Leica] Killing two myths in one post!

The filter myth.

degradation by a filter: good quality--> 2%, not good quality-->10%.
degradation by handholding below 1/125: 50%
degradation by (slight) defocus: 30 - 80%

.....

Erwin

From Leica Mailing List;
Date: Mon, 30 Jul 2001
From: "Mxsmanic" mxsmanic@hotmail.com
Subject: Re: [Leica] Some lens tests

A fascinating test. I'm especially amazed by how poorly the Nikkor 1.8/50 performs at all apertures. I actually have this lens, I think, somewhere, but I don't use it anymore; after seeing this, I'm not tempted to start using it again!

I'm also somewhat surprised by how close the lenses are in the 90mm test. The Leicas are clearly superior at the other focal lengths, but at 90mm it looks like a wash to me.

One other thing I noticed that is significant: none of these lenses, at any aperture, is resolving more than about 40 lp/mm. Indeed, if I downsample your actual-size scan to the size that would have been provided by an LS-2000 (with a maximum resolution of 2700 dpi), then upsample it again to the size provided by the 4000, the result looks identical to the original. In other words, the resolution of the image is no more than 53 lp/mm, and probably considerably less, around 40 lp/mm. It may be a film limitation, since the Summicron 90, at least, is supposed to be able to do far better than this (around 100 lp/mm, even out at the edges of the field, and even wide open). Provia starts to fall off pretty rapidly at 50 lp/mm, if I remember correctly.

Yet more evidence that, for most purposes, you need never go beyond 2700 dpi. And the fact that this was all done on a tripod just underscores this point. And in the case of the Nikkor 50, even a hand scan of a one-hour print would probably extract just about all it can manage!

I don't see any Leica Glow (tm), either.

- ----- Original Message -----
From: "Paul Chefurka" Paul_Chefurka@pmc-sierra.com
To: leica-users@mejac.palo-alto.ca.us
Sent: Monday, July 30, 2001
Subject: [Leica] Some lens tests

> I just finished putting upo some lens test pages that I hope will be  more interesting than controversial :-)
>
> Leica lenses: 28 Summicron, 35 Summilux ASPH, 50 Summicron, 90 APO  Summicron, Tri-Elmar
> Nikon lenses: 50/1.8 AIS, 55/2.8 Micro, 85/2.0 AI
> Cosina lens: Voigtlander 90/3.5 APO-Lanthar.
>
> Mostly shot at two apertures, 4.0 and 8.0.  A complete description of  the test methodology is on the opening page.  Have at it, folks.
>
> http://members.home.net/chefurka/Photo/LensTests/LensTests.html
>
>
> Paul Chefurka

From: "thc" a2toothfairy@mw.mediaone.net>
Newsgroups: rec.photo.technique.misc,rec.photo.technique.people,rec.photo.equipment.medium-format
Subject: Re: Technique of slow shutter hand-holding
Date: Mon, 05 Nov 2001 

An ability to hand-hold a camera during slow shutter speeds varies with
respect to the particular individual and the particular shooting conditions.
Most photographers have good days and bad days for their hand-holding
ability.

Why not use a short roll (12-exposures) of film to test for your own
abilities?

Choose your own combination of film and lens, focus onto an unmoving target,
make a series of tripod-supported exposures and a series of hand-held
exposures at varying shutter speeds, and then analyze the results of your
own ability.

For your information:

Popular Photography (US) magazine (volume 63, number 3, March 1999, page 17)
featured an article about hand-holding versus using a monopod support.   Six
photographers used the same Canon "EOS Elan IIe" 35mm format SLR camera with
a 100mm-400mm f/4.5-6.7 zoom lens set at 400mm, focused at a mannequin head
at 26.5 feet distance, using Kodak MAX 800 color-print film.   A series of
monopod-supported exposures (at 1/8 second, 1/15, 1/30, 1/60, 1/125, 1/250
second) and hand-held exposures (same series, from 1/8 second - 1/250
second) were made (probably using short 12-exposure rolls of 35mm film), and
8x10-inches enlargements were made of the negatives for comparison.

- Results:
- The steadiest on that particular day of the six photographers was blurry
at 1/8 and 1/15, slightly blurry at 1/30, and sharp at 1/60, 1/125. and
1/250 second when hand-holding; was slightly blurry at 1/8 and 1/30, sharp
at 1/15, 1/60, 1/125, and 1/250 second when monopod-supported.
- The average of the six photographers was very blurry at 1/8 and 1/15,
slightly blurry at 1/30, and sharp at 1/60, 1/125, and 1/250 second when
hand-holding; was slightly blurry at 1/8, sharp at 1/15, 1/30, 1/60, 1/125,
and 1/250 second when monopod-supported.
- The poorest of the six photographers was extremely blurry at 1/8 and 1/15,
slightly blurry at 1/30, and sharp at 1/60, 1/125, and 1/250 second when
hand-holding; was very blurry at 1/8, blurry at 1/15, slightly blurry at
1/30, and sharp at 1/60, 1/125, and 1/250 second when monopod-supported.

In summary, these six photographers all were judged to be sharp at 1/60,
1/125, and 1/250 second exposures regardless of hand-holding or
monopod-support of a 400mm lens on a 35mm SLR camera body.  The 1/8 second
exposures by the steadiest photographer were judged to be either blurry
(hand-holding) or slightly blurry (monopod-supported).  The 1/8 second
exposures by the poorest photographer were either extremely blurry
(hand-holding) or very blurry (monopod-supported).  The 1/15 and 1/30 second
exposures were judged to vary from sharp (at 1/15 second, monopod-support,
by steadiest photographer; at 1/30 second, monopod-support, by average
photographer) to slightly blurry (at 1/30 second, both hand-held and
monopod-supported) to extremely blurry (1/15 second, hand-held, by poorest
photographer).

Another article in Popular Photography magazine (volume 63, number 6, June
1999, page 18) confirmed George Lepp's advice that the mirror-lockup feature
on a SLR camera body used with tripod-support generally gives more sharpness
at 1/8 - 1/60 second shutter speeds than not using the mirror-lockup with
tripod-support.

In short:
Monopod- and tripod-support generally outperform hand-holding in the 1/8
second - 1/15 second -1/30 second exposure range (no big suprise here), but
even a tripod-supported SLR camera gains additional sharpness (up to a 227%
gain in sharpness, when measuring line pairs per millimeter resolution) when
the mirror-lockup feature is utilized in the same exposure range.

> Erich Salomon pioneered available light candid photography in the late
> 1920's with the Ermanox.  He often needed 1/4 sec. or longer exposures,
> and initially couldn't use a tripod since he needed to remain incognito.
The (Ernemann A.G. Dresden) Ermanox plate camera of 1924 had either an f/2
100mm or an f/1.8 85mm Ernostar lens, and offered 1/20 second - 1/1,000
second shutter speeds (as described in Rosenblum's "A World History of
Photography," and in McKeown's "Price Guide to Antique & Classic Cameras"),
so Salomon probably didn't have available 1/4 second exposure times with
that particular camera. Maybe the unsharpness and blurriness inherent to his
longer shutter speeds were camouflaged somewhat by the larger film format
(4.5 x 6 cm) of his camera. [Also, I'm uncertain about what size were
printed from Salomon's plates. A 4.5-inch x 6-inch print would be about 2.5x
diameter enlargement. A 35mm frame requires more enlargement for the same
size print.]

> Sure the critical eye will see that some of Weston's famous portraits
> are not critically sharp, yet they're quite good and easily acceptable
> to 8x10.
snip>
> I'll certainly brace myself as often as possible in low light
> situations, but what other tricks, such as implied above, are helpful?
>
It has been shown that tripod-support and mirror-lockup are the gold
standard for truly sharp 35mm SLR photographs.
Thus, any technique not utilizing those two features will result with
photographs deficient in some degree of sharpness.
You state that Weston's unsharp prints are acceptable to you.

So, the question actually reduces to:
What degree of unsharpness are you willing to allow your prints to exhibit?
[Robert Capa's blurry photographs of 1944 D-Day invasion of Normandy exhibit
a captivating subject matter that overcomes their technical problems.]
The correct answer needs you to experimentally determine an exposure
technique that is acceptable to you.
_________


mark@riparia.org says...
> Edward Weston made many of his 4x5 Graflex shots hand-held at 1/10 sec.
> Erich Salomon pioneered available light candid photography in the late
> 1920's with the Ermanox.  He often needed 1/4 sec. or longer exposures,
> and initially couldn't use a tripod since he needed to remain incognito.
> Today I see a photo of him holding his camera against his torso with his
> left hand and holding a cable release with his right hand.
>
> Sure the critical eye will see that some of Weston's famous portraits
> are not critically sharp, yet they're quite good and easily acceptable
> to 8x10.  How many of use can say the same with 1/10 sec. handheld
> exposures?
>
> I've never thought of using a cable release while hand holding.  So, I
> wonder...
>
> Can holding the camera against the chest or abdomen be a more stable
> position than braced against the face?
>
> Is waistlevel viewing inherently more stable?
>
> Is a cable release while hand holding in such conditions advantageous to
> an experienced gentle squeeze on the shutter button?
>
> I'll certainly brace myself as often as possible in low light
> situations, but what other tricks, such as implied above, are helpful?
>



From: bobhickey@webtv.net (Bob Hickey)
Newsgroups: rec.photo.equipment.35mm
Subject: Re: Cognitive Dissonance in camera preference
Date: Mon, 5 Nov 2001 

        Yeah, I see that, but my point is:  I have a Hassy and a Rollei
both with Planars, and I couldn't say which was better,  in spite of
being $1800 apart in price. More importantly, I just don't feel the need
to. I just bought a broken X7, I think, any way, I bought it 'cause I
like black, for about $15, and took it apart. Now it works perfectly. To
me that's exciting, and Sunday I'll know if it's any good. But if the
thing is no good, I'm not going to curse Minolta, or swear I'll only buy
Pentax from now on. Win or lose, I enjoyed myself for $15.
As far as brand loyalty goes, I think that's more a case of laziness
mixed with muscle knowledge ( spine mind ) overtaking the mental
processes. People just naturally don't enjoy taking chances or trying to
learn something new. Suppose they fail? Or are slow, or look foolish?
So, I guess the real question is; would I buy an M6, if I had the money?
In a hot minute.                                                Bob
Hickey

http://photos.yahoo.com/rollei711



From leica mailing list:
Date: Thu, 10 May 2001 
From: Jim Brick jim_brick@agilent.com
Subject: [Leica] Re: Facts revisited

It seems to me (logically speaking) that if a perfectly focused
(non-defocused) lens registers 100 lp/mm while on a tripod and only 20
lp/mm while hand held - to equal a defocused lens while on a tripod - if
you hand hold the defocused lens (already at 20 lp/mm) you simply
exacerbate the problem to worse performance.

You certainly cannot get "better" performance than you started with. And
you certainly can make it worse.

A camera that is NOT nearly dead-on in focusing on the film plane, is
simply going to, in all cases, give you WORSE performance than a camera
the IS dead-on in focusing on the film plane.

Why did Contax make the vacuum film plate? Why do MF and LF photographers
constantly whine about film flatness?

It is because the depth of focus parameter (where the lens focuses in
relation to the film plane) is extremely critical. There is NO fudge 
factor here.

And all of you Leica folks (me too) love to use your lenses wide open. f/2,
f/1.4, and f/1.0 . Without precise focus ON THE FILM PLANE, you will get
crappy photographs.

You can get lucky. You can be off in your focusing and end up with the
subject actually "in focus." But don't rely on luck. Make sure your
equipment is built to do the job the way it is supposed to be done. This is
where heritage and lineage plays an important role. The best fast precision
lenses, on the best precision body. Pieces that actually were made for 
each other. An M3 lens works perfectly on an M6. And a new 90 APO/ASPH lens
works perfectly on an M3. Heritage, lineage, history, and dedication is
what makes it happen.

So do you all still think that the flange to film spec, given to Erwin by
Konica, is an error in engineering or done for a reason? Why won't Konica
mate a Leica lens to a Hexar? Why have people attempted to have this done?
I personally think this is a self answering situation.

Jim

 brougham3@yahoo.com wrote:
>
>So what's wrong with a visual inspection, then?  If you can't see any
>degradation, does it matter if it is present?  From a practical standpoint,
>what's it matter if the theoretical limit is 100 lp/mm for a photographer
>who hand-holds 95% of the time?
>
>I ask because I seek your expert opinion, Erwin.  I'm genuinely curious.
>Does the defocus problem only really matter if you're using a tripod, slow
>film, and trying to get close to the theoretical maximum?  If hand holding
>already degrades the image, how important does the mismatch between Leica
>lenses and Hexar bodies become?
>
>Thanks!


From leica mailing list:
Date: Thu, 10 May 2001 
From: Erwin Puts imxputs@ision.nl
Subject: [Leica] Picture taking technique

"brougham3@yahoo.com" (I am sorry that I cannot address you by your
first name as it is not stated) asked in part:
> So what's wrong with a visual inspection, then?  If you can't see any
> degradation, does it matter if it is present?  From a practical
> standpoint,
> what's it matter if the theoretical limit is 100 lp/mm for a
> photographer
> who hand-holds 95% of the time?
>
> I ask because I seek your expert opinion, Erwin.  I'm genuinely curious.
> Does the defocus problem only really matter if you're using a tripod,
> slow
> film, and trying to get close to the theoretical maximum?  If hand
> holding
> already degrades the image, how important does the mismatch between
> Leica
> lenses and Hexar bodies become?

I am searching for a way to express my thinking that is clear and
consistent. What I tried to say here is this.
With careful technique you can extract maybe 80% of the potential image
qualities of a film/lens/body combination. With a technique that
necessarily introduces severe image degradation  (defocus, hand held
shooting below 1/250), your results will by default much lower.
When you use equipment that has its own sources of degradation (large
tolerances, engineering mismatch between components) you will also
automatically get a degraded image (that is compared to the one you
could produce when all components are spot on).


 From a practical point of view it seems indeed irrelevant which sources
cause the degradation. In both cases you get at best 30% of the feasible image quality.
And if you are happy with the resultant quality, you are right not to
worry what is the cause of the trouble, as you do not seem to have any
concern at all. You like what you get, irrespective of what you might
get if degrading factors are excluded.

My point is that there is a vast and significant difference between the
use of  the method of visual inspection in order to settle the question
whether a photographer is satisfied with the result and the use of this
method to settle the question what is the best image quality attainable.
If you are only interested  in creating pictures you personally like and
are comfortable  with, the method of of subjective assessment (or visual
inspection) is perfectly viable and sound.

If you are interested in getting the best possible results with your
equipment and want to know how to handle the many factors that do
degrade the image, you need a method that can identify the possible
sources of image degradation and a strategy how to cope with this.

It is for this approach that it is of paramount importance to
distinguish between user induced defocus errors and manufacturing induced  defocus errors.
You cannot dismiss this difference by claiming that the result might be
identical. In the first case I can do something about it, in the second
case I am lost.

And I am sure it is of interest to users to know what does cause the
image degradation. In some situations it may be irrelevant as you have
no options or do not care about improvements.

But if you wish to improve your image quality, it does matter.

It may be true that when taking pictures at 1/30 any defocus erors will
hide behind the degradation induced by movement degradation. Or the
other way around. But if I know it is an error of movement, I can
improve the result.

To give a direct answer to your last question: the defocus error
(mismatch Leica lens/ Hexar body) is an error that is constant. It will
be more of a problem if the shutter speed and the speed of the
filmemulsion get faster. Even without a tripod I can get quite vibration
free (satisfactory)  pictures at 1/250 or when using flash, but the
defocus error will have a significant impact. At least at the wider
apertures.

THis is the old trade-off: use a slow shutter speed and a lens well
stopped down and you cover up focus problems and generate vibration
problems. Use a higher speed and a wider aperture and you need accurate
focussing and a well corrected lens.

Any photographer has to specify his/her personal trade off in relation
to his/her demands on pictorial quality. In my view (and that is why I
choose Leica products) this choice/demands should not be limited by
issues of limitations induced by equipment performance.

But is perfectly reasonable to match your  requirements to what the
equipment is able to deliver.


Erwin


From leica mailing list:
Date: Thu, 10 May 2001 
From: henry henry@henryambrose.com
Subject: Re: [Leica] Picture taking technique

Erwin wrote in part:
>But if you wish to improve your image quality, it does matter.
>
>It may be true that when taking pictures at 1/30 any defocus erors will
>hide behind the degradation induced by movement degradation. Or the
>other way around. But if I know it is an error of movement, I can
>improve the result.

Here is another reason for using the best gear you can get even for
handheld photography.

When handholding the camera at slow speeds sometimes you are quite still
and other times you are moving the camera a lot. When you use a
camera/lens combination that can put 30% of the possible image quality on
the film - thats the best you will ever get. When you use a system that
puts 80% of possible quality on the film you will sometimes get the full
80% even handheld! Not always but sometimes. Everytime you press the
button you have the possibility of that elusive 80% with a 30% rig you
have only the possibility of 30%.

Its like comparing target rifles. One is capable of putting all its shots
in the X ring (the very center), the other capable of putting all its
shots in the 8 ring. Which one would you pick to shoot in a match?

Why use lesser gear if you are interested in image quality?
If you don't care, save yourself a lot of money and trouble and buy a
point & shoot.

Henry



From leica mailing list:
Date: Mon, 02 Apr 2001
From: imx imxputs@knoware.nl
Subject: [Leica] Myth and anti-myth

It is remarkable that the idea that there is a significant trade-off
between high contrast and low resolution still rides high in Leica lore.
As far as I know no one who holds his view has ever presented demonstrable
evidence or corroboratable measurements to prove this point. Generally a
high contrast implies a high resolution and the other way around. It may
be that a shift in focus plane may change this relationship to a small
degree, but the general correlation is evident. More contrast is higher
resolution. And statements to the effect that a "slight" reduction of
contrast brings a "slight' improvement of resolution beg, nay scream for
evidence.

Now to kill two more myths. Sometimes I feel like Buffy the Vampire 
Killer.

I have the Kodachrome films which I used as comparison for the 100 to
400ISO slide film test some weeks ago. Results will kill some preconceived
ideas. The King of all slide films is by now the Kodachrome 64, which
resolves easily 90 lp/mm, much more than the E100SW and even close to the
resolution of TP in normal circumstances. Especially noteworthy is the
excellent acutance, the great clarity of detail and the fine grain. A
disappointment was the K25 which at best was as good as the K64, with a
small gain in grain smallness, but not enough to offset the drop in speed.
The fading out of the K25 then is sensible. No added value. Sorry.

Big surprise the K200, which showed as expected a tight but visible grain
pattern, but a resolution that beats the Provia 400F at 70 to 75 lp/mm. So
the idea that fine grain supports high resolution is as false as the idea
that low contrasr supports resolution.

If you want to test the qulaity of your lenses, there is only one easy
way: use K64! and even K200 will show the defects of most lenses. Do some
actual testing!

I also had the opportunity to test the surfaces of filters on an
interferometer. Results will kill another myth. I used four different BW
filters in several colours (not relevant for testing, but to show that
there must be different batches).

Results? Take a deep breath: NO, absolutely NO image degradation by the
filter as all surfaces of the four filters were absolutely plane to the
highest possible degree. At worst only one interferometer stripe for the
experts.

Of course secondary reflections are possible. But the commonly held notion
that the addition of the filter adds two surfaces and by that fact should
degrade the image quality is simply not supported by measurements.
A well made filter in front of the lens will NOT make a drop of image
quality!

These results show that myths are fine if you wish to cling to stories that
seem sensible because they are repeated over and over again and even have
been 'explained' to some degree. But so the flatness of the earth had its
followers and scientifically based stories. But only measurements bring 
the facts.

Erwin


From: "george jiri loun" george.jiri.loun@urbanet.ch
To: Camera Makers Mailing List:
Subject: Re: [Cameramakers] Re Glass platen etc
Date: Sun, 21 Oct 2001 

Hi Brian, I wonder if in your search for John Shaws' sharpness you realize
that actually you compare apples and oranges...  You look at the pictures
in his "Close-ups in nature" (an excellent book, by the way) and compare
them with your prints (or slides?). But don't forget that pictures printed
in a book are always more contrasty and look therefore sharper than
pictures "printed"  on a photo paper!  You will never see a paper photo
picture as "sharp" as a picture printed in a book.  I remember my shock
when I saw for the first time my pictures sent to photo agencies printed
in books - how much "better" they were! Well, it's just the contrast ratio
which is higher on a printed page than on a photo paper. As simple as
that.  If you had your picture printed on a book paper maybe you would
have the same sharpness as that of John Shaw...  Just to help your
frustration...

George



[Ed. note: Thanks to Alex for sharing these tests of filters and impacts on photography...]
Date: Mon, 1 Apr 2002 
From:  alexanderkoz@excite.com
To: rmonagha@post.cis.smu.edu
Subject: Re: thanks for note: re: blind filter tests

Hi Robert!

This is the table I promised. I dropped data on filters geometry, assembling
quality, "freedom of movement" and alike which affected the overall ratings.

I think a blind filter test would be a most interesting effort too

Well, not every German filter got SUPER in the fM test below so the results
look quite "objective" for me. Minolta and Hoya suffer most from the test
although 160 lpm and medium contrast sounds more than enough to be honest.

Best regards,
Alex.


Circular Polarizing Filters
 * fotoMAGAZIN (09/2001)
 Tested with 105mm lens, 400 lp/mm

          Model                 Resolution Contrast Rating
          B+W MRC Slim          300 lp/mm  high     SUPER (5)
          Canon                 400 lp/mm  high     SUPER (5)
          Cokin P164            300 lp/mm  soft     NOCH GUT (2)
          Contax MC             400 lp/mm  high     SEHR GUT (4)
          Doerr Danubia         useless    useless  WENIGER GUT (1)
          Hama HTMC Silver      400 lp/mm  high     GUT (3)
          Hama HTMC Wide        400 lp/mm  soft     SEHR GUT (4)
          Heliopan SH-PMC Slim  400 lp/mm  high     SUPER (5)
          Hoya Pro 1            160 lp/mm  soft     NOCH GUT (2)
          Leica                 400 lp/mm  medium   SEHR GUT (4)
          Minolta II            160 lp/mm  medium   NOCH GUT (2)
          Nikon                 400 lp/mm  medium   SUPER (5)
          Pentax                ??         ??       SEHR GUT (4)
          Rodenstock Jet-Pol    400 lp/mm  medium   SEHR GUT (4)
          Soligor               400 lp/mm  medium   GUT (3)
          Tiffen                300 lp/mm  middle   GUT (3)

MYTabs
ak (2002)


From manual SLR mailing list:
Date: Wed, 23 Oct 2002 
From: "R.C.Booth" r.c.booth@att.net
Subject: Re: [SLRMan] the 100 lpmm lens you may already own ;-)


Just to add a bit more:  I don't think the image quality of a picture taken
with a 50mm K mount (f2) lens is much less than the earlier screw mount
variety.  The best thing is that, for someone on a budget, they can aquire
some quality equipment that is virtually equal to the "best" if they're not
worried about prestiege.

RCB
----- Original Message -----
From: "Robert Monaghan" rmonagha@mail.smu.edu
To: SLRMAN@topica.com
Sent: Wednesday, October 23, 2002 
Subject: RE: [SLRMan] the 100 lpmm lens you may already own ;-)

> regarding the pentax vs. nikkorex "upsale": ;-)
>
> see http://medfmt.8k.com/mf/limits.html#100 - table shows a number of
> brands of 50mm f/1.7-ish lenses readily reach 100 lpmm in resolution tests
>
> the older pentax lenses were only second to leitz in Dr. Skudrzyk's
> tests of lens aerial resolution (in photography for the serious amateur);
> this is why it was no surprise to many of us that at f/8, the pentax 50mm
> f/1.4 equalled the new leica 50mm f/2 at f/8; viz.
> http://medfmt.8k.com/mf/rangefinder.html#keppler
>
> basically, we are a lot more limited by our technique than by our lenses;
> see also http://medfmt.8k.com/third/max.html#handhold on degradation % ;-)


From: "Q.G. de Bakker" qnu@worldonline.nl
Newsgroups: rec.photo.equipment.medium-format
Subject: Re: why MF won't get better ;-) Re: What is depth of focus?
Date: Sun, 10 Nov 2002 


Paul King wrote:

> Possibly dumb question... I have it in my brain that F16 gives 'better'
> resolution than wider F-stops. Is this not the case, and if so, what is
> the 'best' F-stop for the maximum resolution, regardless of DOF. Are
> there other 'variables' involved?

Yes, there are. Aberrations.

In a perfect lens, free from all aberrations, highest resolution is at
widest apertures, with maximum resolution approximately halving every two
stops you close the aperture down due to increased diffraction.

In real-life lenses resolution isn't limited by diffraction alone, but by
residual aberrations too. Stopping down will gradually decrease the effect
of aberrations, until the state of maximum correction possible in a given
lens design is reached.

The f-stop at which this happens however may well be beyond the point where
both aberrations and diffraction are equally bad.

The "sweet spot" of a lens, the "best" f-stop is the one where stopping down
has just led to a reduction of the ill effects of abberations below the
level of diffraction. Stopping down more will perhaps reduce aberrations
further, but you wouldn't notice since the increased degradation due to
diffraction is greater.

Where the "sweet spot" of a lens is depends on the particular design. Some
lenses perform best wide open, others need moderate stopping down. There is
no hard and vast rule.



From: "David J. Littleboy" davidjl@gol.com
Newsgroups: rec.photo.equipment.medium-format
Subject: Re: why MF won't get better ;-) Re: What is depth of focus?
Date: Mon, 11 Nov 2002


"Q.G. de Bakker" qnu@worldonline.nl wrote:
> Where the "sweet spot" of a lens is depends on the particular design. Some
> lenses perform best wide open, others need moderate stopping down. There is
> no hard and vast rule.

It's worse than that: resolution changes _differently_ with aperture at the
center, edge, and corner for every different lens. (In some lenses it seems
the corners are still getting better while the center is starting to get
worse as you go from f/8 to f/16.) I presume you know this site, but for
people who don't:

http://www.hevanet.com/cperez/MF_testing.html

David J. Littleboy
Tokyo, Japan



From: "roland.rashleigh-berry" roland.rashleigh-berry@ntlworld.com
Newsgroups: rec.photo.equipment.medium-format
Subject: Re: why MF won't get better ;-) Re: What is depth of focus?
Date: Sun, 10 Nov 2002


Usually f8 gives the best resolution. Not all lenses will perform the same
though and it is better to check out their MTF graphs if these are
available.But as a rule of thumb, you will get the best resolution at f8 no
matter what the format. Take a look at www.photodo.com in the "products"
section.

I tend to use f16 to give me better depth of field and to overcome film
bulge problems.



From: "roland.rashleigh-berry" roland.rashleigh-berry@ntlworld.com
Newsgroups: rec.photo.equipment.medium-format
Subject: Re: why MF won't get better ;-) Re: What is depth of focus?
Date: Sun, 10 Nov 2002 

...(quotes above)


I would add that as you go to f11 and smaller then contrast gets affected a
lot. If you do a simple experiment and make a tiny hole with two fingers and
a thumb and hold it in fromt of your eye and look through then you will see
this effect of less contrast but increased depth of field. If you use old
cameras for landscape then you maybe are aware of the fall-off you get in
the corners. If you look through the back of the camera through the lens at
the corners then you will see the light being obscured by the lens housing
at about f8 or smaller. It depends on the lens design. Sometimes you have
got to stop down to f16 to avoid this.


From: "Neil Gould" neil@terratu.com
Newsgroups: rec.photo.equipment.medium-format
Subject: Re: why MF won't get better ;-) Re: What is depth of focus?
Date: Sun, 10 Nov 2002


Hi,

Paul King paulking@amaranthstoneware.ca wrote:
> Roland (et al.):
>
> Possibly dumb question... I have it in my brain that F16
> gives 'better' resolution than wider F-stops. Is this not the
> case, and if so, what is the 'best' F-stop for the maximum
> resolution, regardless of DOF. Are there other 'variables'
> involved?

While the rule-of-thumb is that smaller apertures provide more depth of
field, this has little to do with resolution. There isn't a single answer
to the question of resolution by aperture; this performance will be
specific to a lens. One variable which may give a hint at which aperture
will provide the sharpest images is the MTF. However, most lenses don't
ship with a MTF chart specific to that lens, and as has been shown, there
is a bit of variation between samples of the same make and model.

Regards,

--
Neil Gould
Terra Tu AV   http://www.terratu.com
Technical Graphics & Media


From: "David J. Littleboy" davidjl@gol.com
Newsgroups: rec.photo.equipment.medium-format
Subject: Re: How is coc related to lp/mm?
Date: Thu, 28 Nov 2002

"Robert Monaghan" rmonagha@smu.edu wrote:
>
> quoting david:
> I realize that these are often taken as standard. I was pointing out that
> very few people in the world have ever seen prints that actually meet the
> 8 lp/mm standard.
> end-quote
>
> That's why I moved up to medium format ;-)  It all depends on print size.

Since I like numbers, (film size cropped to the A4 ratio: 35mm = 34x24, 645
= 56x39, 6x6 = 46x39, 6x7 = 69x48, 6x9 = 56x79), lets calculate the lp/mm on
the film required for 8 lp/mm on the print.

assume borderless A4, 210mm short side:

(format: ratio = magnification => film lp/mm)
35mm: 210/24 = 8.75x => 70 lp/mm
645: 210/39 = 5.38 => 43 lp/mm
6x6: 210/39 = 5.38 => 43 lp/mm
6x7: 210/48 = 4.375 => 35 lp/mm
6x9: 210/56 = 3.75 => 30 lp/mm

Assume borderless A3, 297mm short side (11.7x16.5)

35mm 297/24 = 12.4 => 99 lp/mm
645: 297/39 = 7.62 => 61 lp/mm
6x6: 297/39 = 7.62 => 61 lp/mm
6x7: 297/48 = 6.19 => 49.5 lp/mm
6x9: 297/56 = 5.3 => 42.4 lp/mm

Assume borderless A2, 420mm short side (16.5 x 23.3)
6x7: 420/48 = 8.75 => 70 lp/mm
6x9: 420/56 = 7.5 => 60 lp/mm

Most films resolve 50 lp/mm at 1:1.6, with the best doing about 60 to 65.
What this means is that for real subject matter:

35mm is hopeless for 8x10 (But we knew that already)

645 and 6x6 are iffy at A3 (oops: so much for my "645 is good enough to
drive the Epson 2200 theory". Ouch. (Actually, I don't think the Epsons can
produce anywhere near 8 lp/mm; 5 lp/mm is an upper limit, since 250 dpi
source image prints are indistinguishable from 360 dpi.))

Only 6x9 can hack it at A2. (Whew! My chuck the 645 and move to a GSW690
when 13MP digital hits US$2,000 theory still flies.)

> Lots of contact prints easily exceed 8 lpmm, up to the paper limits ;-)
> And lots of MF and LF prints to 8x10" or 11x14" can hit 8 lpmm, depending
> on magnification factors ;-)

See above {g}.

David J. Littleboy
Tokyo, Japan



From: hemi4268@aol.com (Hemi4268)
Newsgroups: rec.photo.equipment.medium-format
Date: 28 Nov 2002 
Subject: Re: How is coc related to lp/mm?


>Most films resolve 50 lp/mm at 1:1.6, with the best doing about 60 to 65.
>What this means is that for real subject matter:
>
>35mm is hopeless for 8x10 (But we knew that already)

Not exactly since eye resolution doesn't fall of a cliff at slightly less then
8 lp/mm.

If you did a test with say 40 photographs with a resolution range between 2 to
12 lp/mm in 1/4 lp/mm seperations, the eye would judge any photograph between
about 6.75 to about 9.75 to be about the same.  Anything over 10 would be
judged exceptional and under 6 would still be exceptable.  The eye really does
not sence fuzzy edges until it gets to be way under 4 lp/mm

Larry


From: artkramr@aol.com (ArtKramr)
Newsgroups: rec.photo.equipment.medium-format
Date: 28 Nov 2002
Subject: Re: How is coc related to lp/mm?

...(quotes above posting)


If you take a series of prints of equal lines/mm but some  are of greater
contrast than others, the higher contrast prints will be perceived to be
sharper as proven by Kodak's JND tests.

Arthur Kramer
Visit my WW II B-26 website at:
http://www.coastcomp.com/artkramer


From: "David J. Littleboy" davidjl@gol.com
Newsgroups: rec.photo.equipment.medium-format
Subject: Re: Enlarging limitations
Date: Fri, 14 Feb 2003


"Robert Monaghan" rmonagha@smu.edu wrote:

> now maybe I'm stuck in "lala academic dreamland" ;-), but surely you don't
> mean 55 lpmm is a limit;

Yes, I do. It looks to me the be seriously unlikely that one could get
significantly over 55 lp/mm reliably in actual photography.

> lots of folks using tech pan are waay above that
> "limit", even with (very good) 35mm lenses.
> Tech pan is spec'd at 1.6:1 at
> 100 lpmm, 320 lpmm at 1000:1, so it should do rather better than 100 lpmm
> with typical real world scenes, yes?

But if you look at the lens tests, there really aren't a lot of lenses doing
much better than 60 lp/mm, and that's for high contrast targets. If you are
imaging a low contrast target, and your MTF is too low, you won't get an
image of the target. And the MTF really is way down at those numbers, even
for the Zeiss lenses. At least as I read the MTF charts from Zeiss, the MTFs
are down far enough at 40 lp/mm that even that is going to be hard to use as
a practical matter.

For practical photography, expecting to get over 50 or 60 lp/mm seems
seriously unlikely. Maybe for contrasty images in a studio with controlled
lighting.

> In color velvia is spec'd for 80 lpmm
> at 1.6:1 contrast ratios (50% contrast), and 160 lpmm for 1000:1 ratio
> subjects (test charts)...

But Velvia is horribly grainy. You can't actually use that resolution for
anything other than test charts.

David J. Littleboy
Tokyo, Japan



From: Lassi lahippel@ieee.org
Newsgroups: rec.photo.equipment.medium-format
Subject: Re: Enlarging limitations
Date: Fri, 14 Feb 2003


Robert Monaghan wrote:
>
> now maybe I'm stuck in "lala academic dreamland" ;-), but surely you don't
> mean 55 lpmm is a limit; <...>

I suppose it means the practical limit of not requiring special efforts,
as explained in your own page at http://medfmt.8k.com/mf/limits.html.

BTW, I noted two errors in there:

(quote)

I think a person using 2700dpi scanner cannot see more than 100.
According to Chanon's theory of sampling, 50 line/mm is the max of
2700dpi.


Hiroaki

[Ed. note: Claude Shannon's information theory work is meant, I believe.
Shannon determined that you have to sample at least twice as often (cf.
frequency) as the signal frequency (cf. lpmm) to ensure accurate
sampling... ]

(end quote)

(1) The performance limit isn't two pixels per line pair. You need three
to four, depending on how much contrast you want. At two per pair you
may get a flat grey surface from black&white; stripes, if the pixels
happen to hit the edges. Therefore 2700dpi can only reproduce 25 to 35
lp/mm.

(2) And the sampling limit is named after Harry Nyquist
(http://www.geocities.com/bioelectrochemistry/nyquist.htm). It states
that the sampled signal cannot be reconstructed accurately, if it
contains components, whose frequency is higher that half the sampling
rate. The higher frequencies will be aliased to lower frequencies, and
appear in the image as moiree-like artifacts.

So the above lp/mm performance applies only, if the original image is
low-pass filtered to remove features smaller than the pixel spacing of
the detector. The image must be slightly out of focus, decreasing total
performance even further. Luckily dexterity with unsharp masking will
recover some of the lost sharpness.

To convert from lp/mm to dpi with good contrast, the rule of thumb is to
multiply by 100 (= 25.4 x almost 4).

-- Lassi


From: "Jeremy" jeremy@nospam.thanks.com
Newsgroups: rec.photo.equipment.35mm
Subject: Re: Does the "50 lpmm barrier" still stand?
Date: Fri, 04 Jun 2004 

"Mike Henley" mnhenley@msn.com wrote...
> I found this page interesting http://medfmt.8k.com/mf/limits.html
> though i'm not sure i fully understand it.
>
> First of all, i'm not sure there's an obvious date on this page but
> looking at the site generally i think it's a 1998 date. (some other
> articles are dated)
>
> Has the technology, particularly in 35mm film, improved since to
> change this? well, it has undoubtedly improved, but does the "50 lpmm
> barrier" remain a significantly limiting factor?

Erwin Puts, on his Leica website, suggests that there is a lot more that
factors into the equation than simply resolution. He says that there may be
50 different factors that should be considered.

A good argument can be made that if your image quality requirements are
super-critical, you'd probably do better by giving to medium format. That
seems to be pretty much the model that Pentax uses. They offer several
well-regarded 35mm lenses in their "Limited" line, and they also offer 645
and 6x7 equipment for those photographers that need more.

I doubt that even the best Leica or Zeiss 35mm lenses can produce images
that beat Pentax's medium format lineup.




End of Page