In search of...film flatness and a quiet shutter.

<p>"....much farer...." = much fairer. Twenty lashes.</p>

<blockquote>

<p>I'm looking for cameras that are known for film flatness, a quiet (leaf) shutter...I do understand that film flatness is kind of controversial, as in <a rel="nofollow" href="../medium-format-photography-forum/001awY">this thread</a> in this very forum. I acknowledge that I might be chasing after something that makes no difference. But humor me, please.</p>

</blockquote>

<p>Thanks for so appropriately searching the archive and referencing its relevant content!</p>

<p>Nothing has changed since that decade-old thread. The deniers are still posting and film continues to bulge, making a visible difference on images exposed in the same cameras identified as susceptible ten years ago. :-) At this point, I own a Mamiya 7 with 80mm lens. In my opinion, you won't find flatter film or higher resolution across the image area than that combination:</p>

<p><a href="http://www.hevanet.com/cperez/MF_testing.html">http://www.hevanet.com/cperez/MF_testing.html</a></p>

<p>Sal,</p>

<p>Do you know what was the contrast level at which those resolutions were measured, what is its relationship to actual conditions and was the contrast constant betewen the tests, and how did they (could they?) assure optimum film flatness for each camera type (e.g., some cameras don't like film being used after long residence times in the camera, others less so)? From what I've read, there is a more or less inverse relationship between contrast and resolution.</p>

<p>There is a description at the bottom of that linked page relating some aspects of the tests. I recall that a standard USAF test chart was used, but Christopher Perez should be able to confirm. He could also elaborate on how long the film was left in cameras between exposures.</p>

<p>Even if Christopher powered quickly through the rolls to minimize bulging, results with some cameras were bad enough to conclude that the problem exists. See his Note 1.</p>

<p>I saw the few sentences footnote, but nothing about contrast. I have seen some lenses that have great resolution at 10% or 30% contrast, but which fall off greatly at higher contrasts. I think that is why most modern tests are done using MTF curves instead of resolution. Chromatic aberration measurements and light fall off are also just two other factors that determine a lens's quality. As nice as the list is, it needs other data I think to enable good comparisons.</p>

Those would be special lenses, Arthur, the ones that do worse at higher contrast. ;-)<br><br>MTF is a measure of contrast and resolution. They measure contrast (one variable) of transitions at a given (the other variable) resolution, and express the result as a percentage of the perfect score (i.e. the contrast in the target).<br>The 'output' is given as a percentage of the 'input'. And that takes the contrast of the target out of the equation almost completely.<br><br>Typically, contrast falls off with increasing resolution. That's why with increasing resolutions, the score - as a percentage of perfect - is lower.<br>So they do not have great resolution at 10% contrast, but a poor 10% contrast at great resolution. Better contrast (like 30%) at lower resolution, and the best performance at a low resolution.<br><br>That performance measured in MTF tests is affected by chromatic aberration, and all other aberrations. And because of that, MTF is a great measure for lens quality. The only other two things we'd like to know and are not included in the MTF measurement are light fall off and distortion.<br><br>Present a lens with a high contrast target (at any resolution), and the inevitable contrast fall off will leave some discernible pattern in the image a lens produces, where a low contrast target presented to the same lens will only help to create a featureless blur. So special lenses indeed that do better at low contrast. ;-)<br><br>P.S.<br>MTF measurements, and other lens tests, measure the performance of lenses. Not how flat film is.<br>We measure how flat film is by exposing it, examining it with loupes, and not finding problems. That is, not, until some company trying to sell a film flattener tells us there is a problem. All of the sudden, some people start believing there is something they have never seen before, do not see now, and will never see. But that company said it's there, so it must be there... :D

<p><em>"Those would be special lenses, Arthur, the ones that do worse at higher contrast."</em></p>

<p>QG, I have not done any measurements, but from what I have read the resolution and the contrast performance of a given lens are usually inversely related. Those references are quite old, like the old (1980s) technical test measurements of contrast and resolution by (either) Popular Photography and (or) Modern Photography magazines, prior to the MTF tests. By quoting a simple resolution figure at low contrast, it says apparently little about the performance of the lens at other contrast levels. I stand of course to be corrected if that is not so.</p>

<p>That (the singularity of the measurements) was my reservation about the test link above, that is so often quoted to compare MF lenses. Also, and as we have both noted, the absence of chromatic aberration, spherical aberration and other similar lens performance measurements give only an initial idea of the quality of a lens. There are few complete test comparisons available. I tend to now read the two top French magazines in this regard (Chasseurs d'Image and Réponses Photo) which frequently review a wide variety of camera lenses. </p>

MTF testing is a pretty complete way of testing a lens. MTF performace measures the effect of just about everything that can have an influence on a lenses performance. So that single way of testing provides much more than <i>"only an initial idea of the quality of a lens"</i>. Quite the contrary: it provides a pretty much complete idea of the quality of a lens. All we need to know more is light fall off and distortion. And then you have the complete picture.<br>The only disadvantage of MTF testing is that what it provides is an 'integrated" result, that does not allow singling out the effect of any individual factor that may affect performance. But complete and comprehensive it is.<br><br>The thing with contrast and performance is that anything that decreases performance (apart from fall off and distortion) does so by decreasing contrast. So what you need to know is how the contrast in the test target is lowered in the image produced of it by the lens. How contrast in the target is modulated and transferred to the image projected by the lens.<br>And that's exactly what MTF tests produce: a measure for the decrease in contrast, comparing the one in the target with the one in the image. It's a relative thing, expressed as a percentage. The higher the percentage, the better the performance, because the more 'true' to the original.<br>Now if the contrast in the target is increased, the resulting contrast in the image gets beter too. Still less than in the target, but more than what you would get if the lens was presented with a lower contrast target. And with that higher contrast, the ability to keep fine detail separated increases too.<br><br>So lenses perform better when provided with a higher contrast target. Lenses do worse at lower contrast.<br>If you keep target contrast at a fixed value, the contrast in the image produced by the lens will get worse with increasing resolution in the target. So lenses indeed do worse at higher resolution, because the contrast in the image they produce drops off.<br>So lenses do not do worse at higher contrast. They do worse by producing/showing less contrast.<br><br>Quoting a resolution figure at low conbtrast does indeed not say much. Quoting the contrast produced (whether high or low) at a given resolution (high or low) however tells quite a lot. We know just about everything when we get more of such relative contrast figures at different target resolutions, and at different places in the image.

<p>Q.G.:</p>

<blockquote>

<p>Now if the contrast in the target is increased, the resulting contrast in the image gets beter too. Still less than in the target, but more than what you would get if the lens was presented with a lower contrast target.</p>

</blockquote>

<p>So when they show MTF curves for film or lenses, shouldn't the MTF figure be quoted with a contrast ratio of the chart shot? I'd imagine that the MTF curve would change based on contrast of the target...</p>

No. MTF curves give the relative figure: result/original.<br>30% of quite a lot is still quite a lot. And 30% of almost nothing is almost nothing. But no matter what, 30% is 30%.<br>So if we know that performance figure, and understand how it is a input vs output thing, it is understood that (and how) the input will have an effect on the output.

<p>Oh right. I'm embarrassed at my last post... :-P</p>

<p>From Norman Koren's site, the MTF(f) is 100 x C(f)/C(0), where</p>

<ul>

<li>C(f) = (Lmax-Lmin)/(Lmax+Lmin) for L values at given spatial frequency <em>f</em></li>

<li>C(0) = same as above but L values taken from low frequency region</li>

</ul>

<p>So I figure one could calculate the MTF curve for a # of different charts with different contrast ratios, then fit the data & get a best-fit curve & represent that as the final MTF curve?</p>

<p>Is that how they do it or do they really only calculate the MTF from one chart with one contrast?</p>

<p>-Rishi</p>

<p>How about a leaf-shutter lens on a field camera? Meets most of the criteria given, assuming weight and workflow are not considered. . . :-) Go Horseman!</p>

That's not MTF, Rishi, but an attempt to express the performance of a lens in one (1) 'MTF' number.<br><br>MTF is 'just' the ratio of contrasts, how much of the contrast in the target is left in the image.<br><br>Trying to express the MTF performance of a lens in one single number may look appealing, but, of course, says less about a lens than the two (only !) numbers that are compacted into one.<br>Usual MTF testing only (again: !) plots the MTF across the image (something also thrown away in that compact score: the performance of a lens varies across the image. Center and edge/corner performance usually differs hugely) for three distinct target resolutions, not providing any measured info about how the lens does in between. Or plots the MTF score against a scale of target resolutions, without showing how it varies across the image.<br>Both ways of presenting already reduce the amount of information, with the first being (quite a bit more) preferable over the second. That single score thing can not be more than just a rough indication.<br><br>Again, the input contrast is not that important. We do not need to know what it is, unless we are doing the test ourselves. The resulting figure says more than enough.

<p>How did this get from quiet shutters and film flatness to MTF curves?</p>

<p>Anyway the article about fitting a 'Blad with a vacuum back gave me a chuckle. If the guy wants to copy flat artwork he could have bought a used Littlejohn and a set of process lenses for less than he spent on making that back.</p>

<p>Honestly guys! Worrying about a bit of a ripple in your film when you're photographing 3-dimensional subjects. Really? Cos just rocking back on your heels or your subject breathing probably throws your focus out more than any film bulge. And lenses don't have an exactly flat field either.</p>

My fault entirely, RJ. And i apologize.<br><br>Speaking of vacuum backs for Hasselblads (i missed the reference to such things in this thread), i happen to have a couple.<br>70 mm backs, with customized pressure plates and an air hose attachment on the outside. They were used in aerial photogrammetric/mapping applications (not by me. I just got them for a song when they were replaced.) Though not on Hasselblad's photogrammetric cameras. Go figure... ;-)

<p>Minolta Autocord, Maxwell screen. If it needs work, send it to Karl Bryan.<br>

Silent, relatively small, sensible film path for flatness, superb lens.<br>

Oh, pretty cheap too, even with a CLA.</p>

<p> </p>

<p>Those of you concerned with film flatness in medium format systems would do well to read my post with photos regarding the <em><strong>utter inability</strong></em> of a Phase One/Mamiya 645 system to maintain any semblance of film flatness:</p>

<p>http://www.photo.net/medium-format-photography-forum/00YUob?start=23</p>

<p>Rodeo Joe: 'a bit of a ripple'? Really? You call a 4-8mm section of film bulging out <em>millimeters</em> 'a bit of a ripple'? Do you know what the tolerance of the micro AF adjustments on new dSLRs is?</p>

<p>Here's your 'bit of a ripple' after sitting for <em><strong>only 5 minutes</strong></em> (not overnight, not 1 hour, not 30 minutes, but <em>5 minutes</em>, you know, the time it might take to set up & meter your next shot) on the rollers of a 645 back:<br>

<img src="http://rishisanyalphotography.com/ForumPostFiles/photo.net/Mamiya645FilmBack_Frame4-5min.jpg" alt="" width="800" /><br>

Good riddance. And people wonder why film went the way of the dinosaurs? </p>

<p>-Rishi</p>