Velvia 50 - 14 stops

[peter_wang6]

<p>Vijay, either you are a technical writer or you should be one. That was so well put even I understood it. LOL</p>

[stwrtertbsratbs5]

<p>Edward is right about using a step wedge. That is the correct way to test the dynamic range of a film.</p>

[mauro_franic]

<p>The point of this post is not whether Ektar can capture 12 or 15 stops. It is about getting Velvia colors with twice the dynamic range.</p>

<p>No one argues against a step wedge. If someone want to run the test and post it, most people would appreciate it.</p>

[Ed_Ingold]

<p><em>No one argues against a step wedge. If someone want to run the test and post it, most people would appreciate it.</em></p>

<p>There is no scientific reason to repeat work which has already been published and accepted. Kodak has been performing and publishing characteristic curves for at least 60 years (I have bulletins from the 1940's), using the best techniques and equipment. It is not so obvious to laymen how to interpret these curves. I have attempted to put a "photographic" face on these numbers by relating the scales to f/stops, slope to contrast, and span to dynamic range.</p>

<p>An image of a step wedge would serve little purpose (other than to prove you had one), unless accompanied by a plot of exposure vs density (e.g., a characteristic curve). Web images, or even prints, have too little dynamic range to see the results objectively. The Photoshop "Information" box could be used in lieu of a densitometer, or you could accept the results from DXO software (which I find inadequately documented).</p>

<p>It would be interesting to publish results which are not easily available, as for variations in processing of B&W film, or various digital sensors. DPReview.com now publishes dynamic range data for DSLRs under test, based on step wedge measurements. Again, why duplicate those tests, which appear to be well designed and objectively presented. The results of Ektar 100 are locked in - C-41 processing is not ordinarily subject to variations (other than Tetanal, which I find produces uniformly low quality results).</p>

<p>I can't get too excited about testing color film. As noted, objective data is already available. More important, I don't think about dynamic range, characteristic curves, etc., when I use film. It make more sense to shoot enough and scan enough to get comfortable with a particular emulsion. The "science" is useful to solve problems, predict performance, or (most certainly) to satisfy one's intellectual curiosity.</p>

<p>I am curious about the actual performance of my CFV back. Hasselblad states (as do Phase One and Leaf) that the dynamic range is 12+ stops, but without objective data to substantiate this claim. Stouffer step wedges cost as little as $20. If I do the work, I will share the results. None are small enough to fit against the sensor itself, so it will have to be through a lens, which will degrade the results to some extent.</p>

[hclim]

<p>Mauro,<br>

In such a presentation, wouldn't it help if you had spot metered and recorded the extreme ends that you wanted captured in detail. Then we know for sure the dynamic range you are talking about.</p>

 

[stwrtertbsratbs5]

<p>"The point of this post is not whether Ektar can capture 12 or 15 stops. It is about getting Velvia colors with twice the dynamic range."</p>

<p>Mauro - You're the one making the ridiculous claim of 14 stops. Look at the title that you placed on this thread! And Edward is right again - I'm not about to repeat work that was properly performed by the manufacturer.</p>

[photo_smith]

<p><strong><em>I have attempted to put a "photographic" face on these numbers by relating the scales to f/stops, slope to contrast, and span to dynamic range.</em></strong><br /> <strong><em></em></strong><br /> Edward why attempt to find dynamic range from Kodak curves? Surely you understand they are derived from Stouffer type wedges?<br /> Those wedges were not designed to find out the limits of the dynamic range but rather to find out the films response over a limited range of density values which with a 21 step transmission wedge is around 11 stops (curious coincidence?).<br /> Remember our discussion on why the curves stop on the right hand side? How many stops does a step wedge cover? why don't Kodak plot further along the curve?<br>

Your decision to start counting density at 0.7 (rather than base +0.1) still puzzles me why use slope and contrast build when a set of standards has been in use by Kodak since 1942- just use the Jones point!<br /> Trying to make absolutes from estimates in order to give figures a "photographic face' is pointless if those figures don't give you all the data you require.</p>

[Ed_Ingold]

<p>DPReview uses a Stouffer 1/3rd stop step wedge with 41 steps, covering a dynamic range of 13 stops (<a href="http://www.stouffer.net/TransPage.htm">http://www.stouffer.net/TransPage.htm</a>).</p>

<p>Kodak makes no assignation of dynamic range, nor any standard method for its determination*. I suggest measuring from a point on the curve with a certain slope because the slope is related to contrast, hence the amount of useful information recorded on the film. If the slope were less than about 0.20, it would be difficult to distinguish white from black in the image, let alone any reasonable scene.</p>

<p>This is more generous than measuring from the extrapolated "corner" value (the preferred method from an engineering and mathematical analysis background). At the other extreme, DXO measures from the point where the density is just barely discernible from the noise level of the measurement. IMO, this exaggerates the useful dynamic range by 2 or 3 stops, trivializes any differences between sensors, and fails to consider whether the toe region can actually register an useful image.</p>

<p>* q.v., <a href="http://motion.kodak.com/motion/uploadedFiles/US_plugins_acrobat_en_motion_education_sensitometry_workbook.pdf">http://motion.kodak.com/motion/uploadedFiles/US_plugins_acrobat_en_motion_education_sensitometry_workbook.pdf</a></p>

[photo_smith]

<p>Edward Ingold.<br /> <strong><em>DPReview uses a Stouffer 1/3rd stop step wedge with 41 steps, covering a dynamic range of 13 stops</em></strong> (<a rel="nofollow" href="http://www.stouffer.net/TransPage.htm" target="_blank">http://www.stouffer.net/TransPage.htm</a>).<br /> <br /> No Edward that's wrong the 41 step is still 11.5 stops just that the density increments are smaller 0.5 (1/6 stop) vs . 1.5 (1/2 stop) for the 21 step they also do a 31 step wedge with 1/3 stop increments the density range is the same 11.5 stops. I'm starting to think you've never used a step wedge let alone lecturing others on their use!<br /> <br /> <strong><em>I suggest measuring from a point on the curve with a certain slope because the slope is related to contrast</em></strong><br /> <strong><em></em></strong><br /> That's just plain silly, we have a set of standards already in place that use 0.1 plus base fog as a speed point, anywhere after that density is building. Your method smacks of re-inventing the wheel for the sake of your own arguments-just use the method everyone else does the ISO standard.<br /> <br /><strong> As for the Kodak PDF you link to I was the one who made you aware of it! Please read it and use Kodaks modus rather than you own cludge</strong></p>

[Ed_Ingold]

<p>The Stouffer T4110 step wedge has 41 density increments of 0.1, which are equivalent to 1/3rd stops, as clearly stated in Stouffer's specifications (q.v.). Therefore 41 steps encompass 13 stops.</p>

<p>Kodak does not specify "speed point" as a base line for dynamic range. Kodak and the ISO procedure refers to the determination of sensitivity. In fact, dynamic range is not mentioned in either method. Taking the Kodak paper at face value, the dynamic range would commence at the 0.7 density point - well above the corner value. I have stated my assumptions, which can be reproduced by anyone interested. Furthermore I have stated the justification for these assumptions. If that is "silly" to you, then you have never written nor reviewed a scientific paper.</p>

<p>Kodak has apparently updated their equipment since the 1940's. The Ektar 100 characteristic curve covers a span of 13 stops. Kodak apparently feels that the maximum exposure in these curves is sufficient to describe the useful range of the film. It is, after all, nearly 7 stops above the point which defines the ISO sensitivity.</p>

[marc_bergman1]

<blockquote>Our new Dynamic Range measurement system involves shooting a calibrated Stouffer Step Wedge (13 stops total range) which is backlit using a daylight balanced lamp (98 CRI). A single shot of this produces a gray scale wedge from (the cameras) black to clipped white (example below). Each step of the scale is equivalent to 1/3 EV (a third of a stop), we select one step as 'middle gray' and measure outwards to define the dynamic range. Hence there are 'two sides' to our results, the amount of shadow range (below middle gray) and the amount of highlight range (above middle gray).</blockquote>

<blockquote>This is from DPReview.<br /></blockquote>

[photo_smith]

<p>Edward the Ektar curve shows data for the 21 step as it only goes to density 3.05 or do you think the curves drop off the face of a cliff?<br />The Kodak paper is NOT the ISO paper but you can estimate the dynamic range by looking at the curve, but that curve has been generated by using a 21 step wedge with a maximum density of 3.0 if Ektar is capable of more density then your assertions are incorrect<br /> <strong>Taking the Kodak paper at face value, the dynamic range would commence at the 0.7 density point</strong><br /> Seriously I think you're just being obtuse now, that would place it above the mid tone zone V</p>

[bernard_lazareff]

<p>Mauro,<br>

Whether or not there are 14 stops of DR in Ektar (personally dubious, but haven't run tests), for me the bottom line of your interesting experiment is that one can benefit from the large DR of negative and its flexibility to emulate the color rendition of a reversal film. Congratulation. On emight even think of having a special profile such that right at the scanning step, the file would come out as Ektar-like, or Velvia-like, or... Without the need of a simultaneous shoot with the target emulsion.</p>

[photo_smith]

<p><em><strong>Kodak apparently feels that the maximum exposure in these curves is sufficient to describe the useful range of the film. It is, after all, nearly 7 stops above the point which defines the ISO sensitivity.</strong></em><br /> Edward that's not how ISO is defined, did you read the linked PDF?<br /> Kodak plot curves using a 21 step wedge (page 6 PDF) <br /> ISO sensitivity is not defined by the 0.1 plus base fog that is just the point we measure as the inertia point from that we measure 1.3 log units to the right of that point then follow up to the curve and measure that density value film speed will then be that number divided by the antilog of the point.<br /> For instance if you take 0.1 then go along the curve for 1.3 log E and the value is 0.9 the correct way to determine speed is the antilog of 0.9 (8) divided by 800 therefore the film has a speed of 100<br /> Read the PDF.<br /> Meanwhile asserting that the Kodak curve shows data for 13 stops is obviously incorrect it clearly shows 11.5 and the maximum density is 3.05 on the blue channel the same as the maximum for the 21 step wedge.<br /> o prove that Ektar is only capable of 10 stops you will need to plot your own test as the Kodak curve has insufficient data to draw that conclusion.</p>

[Ed_Ingold]

<p><em>ISO sensitivity is not defined by the 0.1 plus base fog that is just the point we measure as the inertia point from that we measure 1.3 log units to the right of that point then follow up to the curve and measure that density value film speed will then be that number divided by the antilog of the point.<br /></em><br>

True, which place the ISO point on the linear portion of the curve, about 7 stops below the maximum exposure value ;-) On a pragmatic scale, Mauro's data confirms a 10-11 stop dynamic range if you look at just one point on the figurine.</p>

[photo_smith]

<p>Not so fast remember this is colour film with 3 curves the data should be averaged. The figurine is not a good test better one would be to photograph a white fluffy towel for 15 steps 7 each site of the metered point (mid grey) then measure with a densitometer base and d-max then count the steps between the two where detail is present.<br>

I would imagine that about 11 is realistic but you can't make absolutes from estimates, you certainly can't use a curve derived under contact printed 21 step wedge conditions to ascertain what range a film/lens combo will yield.</p>

[Ed_Ingold]

<p>A step wedge in contact with the film/media is the best method in not the most practical. A lens is subject to various aberations, but particularly flare, which can only adversely affect the results.</p>

<p>All measurements are estimates! If you know the uncertainty of your standards and methods, you can assign an uncertainty to the results. The less the uncertainty, the better the estimate.</p>

<p>In the case of Ektar, it makes little difference whether you measure dynamic range from the red, green or blue line or some composite of the three. The density is different, but the curves are nearly parallel in the toe area. The absolute value of the density is irrelevant, just the shape of the curve as it affects the start and end points along the horizontal axis.</p>

[photo_smith]

<p><strong>A step wedge in contact with the film/media is the best method in not the most practical. A lens is subject to various aberations, but particularly flare, which can only adversely affect the results</strong>.<br /> <br /> Which is why the step wedge is of limited value, I generally use lenses when I photograph what about you? They impart their signature to my images which is why the test is more useful to a photographer than the curves issued by film companies- they are real world.<br /> Why do you think ISO standards call for D log 1.3 as a brightness range? do you think they try to factor in flare aberrations? What other curves do they use when working out parameters? Have you seen a quadrant diagram?<br /> What is the flare factor in the latest ISO standard?<br /> <br /> <strong><em>If you know the uncertainty of your standards and methods, you can assign an uncertainty to the results. The less the uncertainty, the better the estimate.</em></strong><br /> <strong><em></em></strong><br /> Sure but in this thread you have berated the OP for not using the Kodak data from the curves, then have formulated your own DR measurements based on poor methodology and limited data.<br /> <br /> <em><strong>In the case of Ektar, it makes little difference whether you measure dynamic range from the red, green or blue line or some composite of the three.</strong></em><br /> <em><strong></strong></em><br /> During this thread you've shown that you don't understand how to read the curves, you are aware the method for reading colour negative is different from mono ones?What difference does the colour masking make?<br /> <br /> <strong><em>The density is different, but the curves are nearly parallel in the toe area. The absolute value of the density is irrelevant</em></strong><br /> <strong><em></em></strong><br /> Sure they are parallel in the toe area, what about the D-Max? do you think there is no info to the right? If we used the D-Max 4 step wedge would there be more data in that part of the curve? Or does the curve drop of a cliff?<br /> The only way to test a film is to test it in the conditions you use, with the camera/lens combination you use.<br /> The published curves are of limited use with limited data, please don't berate people for trying to establish their own working tests.</p>

[Ed_Ingold]

<p><em>Sure but in this thread you have berated the OP for not using the Kodak data from the curves, ...</em></p>

<p>Not quite! I showed how his method added the dynamic range of the subject to that of the film, exaggerating the results by 3 or 4 stops. He is not alone in this fallacy.</p>

[photo_smith]

<p>No Edward what you did is use your own method for calculating dynamic range, one where you start at a density of 0.7 because you 'feel' that is a more relevant method than others use.<br /> If you want to approximate DR with mono you need to use Base fog +0.1, normally this will give a start point of around 0.24 why do you insist on using your contrast method rather than the traditional one? You must have noticed it shaves off almost 2 stops from the DR figures people get with the conventional scientific methods used by people like Phil Davies?<br /> <br /> The fallacy is all yours, if you insist others are wrong in their calculations because they use convention methods to find density ranges recordable by film it makes me wonder why you continue to make these assumptions based on your modus?<br /> Certainly you should defer from saying other exaggerate when you are using figures derived from your questionable methodology.<br /> You state that no measurement for DR exists, that's not strictly true, but then someone who starts reading curves at 0.7 and ignores the possibly of the curve data going beyond the right hand side probably has their own agenda and isn't interested in the truth.</p>

<p> </p>

[Ed_Ingold]

<p><em>Sure they are parallel in the toe area, what about the D-Max? do you think there is no info to the right? If we used the D-Max 4 step wedge would there be more data in that part of the curve? Or does the curve drop of a cliff?<br /></em><br>

You would not need a special "DMax 4" step wedge. Since we are only interested in relative measurements, not absolute sensitivity, it would suffice to increase the exposure by a stop or two to measure well into the DMax region of the film. Perhaps someone cares to perform this experiment and report the results.</p>

<p><em>No Edward what you did is use your own method for calculating dynamic range, one where you start at a density of 0.7 because you 'feel' that is a more relevant method than others use.<br /></em></p>

<p>Actually, I based my starting point on the slope, not some arbitrary density value, which is related to contrast, hence the ability to record detail. Methods improve all the time, otherwise scientists would be limited to the teachings of Paracelcius or Aristotle.</p>

[larrydressler]

<p>I am glad I just take pictures that people like and quit argueing about it... just went out and shot a roll in an old Box camera...... Have a great day.</p>

[Ed_Ingold]

<p>Larry,</p>

<p>I'm reminded of the wind player who gets hopelessly lost in, say, the Beethoven's Sixth Symphony, raises his hand and says, "Can we tune?". Enjoy your box camera :-)</p>

[photo_smith]

<p><em><strong>Actually, I based my starting point on the slope, not some arbitrary density value, which is related to contrast, hence the ability to record detail. Methods improve all the time, otherwise scientists would be limited to the teachings of Paracelcius or Aristotle.</strong></em><br /> <em><strong></strong></em><br /> Yes you based your point (density 0.7) using calculations based on YOUR interpretation of contrast building. The figure 0.7 BTW is very close to mid grey certainly higher than open shadow on a sunny day. Why start so high up the curve? do you think no detail is recorded between the inertia point and 0.7?<br /> <br /> Yes scientific methods improve all the time, but your calculations are not an improvement are they! We have the International Standards Organisation and scientists that work on film speed standards; they have set 0.1+ base fog as the point in which contrast builds- why do you think they are wrong? why should we chop off the lower 2 stops.<br /> I know! so we can have retarded arguments on the internet about why we feel our own standards are better than those of the ISO.<br /> I'm done.</p>

[mauro_franic]

<p><a href="../photodb/user?user_id=2389883">Bernard Lazareff</a></p>

<p>I tried it, but the color match profile changes as the lighting conditions change. It only works acurately with two cloesly timed and composed shots.</p>