Dynamic Range of a Film?

<p>hello --<br>

how can i determine the dynamic range of a specific film? i thought i would be able to find that information from the product description e.g.:</p>

<p>http://www.kodak.com/global/en/professional/support/techPubs/f4036/f4036.pdf</p>

<p>however, i think i might just not understand how to read those curves and how to translate them into f-stops or an equivalent of measurement, which would help me when metering a scene. any help would be greatly appreciated.</p>

<p>thanks in advance for your time.<br>

patrick.</p>

 

<p>Depends on film you use. Normaly processed bw like Kodak T-max and Ilford Deltas are about 6 - 7 f-stops wide, old style bw is about 6, color negative material is about 4 - 5, chromes 3 - 4 f-stops wide. You can determine exposure latitude by shooting same scene at -2EV, -1EV, N, +1Ev and +2EV in same light. Technically speaking there is no such a thing as films dynamic range but exposure latitude.</p>

<p>Look at the characteristic curve. The horizontal axis is exposure, in logarithmic units. 0.3 units there is one stop more exposure, which is twice as much exposure. (log(2) = 0.30102999.) The vertical axis is the developed density of the film, again in logarithmic units, so 0.3 units is twice as dense.<br>

The straight portion of the curve is the part where you get no compression of tones. So the straight line portion for this film is about 3.0 log exposure units, or 10 stops. There's another compressed stop in the shadows.<br>

Depending on what you're doing with the film, you may find densities of 2.4 in the negative a bit hard to deal with. For instance, quite a lot of film scanners would be challenged by that, and you might get crushed highlights. But optical printing can deal with that.<br>

Note also that the density range of the negative is less than the scene brightness range of the subject. But photo paper is more contrasty than paper, so you get some back. But the density range of paper is less than the brightness range of the real world, so you have to choose some highlight and shadow details to lose.<br>

So while you can capture 10 stops of dynamic range on the negative, part of your artistic challenge is deciding what part of that to throw away on the print.</p>

<p>Patrick, color negative and B&W negative have in general over 12 stops. Transparencies between 5 and 7.</p>

<p>in practice, i've found that the latitude one can achieve depends to a high degree on the development and the way the curve is manipulated (through placement on the zone system). the curves tend to be guides for normal development, so of course this should be adjusted to suit your needs and the light by experimenting with developers, dilutions, etc.</p>

<p>here is an example of what seems to me to be good latitude from neopan 1600 @ 1000 in microphen 1:2 11 mins, considering in fact this is a 1 stop push. my experiments with this film in stock solution didn't come close to this latitude.</p>

<p><img src="http://rjjackson.com/images/temp/vic_honey.jpg" alt="" /></p>

<p> </p>

<p>While it may be helpful to know how many 'stops' your film can handle it doesn't really address your use or rather how you can utilise them.<br>

Most B&W films can handle around 8-10 stops easily with normal processing but as alluded in earlier posts paper handles less, a good printer on grade 2 paper can normally get 8 separate zones (not counting black and paper white) Paper generally has a very steep curve, which can provide better shadow and highlight separation. You'll need a quadrant diagram to plot where the tonal compression takes place.<br>

What is more helpful is to think of the film as a compression medium, your scene can have an enormous range of tones the average is about 1:160 (Log 2.2) most films can record up to a ratio of 1:500 with careful development and exposure. <br>

Tonal compression is not a bad thing, and the eye readily accepts it in the shadow and highlight portions of the image. <br>

So when people talk about the 'dynamic range' of a film there is some ambiguity about the nomenclature; do they mean how many stops can be recorded on the actual film as measured by a step wedge? Or do they mean the maximum SBR the film can handle with careful exposure and development.<br>

I once saw a demonstration of the latter where a book was placed in a black box lined with velvet, a light with a bare bulb was placed on the box and was the only illumination in the room, the range was huge.<br>

In the resulting print you can clearly read the text on both the bulb and the book. The results were possible with careful exposure and development.</p>

Set up a gray card and take a normal exposure of it. Then open up one stop more and take another exposure, then open up another stop and take another exposure... etc. Then close down one stop from the gray card exposure and take another exposure, then close down another stop and take another exposure...etc. Develop the film and you can see how many stops it is from black to pure white, the range of the film.

<p>hello all -- thank you for your time in clarifying this to me. i guess i didn't factor in the output or input factor (print & scan) as much as i had thought. though i take from your responses that reaching the limits of film is possible through careful management of development and measurement, it is more likely that paper and scanner will compress the range which will be present. thank you clarifying all of this to me and providing some averages from the different types of film. </p>

<p>patrick.</p>

 

<p><b>@John Shriver:</b> Thank you for explaining in great detail how to read the charts. I was at a loss about the logarithmic nature used to chart it. Your message was very helpful and i appreciate very much you taking the time.</p>

<p><em>...reaching the limits of film is possible through careful management of development and measurement,...</em></p>

<p>The dynamic range is represented by the characteristic curve, which in turn depends on the way the film is developed. If you change the development, you change the characteristic curve. If you were to draw the new curve, using a step wedge for example, you would see the effects of development in quantitative terms. Exposure does not change the curve, merely which part of the curve you are using.</p>

<p> </p>

<p>Edward <br /> That is a very simplistic view, and one that doesn't take into account that although the film curve does change with development, dynamic range is not 'fixed' just simple push/pull will lessen or increase dynamic range (although this is a blunt instrument) <br /> Exposure changes the dynamic range of the material you use. When scene dynamic range is large and you want to record it your effective speed becomes lower. This is because exposure and development affect the negative differently. Most film curves you see are ISO standard 1:160 SBR with a flare factor added in.<br /> You can easily find the dynamic range by dividing the density expressed as stops (D/0.3) by the gamma, gamma is development dependent and density is mainly exposure dependent (obviously development plays a part).<br /> For a film with a gamma of say 0.6 bar (normal contrast) and a density of around 3 which works out to 10 stops you should get a DR of about 16 stops. <br /> So if you want to record a scene with a high SBR you will lower you film speed to obtain the density you need at the lower gamma (less development)<br /> Most of this is basic stuff and I'd recommend you read "Beyond the Zone System" by Phil Davis which updates the original zone system for modern use, and moves the system in a direction Adams had been considering before his death.<br /> Mark</p>

<p>Are you referring to "latitude" or to "dynamic range"?<br>

"Latitude" indicates how far from the correct exposure you will get a negative/slide still able to render an "acceptable" image.<br>

Dynamic range is how many "tone" or "separate zone" (from the brightest bright to darkest dark ) the film can capture.<br>

Anyway latitude and dinamica range are related each other, in fact a filme with a low latitude (i.e. slide film) has a low dynamic range too; a film wich has a wide latitude (i.e. B&W negative) has a wide dynamic range). </p>

<p>I don't think "latitude" is a concept that is a property of a film alone, but is a function of the relationship between the dynamic range of a film and the subject brightness range you are trying to squeeze onto that film. </p>

<p>For example if you have a film with a dynamic range of eight stops and a scene where you want detail in two parts of the subject eight stops apart, then you have no latitude at all since there is only one exposure value that will capture the scene in totality. If OTOH you have a subject brightness range of seven stops and a dynamic range of ten stops then you have <em>up to</em> three stops of latitude available depending on where you have the exposure set. In other words your "latitude" may be three stops in one direction and zero in the other. </p>

<p><em>That is a very simplistic view, and one that doesn't take into account that although the film curve does change with development, dynamic range is not 'fixed' just simple push/pull will lessen or increase dynamic range (although this is a blunt instrument)</em></p>

<p>I think "realistic" is the correct term. The characteristic curve represents the output (negative density) versus the input (exposure to light) in absolute terms. The range of exposure over which you get a useful* change in density is the dynamic range of the film. The slope of the center of the curve is called "gamma", a measure of contrast. If you change the development, that curve changes - it is as simple as that. Actually, the dynamic range stays about the same, but the gamma changes. Hence an high-contast subject becomes easier to print if you "pull" the development because the gamma is reduced. The film "speed" also changes, because the density for a given exposure is less in that event.</p>

<p>I see several references to "latitude", which is the amount you can vary the exposure for the scene and obtain a useful image. If the dynamic range of a scene is less than the dynamic range of the film (or medium), you can vary the exposure (placement) of that scene as long as the darkest or lightest portions for which you wish to see detail don't go outside the boundaries of the characteristic curve. In otherwords, latitude is the range over which you can shuffle a box within a larger box. As a corollary, the greater the dynamic range of the scene, the less the latitude (the little box fits more tightly in the larger one).</p>

<p>* A "useful" change in density is not defined in the literature. However at the low (and sometimes the high) ends of the characteristic curve, there is little change in density with exposure - the line is nearly flat (the slope approaches zero). There is typically about one stop difference if you measure dynamic range from the point with a 10% slope versus a 50% slope.</p>

<p>"If you change the development, that curve changes - it is as simple as that. Actually, the dynamic range stays about the same".<br>

<br /> Really are you sure? <br /> So if the development changes the 'curve changes' (for that I presume you mean gamma) but the dynamic range recorded by the film remains the same?<br /> That is plain wrong.<br /> Ever exposed Tech pan at 250 and developed in Dektol? still think the DR will be the same? guess how many tones you get from that combo?<br /> No - exposure and development are key in obtaining the required DR.<br /> <br />"The characteristic curve represents the output (negative density) versus the input (exposure to light) in absolute terms".<br /> <br /> Half right. The characteristic curve is the product of three things 1. The films characteristic 2. The exposure. 3.The developer/Agitation. It is possible to have many curves (these are called a family) generated from one film by changing development or developer.<br /> Optical DR vs. Scene recording DR<br /> Optical dynamic range is the range of density recorded divided by o.3 to give the range density stops, you then MUST divide that figure by the Gamma normally 0.55 - 0.65 for mono negs and 1.2-1.4 for transparencies.<br /> So rather than dynamic range being 'fixed' it can be varied by the gamma (development time/agitation) and the exposure.<br /> Further more the effects of development and the ability to record tones varies from developer to developer some pyro developers give more separation in upper tones than other developers.<br /> For instance there are two types of DR for film 'Optical' and 'recording' One pertains to the range of densities (high Optical DR films include Velvia) in the film between d-min and d-max and the 'recording' DR which has the ability to compress a scene with a high SBR onto a film using exposure and development (mono films fall into this category)<br /> Normal SBR is around 7 but adjusting your exposure and development can give you the ability to record an SBR of over 12 (you will lose speed = EFS will probably half)<br /><br /> The whole situation is not as simple or 'absolute' as you pretend Edward.</p>

<p><em>Ever exposed Tech pan at 250 and developed in Dektol? still think the DR will be the same? guess how many tones you get from that combo?</em></p>

<p>It was an oversimplification to state that the dynamic range doesn't change. However, it doesn't change much with moderate push/pull modifications, as you can see in various technical data sheets. Dektol is extreme. I have used 1:1 Dektol on Tri-X, for basketball tournaments at ISO 1000. The contrast (gamma) is greatly increased, but the dynamic range is shortened significantly. Fine-grained film like Technical Pan tends to be contrasty anyway, so I presume the tonality is degraded (a purposeful understatement).</p>

<p>Pulling the development broadens the curvature in the toe region, which improves shadow detail. However, the maximum density is generally reduced at the same time. Nothing you do will increase the maximum density beyond a certain point - that's a fundamental property of that emulsion. When you push the film the gamma increases, the fog level is increased, the curvature in the toe is reduced and you introduce a shoulder area at the top end, which combine to significantly reduce the dynamic range.</p>

<p><em>Optical dynamic range is the range of density recorded divided by o.3 to give the range density stops, you then MUST divide that figure by the Gamma normally 0.55 - 0.65 for mono negs and 1.2-1.4 for transparencies.<br /></em><br>

The dynamic range is what you can squeeze on the film from the scene being photographed. The density range of the film is not a consideration in this regard, since it is after the fact and only affects your ability to scan or print the results. The density range is what it is, expressed in terms of f/stops if you divide the range by log2 (=0.301). Gamma has nothing to do with it.</p>

<p>The characteristic curve is "absolute" in the sense that it completely describes the behavior of the film when processed as specified. The horizontal scale is only absolute* when you have the resources of Kodak or Fuji to calibrate the light source, in order to measure the sensitivity of the film (i.e., determine the ISO value). Density, on the other hand, is a ratio which can be easily determined. Since dynamic range is also a ratio, the absolute exposure value is irrelevant.</p>

<p>* You can use a step wedge to divide the exposure range into convenient steps, but not to determine an absolute intensity at any point. Since dynamic range is a ratio, a step wedge is sufficient for its determination.</p>

<p><em>The characteristic curve is "absolute" in the sense that it completely describes the behavior of the film when processed as specified</em><br /> <br /> The key here in your argument is 'as processed as specified' think about it! When confronted with a SBR of say 7 (60:1) you'll instantly know that it will fit into your curve as long as your developer will yield a gamma that gives at least log 1.8 along the horizontal axis.<br /> Obviously larger a SBR will need a lower gamma, and it is important to remember that the effective film speed drops when you do that.<br /> <br /><em> Gamma has nothing to do with it.</em><br /> <br /> On the contrary gamma is crucial to the dynamic range (SBR and optical) you wish to record. Remember the film pushed and developed in Dektol? <br /> The point of that was you can see that increasing the gamma reduces the range (SBR) the film records (in that example to 2) this has the effect of increasing Optical DR which is measured as the spread between your speed point and D-max, expressed in stops divided by the gamma (or contrast index).<br /> So you need to be aware there are two types of DR with film one optical:<br /> http://www.astropix.com/PFA/SAMPLE1/SAMPLE1.HTM<br /> As I earlier pointed out Velvia has a very high Optical DR (spread between d-min and D-max)<br /> Then there is the dynamic range a from scene a film can hold and be useful information, this depends on your technique : exposure, agitation, developer type, strength, time and temperature. vary any of those and you will get a slightly different curve. <br /> Gamma (or contrast index) is crucial to working out your DR with respect to the SBR. I know you like to read and quote so here is a link telling you how to work out the DR in film:<br /> http://www.dantestella.com/technical/dynamic.html<br /> Edward, I think we are going to have to agree to disagree on this.<br /> <br /></p>

<p>When people talk about the latitude of negative film, I believe what is really meant is the latitude of the combination of film exposure and paper exposure.</p>

<p>i.e. if you have a film with ten stops dynamic range and a subject also with ten stops dynamic range then there is no latitude. There is only one correct exposure.</p>

<p>If that same film is used to record a scene of eight stops of dynamic range then those eight stops can be placed centrally i.e. with a stop to spare in the highlight and shadow regions or at one of the two ends i.e. with all the 'spare' two stops at the highlight end or the shadow end (or any point in between).</p>

<p>Only when it is printed does 'latitude' come into effect, taking those eight stops of range wherever they may be within the films dynamic range and translating them into a print.</p>

<p><em>On the contrary gamma is crucial to the dynamic range (SBR and optical) you wish to record. Remember the film pushed and developed in Dektol? <br /></em></p>

<p>Gamma is part of the characteristic curve - related to the slope. If you change the gamma via development, you change the curve. There is no "one, true characteristic curve". It is the product of the film and the process it undergoes.</p>

<p>You are looking at two sides of the issue (1) How to capture the scene on film (dynamic range) and (2) How to print the negative on paper (DMax and gamma). These are two separate and distinct issues, but you are switching your argument from one side to the other as it suits you.</p>

<p>Rather than spending another "thousand words", I have attached a diagram taken from the technical specification sheet for Kodak T-Max 100. I have annotated the diagram to show how dynamic range and density range are measured on the graph. You will note that there is a family of curves representing the effects of development time. Dektol is not among them (surprise). I have simplified the calculation of gamma, which is actually measured from a point in the toe region to a point in the shoulder, making it somewhat less than the gamma at the inflection point as shown. In this example, I measure the exposure from the 20% contrast point in the toe.</p>

<p>Notice that reducing the development time broadens the curvature in the toe region, which slightly increases the dynamic range (about 0.2 stops), but has no effect on the DMax.</p><div></div>

<p>I don't know why the image doesn't display - it is well within the size limits. Rather than fool around, I placed it in my portfolio under the same title.</p>

<p>I get it! Adobe Illustrator defaults to CMYK, which chokes Photo.net. Trying again with the illustration...</p><div></div>

<p>I mispoke. Development has a profound effect on the DMax. What I meant to say that it has little or no effect on the dynamic range at the high end based on this data.</p>

<p><em>"Gamma is part of the characteristic curve - related to the slope. If you change the gamma via development, you change the curve. There is no "one, true characteristic curve". It is the product of the film and the process it undergoes".</em><br /> <br /> Yes which is what I have stated all along...<br /> <br /><em> You are looking at two sides of the issue (1) How to capture the scene on film (dynamic range) and (2) How to print the negative on paper (DMax and gamma). These are two separate and distinct issues, but you are switching your argument from one side to the other as it suits you.</em><br /> <br /> No that marks your misunderstanding of the issues. The human eye has a gamma of 1. About 100 years ago it was found that when a negative was printed it often looked flat at that target gamma. After much experimentation a final gamma was decided on of about 1.5.<br /> To find how your range of tones is calculated you will need a quadrant diagram, this will give you the product of both film and paper your output.<br /> In your curve diagram from Kodak your DR you dynamic range calculations are OK ballpark wise, but you seem to be picking an arbitrary start point on the x-axis.<br /> The ISO standard for DR uses the y axis (where you state density range) divides that by 0.3 then by the gamma. You can see the results will be similar and you can also see that the gamma and final d-max ARE important <br /> You're nearly there.<br /> Why don't you accelerate your understanding by reading Beyond the Zone System by Phil Davis.<br>

<br /><i>"Notice that reducing the development time broadens the curvature in the toe region, which slightly increases the dynamic range (about 0.2 stops), but has no effect on the DMax".</i><br>

<br /> Look at the curve again, each increase in development increases the D-max by about 0.4 units.<br /> If you read the book I suggest it will show you another curve, the one that relates exposure, development to the curve in order to capture different SBR.<br /> The curve you show above is for a SBR of 60:1 which is ISO standard change the SBR and you'll need to change development and exposure in order to record the relevant SBR.<br /> Here is a link to help your understanding of how it works:<br /> http://btzs.org/Articles/HowToReadFilmTest.htm<br /> <br /> Look at fig.5 which shows a graph for Pan F (50 ISO) you'll notice that as the SBR increases to 11 (1000:1) the film speed is reduced to EI 20 at the same time the development time must go down from 7:30 to about 4 <br /> The standard curves are a good start point, but they only reference the ISO standard SBR which is 7 (60:1) or 1.8 log E<br /> That's my take on it; of course you can debate all you wish but I use the above method, have tested all my films in various developers.<br /> I have seen a 15 zone negative developed, you may doubt it and that's fine.</p>

<p><em>"I mispoke. Development has a profound effect on the DMax. What I meant to say that it has little or no effect on the dynamic range at the high end based on this data".</em><br /> <br /> I see you have edited you post Edward!!<br /> The key here is your phrase 'based on this data'! (i.e ISO standard stated SBR)<br /> Remember that it has little effect on dynamic range given the same exposure, if you increase the exposure the dynamic range increases WRT the gamma (development time).<br /> This allows us to adjust the amount of range the film can capture depending on the SBR.<br /> In simple zone parlance to are expanding or contracting then giving N-1 development, only AA considered film speed fixed (for ease of use) which he later acknowledged is not correct. Other methods use the effective film speed method EFS for capture of wide scale subjects.<br /> Read the link (btsz.org) all will become clear.<br /> N.B on the kodak curve the input (log exposure) is not plotted for this data, those curves can go further to the right with added exposure.</p>