BTZS and low contrast scenes

jdef said:

Donald,

Are you familiar with BTZS? There is a procedure by which the SBR can be extracted from a curve produced by printing a stepwedge onto the film in question, and reading the resulting densities with a densitometer, and plotting them against the densities of the stepwedge itself, which represents the luminance range of a theoretical scene. The resulting SBR value is often not an even number. By processing identically exposed films to varying extents, a range of SBR values results. SBR is just one value of many that can be targeted.

Jay

Click to expand...

This is wrong. SBR was defined correctly by Don and it is the inherent contrast range caused by the illuminance in the subject. When you plot identically exposed films to different times you obtain a range of what Davis calls average gamma, not a range of SBRs. If you wish, you can then plot SBRs against average gamma to obtain a correlation. SBR is not a value you can target, IOW you cannot say I am going to develop for an SBR of 6. AT least not in the way Davis has explained the BTZS.

jdef said:

Donald,

I do have a densitometer, and I can post the curves, although I don't have the Winplotter software, and must use Excel, with which I am not very familiar. If you have the winplotter software, I would be happy to provide the raw data, so that you can plot the cuves for yourself, and see all of the data that program generates. One caveat; I use a 31 step wedge calibrated in .1 density increments, and I don't know wether or not Winplotter can process that data.


Jay

Click to expand...

Jay, Why don't you post the numbers that you have and I am sure that someone here will be able to verify your results. I am really interested in this in light of my experience with HP5.

Kirk Keyes said:

Here's my data for the Acros if anyone with the Winplotter software would care to crunch it for me.

Fuji Acros, Xtol 1+1, 30 minutes, 20C. I get an SBR of 2.95, and a gradient of 1.19.

Step Tablet density, film density measured with blue channel
0.05, 3.59
0.19, 3.35
0.32, 3.17
0.46, 2.98
0.60, 2.84
0.75, 2.67
0.89, 2.49
1.02, 2.31
1.18, 2.11
1.33, 1.93
1.46, 1.79
1.58, 1.63
1.73, 1.44
1.88, 1.24
2.02, 1.04
2.16, 0.84
2.30, 0.60
2.43, 0.45
2.59, 0.30
2.73, 0.20
2.87, 0.16
b+f, 0.09

Click to expand...

SBR 3.6
average gamma 1.33

These are results for a paper scale of 1.45, for silver the SBR would be much lower.

sanking said:

The important thing to understand about the scene you describe is that neither a reflected reading nor an incident reading will give you the correct exposure to reproduce the tonal values as you want them without interpretation.

Click to expand...

I agree completely. Anytime you are shifting tones away from a direct approximation of the original scene requires interpretation. Never said otherwise.

sanking said:

One option is to go to one of the shadow areas of the scene and take a meter reading at that spot, note how much this EV values differs from the EV value of the first reading, and subtract from the SBR of 5 to give you a new SBR value, which could be 3 or 4. If there is not a real shadow area large enough to use for the reading you can simulate the shadows by shading the cone of the meter, take a reading, and then do the same calculation as above. Base exposure on an average of the shadow (or shaded) reading and the reading taken in full light. Or you could just double the EFS of the film and make the exposure based on the shadow reading.

Click to expand...

Thanks for this technique - this is what I'm asking about. However, I'm not sure about it. I had no shadow areas with the sand pattern or the rock art. I guess I'm not visualizing what you are suggesting to do - here's what I think you are suggesting:

1) I have a subject with no shadow areas, one that is of low contrast.
2) Take a reading in the sunlight with the incident meter at the subject.
3) Take a second reading in the shade of my hand.
4) Subtract the two readings.
5) Exposing with the average reading.

This seems like a kludge. Is there a basis in the BTZS method for this, or just something that has been found to work fairly well? It seems like you could get a lot of variation based on how much you cover the meter sensor with your hand.

Jorge said:

SBR 3.6
average gamma 1.33

These are results for a paper scale of 1.45, for silver the SBR would be much lower.

Click to expand...

Jorge, thanks! I should have mentioned my paper scale - sorry. I use 1.05. Could you re-crunch them?

Kirk

Jorge said:

If you wish, you can then plot SBRs against average gamma to obtain a correlation.

Click to expand...

OK - sorry for my misuse of terms, But this is what I was doing. It also shows that SBRs do correlate to average gamma, and therefore average gradient, and even CI.

jdef said:

Jorge, I'm surprised to see you respond, I thought I was surely on your ignore list. As I've said, SBR values can be extracted from curves derived from stepwedge prints. The stepwedge itself represents the luminance range of a theoretical subject. Don's definition doesn't recognize that the stepwedge represents a theoretical subject, the luminance range of which does not need to be metered, because it is a known value which can be measured with a densitometer.



Are you saying that the SBRs derived from these curves are all the same? If not, then they must represent a range.



I can target any value I like. If I develop a test negative from which I derive an SBR of 5, but want to determine the appropriate development for an SBR of 7, I can develop a second test negative for a period determined by extrapolation, guesswork or voodoo that I reason might result in an SBR of 7. Wether or not I achieve the targeted SBR is irrelevant, the point is that I targeted that value.

Jay

Click to expand...

Well yes you are in my ignore list, but in this case I was made aware of your erroneous post. Your first paragraph is enterily wrong. You cannot obtain a range of SBRs from a step wedge test, what you obtain is a range of average gammas. As I said, you can the plot averag vs SBR for your specific tests. IOW, an SBR of 6 is the same for you, me, Kirk or anybody using the BTZS. Development of an associated average gamma to that SBR is specific for each user.

If you are going to correct people it would be nice if you at least learn the terms as they have been explained by the creator of the system.

Kirk Keyes said:

Jorge, thanks! I should have mentioned my paper scale - sorry. I use 1.05. Could you re-crunch them?

Kirk

Click to expand...

SBR 2.6
Average Gamma 1.33
film speed 250

Kirk Keyes said:

I agree completely. Anytime you are shifting tones away from a direct approximation of the original scene requires interpretation. Never said otherwise.

Thanks for this technique - this is what I'm asking about. However, I'm not sure about it. I had no shadow areas with the sand pattern or the rock art. I guess I'm not visualizing what you are suggesting to do - here's what I think you are suggesting:

1) I have a subject with no shadow areas, one that is of low contrast.
2) Take a reading in the sunlight with the incident meter at the subject.
3) Take a second reading in the shade of my hand.
4) Subtract the two readings.
5) Exposing with the average reading.

This seems like a kludge. Is there a basis in the BTZS method for this, or just something that has been found to work fairly well? It seems like you could get a lot of variation based on how much you cover the meter sensor with your hand.

Click to expand...

The comment about interpretation was simply stated as a point of reference, not to contradict anything you previously said.

As for the technique I suggested here are some things to think about.

1. Unless you are photographing a flat object, such as a gray scale, objects in nature will always have shadow values. You may not be able to measure them because of their size, but they exist nevertheless. Even rock art has shadow values, say in the crevices between the rocks, between the lines that outline the image, etc. You may not be able to meter the shadows, either with a reflected or incident reading, but they are still there.

2. If all of the areas of the subject are illuminated by the same light, say the sky, it will not make a lot of difference whether the cone of the meter is completely covered or not, but you do want to make sure that the cone does not see the sky. This should provide a very reliable base to calculate the SBR. Have you ever noticed how little shadow values change with incident readings with the sun out or the sun obscured by clouds? Some, but very little. I think the same rationale applies here. Whether you entirely cover the cone of the meter, or only partially cover it, there will not be a big difference in the reading.

3. How well does this method of incident metering in flat scenes work? In my experience surprisingly well. The subtext of some of the messages appears to suggest that this kind of scene is difficult or impossible for BTZS incident metering. My own experience suggests otherwise.

Your resume of my suggested technique is correct as to the method for determining exposure, but not for determining SBR.

Here is what I suggested. The scene is this. You use your incident meter and take readings, but regardless of where you point it you get the same EV reading, which suggests an overall SBR of 5. To compensate, do this. Record the SBR as 5 and then take another reading, covering the cone of the incident meter with your hand so that it does not see the sky. Record the EV reading and subtract it from the value that give you the SBR of 5 suggested earlier. This will give you the adjusted SBR that will be used to determine time of development.



Sandy

"Thanks for this technique - this is what I'm asking about. However, I'm not sure about it. I had no shadow areas with the sand pattern or the rock art. I guess I'm not visualizing what you are suggesting to do - here's what I think you are suggesting:"

Kirk,
Phil goess into this in his BTZS Lite. Metering for 2 dimmensional objects. I think that may help you here.

Kirk Keyes said:

I agree completely. Anytime you are shifting tones away from a direct approximation of the original scene requires interpretation. Never said otherwise.



Thanks for this technique - this is what I'm asking about. However, I'm not sure about it. I had no shadow areas with the sand pattern or the rock art. I guess I'm not visualizing what you are suggesting to do - here's what I think you are suggesting:

1) I have a subject with no shadow areas, one that is of low contrast.
2) Take a reading in the sunlight with the incident meter at the subject.
3) Take a second reading in the shade of my hand.
4) Subtract the two readings.
5) Exposing with the average reading.

This seems like a kludge. Is there a basis in the BTZS method for this, or just something that has been found to work fairly well? It seems like you could get a lot of variation based on how much you cover the meter sensor with your hand.

Click to expand...

You are absolutely correct, as I mentioned before here is where experience counts. In the same way that many of us found in the ZS that placing the shadows in zone III many times resulted in shadows without detail, the way you place your incident meter counts a lot.

Once again and as a last post, if you check your curves when you are making your measurements you can use the effective film speed given by them when you decide to use lower than 5 SBRs, but knowing which lower SBR number to use is a crap shoot. Given that variations in film speed are very small at this range of expansion, presumably using an SBR of 2.5 or an SBR of 3 amounts to a very small error in exposure.

jdef said:

Jorge, I'm surprised to see you respond, I thought I was surely on your ignore list. As I've said, SBR values can be extracted from curves derived from stepwedge prints. The stepwedge itself represents the luminance range of a theoretical subject. Don's definition doesn't recognize that the stepwedge represents a theoretical subject, the luminance range of which does not need to be metered, because it is a known value which can be measured with a densitometer.

I can target any value I like. If I develop a test negative from which I derive an SBR of 5, but want to determine the appropriate development for an SBR of 7, I can develop a second test negative for a period determined by extrapolation, guesswork or voodoo that I reason might result in an SBR of 7. Wether or not I achieve the targeted SBR is irrelevant, the point is that I targeted that value.

Jay

Click to expand...

Jay,

Normally I would not choose to respond to your quest for individuality. However it appears that you made assumptions that are erroneous. The first is that you assumed that I did not consider the density of the step tablet. I found nowhere that I have stated any thing of the sort.

Beyond this I will no longer choose to discuss this subject with you. It is quite apparent that you have adapted the BTZS to your own frame of reference. That you use it in ways that most of us don't. It is because of this that you and I can not communicate about BTZS as it was formulated and presented in a clear and concise manner.

sanking said:

Or let me put it another way. OK, so you decide to "develop your film to a given CI", but after development you discover that the CI is different from the one to which you "developed your film."

OK, to what CI did you really develop your film? And is that question different than asking, "to what CI was your film developed"? Sandy

Click to expand...

Sandy the questions in your post here indicate to me that you (and probably many others) are having a disconnect between the act of measuring the CI of a piece of film, and developing that piece of film to a predetermined CI. They are two different acts.

sanking said:

Or let me put it another way. OK, so you decide to "develop your film to a given CI", but after development you discover that the CI is different from the one to which you "developed your film."

Click to expand...

Good question! I'm certain that happens almost every time you, or I, or anyone develops film. You've made a measurement in the field of your subjects luminance range, be it with the Zone System, BTZS, or any other system that you have devised to measure it. You go back to your darkroom, unload your film, and get ready to process it. In the past, you've done some tests and you have a table or graph that you have associated development times to your subjects conditions. So based on that measurement of the subject luminance range, you pick a developer time, temp, method of agitation... By doing this, you have decided to develop your film to a predetermined level of overall negative contrast. You then go and process that film. You do things as carefully and consistantly as possible, and because you are a good darkroom technician, you try to minimize any influences on the development process. Afterall, if you goof, you will most likely not achieve that desired level of development, your target CI.

The key here is to have good process control - and I don't mean "process" as in "film processing", I mean it in the quality control sense. All your development times are based on measurements that have been made under a (hopefully) controlled environment and set of conditions. The accuracy of your time measurements, the precision and readability of your thermometer, the strength and age of your developer, its actual pH, the remaining buffering capacity..., your ability to agitate your film in a consistent manner, the ratio of film to developer volume, the total amount of silver that was exposed on the sheet. All these things and more affect the final amount of development your film recieves - they will affect the actual CI.

Depending on how well you can do these things, you should be able to get pretty close to your desired amount of overall negative contrast. Many of us here may actually not be able to, someone that is using a large dip and dunk processor will probably be able to hit the target CI. Especially if they have calibrated the system recently.

Do you need to have measured the actual CI you set out to hit? Maybe not. Maybe one guy had a target CI was 0.60, and he did run a control strip with the run and with 5 runs got measured CIs of 0.59, 0.63, 0.58, 0.70, and 0.61. Maybe the guy with the dip'n'dunk can consistently get CI results centered between 0.58 and 0.62. And I mean consistently, centered around 0.60. The dip'n'dunk guy would be in better control than the other guy. The guy with the dip'n'dunk can come to the reasonable conclusion, that when the dip'n'dunk machine is in control, he can get a CI of 0.60, or more precisely stated 0.60+/-0.02. The other guy, cannot be as confident in his results, but even he has the beginning of some statistical basis to make a claim that he can get hit a CI close to 0.60 (the actual average was 0.62, but his standard deviation (0.05) is not that good). He needs to work on his process control.

I hope you haven't fallen asleep yet, as I'm ready to bring it together.

sanking said:

OK, to what CI did you really develop your film? And is that question different than asking, "to what CI was your film developed"? Sandy

Click to expand...

It is only really different if you do not have very good process control. If you do have good process control, you can say, with good confidence, that you did develop any particular film to some particular CI. If you have poor process control, then you may have an idea of a probably range of CIs that were achieved.

So ultimately, you will have to decide if you have good process control. If so, you can make these conclusions. And you don't need to run a control strip alongside every piece of film, either.

If you can't get consistent, then it's probably better not to.

Please note at no time in this did you have to have a full range of exposed tones on any of the film that you processed (except the ones that you actually used to make the CI calculations) - all the film will have a CI of whatever CI the run was at. Even if the film was blank...

jdef said:

Jorge, you're reaching. We are discussing SBR values derived from film curves, not SBR values as metered in the field. There is a difference. From BTZS fourth edition:

"..find the SBR for each curve by measuring the horizontal distance between its IDmax and IDmin, as illustrated in Figure 4-12. Then label each curve with its SBR in stops, as shown." The accompanying illustration shows a family of curves with a range of SBRs labeled as follows:

3.7
4.7
5.8
7.4
11

(Don, those are SBR values in tenths, used by someone familiar with BTZS)

The above is a range of SBRs derived from film curves. You can qualify that any way you like, but that's what it is, according to Phil Davis. You can go back to ignoring me now, if you like.

Jay

Click to expand...

Yeah, I will go back to ignoring you as apparently you did not understand what you were reading. What were you plotting in the paragraph you mention?

You are plotting a curve to find out average gamma, which you can then associate to a SBR value that you will obtain when taking a photograph.

My intention was to clarify this for those who are reading the BTZS and want to learn it correctly. I have no intention of continuing to argue with someone who 2 years ago was asking "what is azo paper?" and now is a master of all things photorgaphic and even names developers after himself....Unlike you, what I know, I know well, but I am not an expert on everything.....

Jorge said:

SBR 2.6
Average Gamma 1.33
film speed 250

Click to expand...

Thanks Jorge, Even higher than I thought!

By the way, your film CI is probably off, as I did not give you any exposure information, so I'll take that calculated result with a big grain of salt. I most likely did not meet the exposure conditions that Phil recommends.

Kirk Keyes said:

Thanks Jorge, Even higher than I thought!

By the way, your film CI is probably off, as I did not give you any exposure information, so I'll take that calculated result with a big grain of salt. I most likely did not meet the exposure conditions that Phil recommends.

Click to expand...

I would not know Kirk....

It seems that there are actually two discussions going on here.

First, how is a subject of low SBR metered and interpreted? Incident metering requires more experience to deal with the nuances of relatively low contrast. Reflective metering may have an advantage in that there is less interpretation required, assuming that shadow values and high values are large enough to measure (spot meter).

Second, how much exposure and development are necessary to get a firm print in this situation? How is the film developed after exposure to yield a scale sufficient to make the print fit the photographer's intent and pre-visualization? Fortunately, there are several methods available to evaluate this scenario, zone system, BTZS or a wild guess based on prior experience.

I propose a test. Someone please take a picture, develop and print it. List the SBR, zone numbers or foot candles measured to evaluate the exposure. Project a range of values, then process to a given target value. List the film, developer and whatever numbers will show that theory and practice can be effective. Finally, print the film showing the results in a finished print. (Mike Pry's "early" BTZS pond picture comes to mind at this point, well done Mike) This would illustrate the question in terms more readily understandable than the jargon being bandied about.

I know this is not nearly as exciting as talking about it, but please gentlemen, some civility is in order. It is possible to have a discussion without the personal rancor shown in some of these posts. "Deliver us from ego" is still a nice prayer. tim

"Is familiarity with a particular pritning paper a prerequisite to entry into the photographic elite?"

No, but an in-depth understanding of it is certainly a good start. tim

noseoil said:

I propose a test. Someone please take a picture, develop and print it. List the SBR, zone numbers or foot candles measured to evaluate the exposure. Project a range of values, then process to a given target value. List the film, developer and whatever numbers will show that theory and practice can be effective. Finally, print the film showing the results in a finished print. (Mike Pry's "early" BTZS pond picture comes to mind at this point, well done Mike) This would illustrate the question in terms more readily understandable than the jargon being bandied about.

tim

Click to expand...

You are on....

Any paper but toilet paper.

Well the results of my knowledge are posted......anybody else wants to back up their claims with a final print?

jdef said:

You're so macho, Jorge.

Jay

Click to expand...

Uh huh....just what I thought....c'mon big boy, put your money where your mouth is....dont tell me you are just like MS, all hot air....

uh huh........well I think this is settled.....

Kirk Keyes said:

Sandy the questions in your post here indicate to me that you (and probably many others) are having a disconnect between the act of measuring the CI of a piece of film, and developing that piece of film to a predetermined CI. They are two different acts.

Click to expand...

Kirk,

My connect is with the literature. I have looked carefully at the definition of CI in quite a number of technical books, including several devoted entirely to sensitometry, in order to better understand the issue, and I can find no example where the term CI, when defined, is not associated with a slope or gradient, which clearly shows that it is a result, not a process.

Even in cases where the term is used to in such a way that might suggest that it is a control or index number for processing, subsequent explanation always makes it clear that that CI is something to be obtained, not the process itself. I could agree that the phrase "to develop film to a given CI" describes a process, but the phrase does not redefine the meaning of CI. It describes only the reality of the phrase itself as something that takes place. And even within the context of the phrase in question the literature reaffirms the definition of CI itself as a result, i.e. a slope or gradient that defines two different points of density. Blank film, unless it is unevenly fogged, can not have two different points of density; it can not have a slope or gradient. But if you see differently we will just have to agree to disagree and let others conclude what they will.

In any event there has been much to much obsessing on this subject, so with that in mind know that this is my absolute last message on CI, at least within the specific context of this discussion. But thanks for the civil discussion, and at least I dodged some of the bullets whizzing around in this thread.

Sandy

Jorge said:

Well the results of my knowledge are posted......anybody else wants to back up their claims with a final print?

Click to expand...

Jorge,

I had the opportunity to view both an example of your work and also an example of another photographer (who shall remain unnamed) pertinant to this thread in the traveling portfolio. I must say there is no comparison. Your print was outstanding. I wish that I could say the same about the other print. However, in good conscience I cannot do that. I was absolutely dismayed at what I saw in the other print, to be completely honest.

A "quick study" apparently does not equate to good prints. There is a vast disparity between "talking the walk" and "walking the talk". In other words anyone can spout jargon...not many can turn out good work.

As I have found in my life's experience...there has been nothing that has caused me more trouble as when I thought that I knew something when in fact, as I later found out, I didn't have a damned clue.