Ilford MGRC V Deluxe has a little reciprocity error

In his book, Way Beyond Monochrome, Ralph Lambrecht states that his testing shows that papers exhibit reciprocity failure, ranging from 1/12 to 1/25 stop per doubling of exposure time. I was curious about Ilford MGRC V Deluxe, so I performed an experiment: I photographed a patch for 64 seconds (the top patch in the photo below), and six more patches at 32 times the light intensity starting at 2 seconds, and going shorter. That's a 5-stop gain in intensity coupled with a 5-stop loss of time, so the leftmost patch in the second row below should be identical to the upper patch that got the long exposure.
It isn't. There was a 0.1 stop density-difference, because the lower patches are spaced 0.1 stops apart (f-stop timing).

Thus, Ilford MGRC Deluxe loses 0.1 stops for every 5 time-doublings, which is 1/50th stop per doubling. This is a good result that exceeds the best that Lambrecht got with his papers. I see why Ilford tells us to not worry about reciprocity failure. You would only need to worry about it if exposing for 8x or more, which I doubt anyone would do without creating a new test strip.

I changed intensity by 32x using a LED-head that is PWM'd -- pulse-width modulated. I don't know whether using PWM'd bright light yields the same results as with constant-intensity dim light, but with a loss of only 1/50 stop per stop, I'm happy.

Mark Overton

The timer in the Analyzer Pro (by RH Designs) automatically compensates for reciprocity failure of paper. But neither the user's manual nor the calibration manual provides a way of disabling that feature, or of calibrating it to a paper. So I suspect the compensation-rate cannot be changed. It seems to me that that will be a problem, unless you always make a test strip at your final enlargement. But the purpose of the Analyzer Pro is to eliminate test strips, as the manual says, "The instrument can easily be calibrated to suit different materials, equipment and methods, enabling you to produce high quality prints without the need for test strips."
But if its reciprocity correction is wrong, you'd better make test strips.

Mark Overton

Have you found that the reciprocity characteristic of paper is the same irrespective of what color filtration is used? I ask this because reciprocity
failure would also be of interest to those who shoot paper in cameras and develop them as negatives. As paper has low speed (e.g. 3 or 6), long exposure is inevitable in most circumstances and exposures can easily be a few seconds to few minutes. It's interesting in this context whether reciprocity failure of paper is a function of the spectral composition of light.

Raghu Kuvempunagar said:

Have you found that the reciprocity characteristic of paper is the same irrespective of what color filtration is used? I ask this because reciprocity
failure would also be of interest to those who shoot paper in cameras and develop them as negatives. As paper has low speed (e.g. 3 or 6), long exposure is inevitable in most circumstances and exposures can easily be a few seconds to few minutes. It's interesting in this context whether reciprocity failure of paper is a function of the spectral composition of light.

Click to expand...

Well....
Technically, reciprocity failure (of this type) isn't caused by the length of the exposure, but rather by the low intensity of the exposing light.
It certainly seems possible that the reciprocity characteristics of the different emulsion components (blue sensitive, green sensitive, cyan sensitive) would differ.

Raghu Kuvempunagar said:

...whether reciprocity failure of paper is a function of the spectral composition of light.

Click to expand...

As mentioned in a separate thread - reciprocity and intermittency aren't an issue with the normal exposure of printing paper:
http://www.darkroomautomation.com/support/appnotereciprocityandintermittency.pdf

Effects can be caused by improper compensation of lamp warm-up/cooldown:
http://www.darkroomautomation.com/support/AppNotePH212LampDelay.pdf

If you really have reciprocity failure - say a light level that meters to a 10 second exposure on Tri-X - then the HD curve of the material will change. The shadows loose detail while the highlights are not affected. As Matt mentioned above, reciprocity failure isn't a function of time but of light intensity at the film plane.

Reciprocity effect will show itself at the margins of a material's spectral sensitivity - but here we are talking about being at the margins of the margins.

There is quite a good book published on reciprocity failure, by Fuji I think, but I can't put my hands on it. It's somewhere ...
If nothing else, there is the Wikipedia article: https://en.wikipedia.org/wiki/Reciprocity_(photography)

Nicholas Lindan said:

As mentioned in a separate thread - reciprocity and intermittency aren't an issue with the normal exposure of printing paper:
http://www.darkroomautomation.com/support/appnotereciprocityandintermittency.pdf

Click to expand...

The report linked above includes photos of test strips proving there is no reciprocity failure over an 8-stop range. Yet my test showed slight RF over 5 stops; Lambrecht's book states that his testing revealed RF in the range of 1/12 to 1/25 stop per time-doubling; RH Designs incorporated RF-compensation into their timer, so they apparently observed RF. We have a paradox.

BTW, here are two informative threads on this topic from around a year ago. I started those threads, but had forgotten about them.
https://www.photrio.com/forum/threads/paper-reciprocity-failure-after-only-30-seconds.182370/
https://www.photrio.com/forum/threa...-for-paper-before-failing-reciprocity.179638/

One thread describes what Matt mentioned above: RF is caused by low light-intensity, and is thus acronymed as LIRF -- Low Intensity Reciprocity Failure.
Mark Overton

From what I read about reciprocity, unlike film, paper has two kinds of reciprocity failures - one that affects speed and the other that affects contrast. Here is a snippet from a post by PE on this:

Photo Engineer said:

Paper has 2 kinds of reciprocity failure, speed and contrast.

In speed reciprocity failure, the paper loses speed with exposures much above about 1 minute and it gains speed below about 1 second.

In contrast reciprocity failure, contrast goes down with longer exposure and up with shorter exposure.

These effects vary vastly among papers, but can be disastrous with VC papers where you are trying to design matched reciprocity failure between 2 emulsions.

Click to expand...

Regarding long exposures and reciprocity, here is a snippet from a previous thread:

Nicholas Lindan said:

Reciprocity failure isn't a function of time but of light intensity falling on the media. But in photography the light intensity falling on a given media is proportional to the shutter speed/exposure time: we want the integrated exposure to be constant, so as the light intensity on the media falls the exposure time increases. And so time becomes a convenient stand in for the intensity of the light falling on the media

Click to expand...

I think all this is interesting in the context of paper negatives, but that's beyond the scope of OP. Sorry for the digression.

Nicholas Lindan said:

Reciprocity failure isn't a function of time but of light intensity falling on the media. But in photography the light intensity falling on a given media is proportional to the shutter speed/exposure time: we want the integrated exposure to be constant, so as the light intensity on the media falls the exposure time increases. And so time becomes a convenient stand in for the intensity of the light falling on the media

Click to expand...

albada said:

my test showed slight RF over 5 stops; Lambrecht's book states that his testing revealed RF in the range of 1/12 to 1/25 stop per time-doubling; RH Designs incorporated RF-compensation into their timer, so they apparently observed RF.

Click to expand...

Well, that's why different people perform (hopefully) the same experiment over and over until they all get the same result.
Doing the same thing over and over again trying to get a different result is the life of the research worker; or the life of a gambler.

I did a whole mess of reciprocity/intermittency tests when coming up with the Darkroom Automation products. There is nothing worse than getting to the end of a long product development cycle and then discovering "Oh! The physics don't work!" The reports on the DA website show the results of those tests.

I do know there is no point doing the test unless you have lamp warm-up-cool-down compensation set right. It was at this point in tearing out my hair that w-u-c-d compensation got added to the timer. This test does presuppose that intermittency is not a problem.

albada said:

"We have a paradox."

Click to expand...

I believe there are no paradoxes. It's just the word we use when knowledge is hidden. If we knew all the truth of the matter there would be no paradox.

We all have confidence in our results. I'm not sure how to resolve this, though the passage of time often presents with the answer to the question (or shows the question useless).

I think we can agree that the difference in our two experiments if of no practical importance.

It is hard to disagree with Ralph Lambrecht, but I'm afraid I have to. The results shown in the book show a set of parallel HD curves marching to the right along the time axis. If reciprocity failure was the mechanism then the shape of the HD curve's toe would change - well disappear, really - before anything happened to the shoulder; it is the same mechanism as the loss of shadow detail and increase in contrast when film exhibits reciprocity failure.

What R&H put in their gear is a mystery to me. Chris Woodhouse, Ralph's coauthor, worked on the RH designs products - it may be the same reciprocity results were incorporated into their Analyzer.

Raghu Kuvempunagar said:

From what I read about reciprocity, unlike film, paper has two kinds of reciprocity failures - one that affects speed and the other that affects contrast. Here is a snippet from a post by PE on this: ...

Click to expand...

Hmmm, not too sure about this. Any change in reciprocity due to illumination would also show as a change in contrast. The toe of the HD curve will exhibit reciprocity effect before the shoulder - the illumination at the toe is 5-7 stops dimmer than the shoulder and therefore the toe will encounter "illumination reciprocity" first - thus changing the shape of the HD curve.

I read a book of research out of Fuji (?) on reciprocity effects last summer and I can not lay my fingers on it.

I've seen a graph in the book Basic Photographic Materials and Processes (Figure 5-42 in page 182, 3rd edition) which mentions and shows contrast reduction as a consequence of reciprocity failure. The book, however, does not explain why the contrast goes down unlike film where reciprocity failure results in increase in contrast. So paper is different from film in this regard.

Raghu Kuvempunagar said:

I've seen a graph in the book Basic Photographic Materials and Processes (Figure 5-42 in page 182, 3rd edition) which mentions and shows contrast reduction as a consequence of reciprocity failure. The book, however, does not explain why the contrast goes down unlike film where reciprocity failure results in increase in contrast. So paper is different from film in this regard.

Click to expand...

In my 1990 edition this is figure 4-53, pg 124 - or at least I hope it is, if we are cross-talking about different figures/chapters things are going to go wonky in a hurry.

The loss in contrast due to very long exposures when making prints may not have anything to do with reciprocity failure. For loss of contrast to happen the sensitivity of the paper would have to increase at lower illumination - something which would be very strange indeed.

Loss of contrast in long exposures, especially when making very large prints, is due to stray light. The stray light coming off the enlarger is constant and doesn't depend on the negative density, f-stop or print size. However, as exposure time increases the effect of this light on the paper increases - leading to veiled highlights and muddy tones in general. This effect is especially pernicious when making large prints - most of the stray light isn't from the enlarger itself but from the light bouncing off the paper, illuminating the darkroom, bouncing off the walls and ceiling and then in turn re-illuminating the paper. Again, the intensity of this light is constant with print size and so as print size increases the effect of this stray light increases. Ansel Adams complained about the loss of contrast when making murals in his book The Print.

Nicholas Lindan said:

The loss in contrast due to very long exposures when making prints may not have anything to do with reciprocity failure. For loss of contrast to happen the sensitivity of the paper would have to increase at lower illumination - something which would be very strange indeed.

Click to expand...

Strange it may seem, but the book I mentioned earlier as well as PE have talked about the contrast reduction effect of long exposure reciprocity failure in photographic paper. I believe they had access to reports of research studies on this subject and were not misattributing the cause of contrast reduction to reciprocity failure. Unfortunately, I am not able to find any research paper on this topic yet and I'll keep searching for it. As you rightly said, this phenomenon might not be of any practical concern in the darkroom.

Edit: Quoting from the book:
"In addition to a loss in speed with printing-paper reciprocity failure, one can also expect a loss in contrast. Whereas one can expect reciprocity failure to cause increase in contrast for films exposed for relatively long times (low illuminances), a decrease in contrast for print papers can be expected."

I ran more tests very carefully today to determine (1) does reciprocity-failure (RF) occur in today's papers, and if so, (2) does constant-light versus PWM-LED make a difference.
The answers from these tests are:
1. Ilford has very slight RF, consistent with the first posting to this thread,
2. FOMA is worse, and agrees with Lambrecht's testing,
3. Both papers behave the same in constant-light and PWM-light.


Each paper has three rows in the image above. The middle row was exposed for two seconds under bright (non-PWM) light. The top and bottom rows were exposed for 64 seconds under dim light provided by PWM (top row) and constant-light by closing aperture (bottom row). I used the light meter from DarkroomAutomation to verify that the dim brightnesses were equal, and that the bright light (middle row) was 5-stops brighter. These conditions were true within +/-0.03 stops, so these results should be accurate.

For Ilford MGRC V Deluxe (upper three rows), there is only about 0.1 stop loss due to RF -- insignificant.
For Foma RC, there is about 1/3 stop loss over the 5-stop change, which is same as the typical RF-loss that Lambrecht reported in his book, Way Beyond Monochrome.

Conclusion: I like Ilford.

Mystery: If RF is caused by low light-intensity, then why does it occur with PWM? PWM creates pulses that are short and bright, not dim.

Mark Overton

albada said:

Mystery: If RF is caused by low light-intensity, then why does it occur with PWM? PWM creates pulses that are short and bright, not dim.

Click to expand...

There are two types of reciprocity failure - the above referenced LIRF - Low Intensity Reciprocity Failure - and the less commonly encountered HIRF - High Intensity Reciprocity Failure. The latter arises when the exposures are of very short duration and (usually) of high intensity.
The roll papers that Harman sells are stated to be designed for digital exposure. In essence, they are optimized for the sort of exposure that the Lambda printers use - very brief lasers - and they are designed to avoid the HIRF.
There may be some latency effects with PWM that mean that the actual exposure is less than the brightness would indicate, but otherwise I expect that the problem is with HIRF.
A relevant thread from 12+ years ago: https://www.photrio.com/forum/threa...reciprocity-failure-explanation-please.50636/

albada said:

I ran more tests very carefully today to determine (1) does reciprocity-failure (RF) occur in today's papers, and if so, (2) does constant-light versus PWM-LED make a difference.
The answers from these tests are:
1. Ilford has very slight RF, consistent with the first posting to this thread,
2. FOMA is worse, and agrees with Lambrecht's testing,
3. Both papers behave the same in constant-light and PWM-light.

Ilford+Foma Reciprocity-Failure Tests

• albada
• Jan 2, 2022

These tests show that (1) Ilford MGRC V Deluxe has only slight reciprocity-failure over 5 stops...

Each paper has three rows in the image above. The middle row was exposed for two seconds under bright (non-PWM) light. The top and bottom rows were exposed for 64 seconds under dim light provided by PWM (top row) and constant-light by closing aperture (bottom row). I used the light meter from DarkroomAutomation to verify that the dim brightnesses were equal, and that the bright light (middle row) was 5-stops brighter. These conditions were true within +/-0.03 stops, so these results should be accurate.

For Ilford MGRC V Deluxe (upper three rows), there is only about 0.1 stop loss due to RF -- insignificant.
For Foma RC, there is about 1/3 stop loss over the 5-stop change, which is same as the typical RF-loss that Lambrecht reported in his book, Way Beyond Monochrome.

Conclusion: I like Ilford.

Mystery: If RF is caused by low light-intensity, then why does it occur with PWM? PWM creates pulses that are short and bright, not dim.

Mark Overton

Click to expand...

there are low- and high-intensity reciprocity failures.

RalphLambrecht said:

there are low- and high-intensity reciprocity failures.

Click to expand...

The "high intensity" of my PWM is the same intensity as was used in the middle rows of those tests. This is because, with most LED drivers, LEDs are either on with a fixed brightness, or off. So the high intensity is not especially intense, and if HIRF did not affect the middle rows, it should not have affected the top rows that used PWM. Strange.

In case anyone cares, I chose a PWM frequency of 122 Hz because it's high enough that moving a dodge/burn tool should not leave tell-tale lines, and low enough to avoid extremely short exposures. My shortest is a PWM of 1/64 (6 stops of dimming), which results in a pulse-width of about 1/8 millisecond.

Anyway, it's good to know that a PWM light-source doesn't appear to hurt anything compared with tungsten.

Mark Overton