hp5 Reciprocity Effect Question

[patrick_gainer]

The data I see are not the data that Mees or any of you have worked with, I gather. I don't understand why you are so dead set against the empirical curve fitting. If the form of the equation I chose to use did not fit well, and in fact almost exactly to the Ilford published form, I would not use it. The fact that I do not propose a theoretical basis for that form or that someone else used it before me without success has no bearing, nor should it. If I am going to do research to fill in the blanks, I would rather they were my own blanks.

 

An actual theoretical curve of that type would probably involve the Boltzmann constant, the absolute temperature, and one or two other physical constants. It might be that if I knew what combination of those to put in the equation, you would call me a genius instead of just an engineer.

 

Ryuji, I have studied the excerpts you sent me over and over. I keep them beside my bed. I really appreciate your present because I live in the boondocks with only a small college library at hand, and enough arthritis that I find it difficult to walk up the hill from the parking lot to the building.

 

To make matters worse, I had a year or so ago an attack of viral meningo-encephalitis that took away some of my memory accessing ability. I think the memories are still in there, but the access paths have to be reestablished. My favorite computer card game is Freecell. Before encephalitis my average was 84%. It is just recently coming back to 77%.

 

Ron, no matter how many famous people he knows, has decided to take issue with everything I post. In this case, neither he nor you know the extent of what I have written, nor my rationale for writing it. I had arrived at the form of the equation I used long before you sent me the excerpts from The Theory of the Photographic Process. At the time, all the data I had was what Kodak supplied in one of the handbooks. Nothing I have encountered in the meanwhile has done anything but lend statistical significance to that form. So, when you tell me it won't work because someone else says it won't, I remember that visiting professors from our most prestigious technical university were asked to comment on some supersonic wind tunnel data, they said the pressure tubes must have been clogged because the data did not agree with linearied potential flow theory. They did not notice that the wing sections being tested had blunt leading edges which were never considered by the originators of the theory to be within its scope. I found that the theory could be put to use if it was considered that there was a detached shock wave ahead of the leading edge that changed the parameters considerably.

 

Great brains do not always have great ideas to work with. People with great ideas do not always have great brains with which to make them work. If you want to contribute positively, you could take the data I have available, combine it with what you have, and see which theory fits best. At thw age of 77, I don't have enough time left to do it.

[rowland_mowrey]

Patrick;

 

I don't object to your work or your presentation.

 

I do object to your not looking up the prior art. I understand that you cannot get to a decent library and don't have one of your own, but that is what the internet and google.com are for. You could research things.

 

I do indeed know all of the individuals mentioned by Ryuji above. Tadeki Tani works for Fuji, not EK though (last I talked to him anyhow). It is not that I did not know the material, it is that I, like you, forget things and had to re-read the text.

 

This points to the last thing I don't like in general. People who have a good education and a liking for photography believe that makes them photographic engineers. Patrick, if I had an idea regarding aeronautics or something in your field, I would have the work VETTED by an expert in the field, and have it done privately, or if published, I would expect comments like mine given to you by experts in the field who know or can easily look up the material.

 

And, BTW, having the work vetted by an expert does not give it a stamp of approval as I have also seen in published works. Having been asked by experts to vet their work, I know personally what can happen at both ends of the deal. I have carried a briefcase full of galley proofs home every night reading and editing material to be published by coworkers. I have read and written hundreds of technical reports on photographic materials that had to be completely cross refernced. My critics were mainly supervisory and I had to learn to take their comments stoically, somtimes in public at meetings of my peers.

 

If you can't stand the heat, then get your work vetted or do some research of the prior art on the internet.

 

Your work is correct. It is merely a reiteration of facts long known in the prior literature. You and Bond didn't know that. You should have if you wanted to publish it, so that you could avoid any possible taint of plagiarizm. And, BTW, I accused no one of that. I merely feel that you were uninformed on this subject just as I had forgotten about it.

 

Ron Mowrey

[ryuji_suzuki]

"I do object to your not looking up the prior art. I understand that you cannot get to a decent library and don't have one of your own, but that is what the internet and google.com are for. You could research things."

 

Aren't there good document service companies? Internet is full of crap and noise and some good stuff are embedded among those. A lot of things I learned about photography are those I could only find in print media... so I'd think a good library is a must, unless you have a very competent consultant... of course google is good for a starting point though.

 

"This points to the last thing I don't like in general. People who have a good education and a liking for photography believe that makes them photographic engineers."

 

I agree, there are a lot more than what's printed in popular photo manuals and Kodak brochures. This fact makes me worry if I know enough and the answer is always no. If you think yes, good knowledge ceases to grow and bullshits beging to self regenerate on the internet. Actually, much of good stuff on the internet is pirated without proper credit and with added crap and found on other sites, forums, etc.

[patrick_gainer]

Ron,

you said "Your work is correct. It is merely a reiteration of facts long known in the prior literature. You and Bond didn't know that. You should have if you wanted to publish it, so that you could avoid any possible taint of plagiarizm. And, BTW, I accused no one of that. I merely feel that you were uninformed on this subject just as I had forgotten about it."

I beg to differ. Kodak furnished the film for Howard's project. The data he published was not to be seen anywhere else. The data in Ryuji's dissertation on reciprocity is not in tabular form. From what I can tell by trying to extract it from the graph, the equation I have been using fits it well.

 

I did not tell you all I know. If you want to be a peer reviewer of my work, I will send you the complete thing. I know the scientific method theoretically and from practice, and it galls me a little to be lectured about it by someone who has only seen my informal comments. I would like to carry on a corespondence on these subjects, but not in this forum. My e-mail address is pgainer@rtol.net. My snail mail is HC77 Box 86, Glenville WV 26351.

 

I have ordered a used copy of Theory of the Photographic Process, 1969 edition, so I can see if it says anything I didn't know about this subject.

 

Ryuji, I know you to be a man of many talents and interests. I cannot however get what I want to know from the plot of reciprocity effects at your web site. I need numbers in a spreadsheet sort of array so that I can filter out the indicated exposure. My equation only deals with the correction to be added to the indicated exposure. The adjusted expoures do not plot as a straight line on log-log paper. I take it those are experimental data you plotted. I would appreciate it if you could send me the numbers that went onto the plot.

 

I will be glad to send either or both of you copies of anything I plan to publish, and I will take seriously any genuine technical criticism as I had to do at NACA-NASA. it's a little to copious for e-mail or this forum, so it will have to be by snail mail.

[rowland_mowrey]

Patrick;

 

I don't want to review your work, nor that of anyone else. I merely wanted to comment on the fact than any publication requires diligent research into prior art, citations, and peer review.

 

In your last post, I didn't understand this statement:

 

"I beg to differ. Kodak furnished the film for Howard's project. The data he published was not to be seen anywhere else."

 

Once published, how could it fail to be seen everywhere the publication was available, as well as where you published data and plots yourself(namely APUG.ORG). And, how is that connected with the fact that all of us should check for prior art and cite references?

 

When I give a formula, or even refer to a particular piece of technology, I research the patent literature or my own library of photographic text books. I often sit here typing with the reference sitting next to me and open to the page, but I sometimes, often as not, refer the readers to the literature encouraging them to read (an oft forgotten art among the young).

 

You did good work, but you didn't research its background. Others have the priority date of publication and you didn't even know it.

 

I had to research the prior art for all of my inventions at EK and submit a patentability report before we went ahead and spent the big bucks on trying to get a patent. Nothing makes a person look worse than to have a great invention that was patented or published by someone else, and you failed to find their work.

 

If you look at any patent, you will see a list of 'cited patents' at the head or tail of every patent that lists those patents that the US Patent office found as being very close to your invention. In each case, a separate work had to be done to prove that we did not copy their work and that ours was different or better than the cited work.

 

One of my most lengthy pieces of research at EK was just too close to another publication and so our attorneys judged that we should not go ahead with the attempt to patent it. OTOH, they judged that it avoided the other patent quite easily and we were free to use my work in a product.

 

This was only possible by reading the literature. I recommend it to you and hope you enjoy Mees book.

 

With sincere high regards.

 

Ron Mowrey

[rowland_mowrey]

To all;

 

A comment was made above regarding my statement about using the Bond-Gainer method (or the Kron equation if you accept the priority of the 1913 publication). I suggested caution.

 

Here is the reason. Using a Bond-Gainer plot, an error that is less than 1 second in calculation using incorrect constants will result in errors of up to 1 stop (~100 seconds vs ~200 seconds) at longer exposure times.

 

The plot is so poorly reproduced in the graphic posted on APUG that I cannot tell you which films among the 5 cited that this represents. I do encourage those interested to look at the plot first hand and see for themselves the fact that the graphs are very close together at 1 second and diverge at longer and longer exposure times (an expected result). The problem is that a tiny error in your derivation of the constant and the correct exposure time at one value can lead to a one stop error at the other end of the scale.

 

The bottom line is to use it with caution, and try to stay below a 10 second exposure time to minimize errors. OTOH, at 10 seconds, the errors over all 5 films is less than ~1/10th of a stop which is hardly worth worrying about and therefore any of the 5 equations will give a close approximation for any of the films.

 

At short exposures, one curve fits all, and at long exposures, any tiny error is magnified by up to a stop. This is essentially what is discussed by Mees and shown in his graph cited by me above.

 

Use the Bond-Gainer (Kron) method with caution. Don't use this method with color film.

 

Ron Mowrey

[patrick_gainer]

"And one can certainly take pictures without knowing all these. So can he take pictures without fallacious statements." Or even truthful statements. My whole object in this subject from the start has been to provide a simple basis for finding your own reciprocity corrections when they are not available elsewhere for some film you may be using, and when you do not own a densitometer, and when the film you are using is a precious commodity. You have some in the freezer that is no longer being manufactured and you want to maximize the probbility of getting good pictures of caves and cathedrals. I never, ever claimed to be trying to break new theoretical ground. I found an equation by commonly used analytical methods that was a close fit to the data I had at hand and contrived a method for using that equation for the purpose at hand. That method has been used since we first started imagining that Nature followed rules. My observation that the equation, with only two arbitrary coefficients, fitted available data

and that one of the coefficients seemed to be the same for all films tested made it easier for anyone to find the necessary algorithm to define with photographic accuracy the corrections for long exposures.

 

Now tell me again what we are arguing about. You say I should look at the work of others. The others were not interested in simplifying anything. The data I found supplied by Ryuji cannot be used as published for testing my method, because they are in the form of a plot of poor resolution and numerical data are not tabulated. Furthermore, the plot is not in the form that I would make to illustrate any method, certainly not mine.

[patrick_gainer]

Ron,

I said"I beg to differ. Kodak furnished the film for Howard's project. The data he published was not to be seen anywhere else." A bad choice of words on my part considering current diction. I did not mean the imperative, I meant tha actual. The experiments Howard did contained data that no one else had published at the time. He probably saved Kodak and Ilford 2 or 3 bunches of money by doing the experiments. You really ought to read his paper in Photo Techniques, July/August 2003 and mine in the SeptemYou saidber/ October issue.

[patrick_gainer]

I don't understand why you refer to my stuff as "The Gainer-Bond Method". It is a generally accepted engineering method. You do not seem to know the meaning of engineering. According to one of my former professors, "An engineer can do anything." Engineering is a generic term for problem solving. When I studied aeronautical engineering, most of our courses were in the mechanical engineering department. We studied all kinds of things, including surveying and internal combustion engines. I had a number of chemistry courses that served as electives.

 

At NACA, the first thing I did of any note was to save the overtime efforts of some others who had collected miles of oscillograph recorings of flight data and were copying the stuff on an Ozalid machine. Do you remember those? It turned out that the developing part of the reproduction had not worked. All you could see on the Ozalid paper was a weak yellow line. I knew from past experience that Ozalid is developed by ammonia vapors, so got some household ammonia from the janitor's closet and a small paint brush, thinking to save these guys from more night work. As soon as I got the brush near the trace, the black line popped out. Soon after, the large bottom drawer of each desk had an open container of ammonia and a loosely coiled roll of ozalid paper. Also, everybody's sinuses were clear for days. That escapade earned me the nickname "Gadget Gainer."

 

I learned photography on the job as well as by my hobby. I learned it as a tool for solving problems at NACA and also solved some photographic problems. All the time I played the oboe in the local symphony orchestras I took photos of guest artists and other orchestra members. I was in effect a working journalistic photographer. Don't insult me by implying that I just recently began looking at photography as both art and science.

[gaius1]

If you want to call it the Hammond Method instead, that's fine by me :-)

[rowland_mowrey]

Patrick;

 

The graph posted on APUG.ORG was labeled "Bond-Gainer Reciprocity Chart". That is why I used the wording I did - to be familiar to those who read those posts. (not that you used that wording)

 

I know you have worked with photographic materials for years. I don't deny that. You also have done a lot of good work. You just are not a "Photographic Engineer". I build model trains, but am not qualified to repair or run a full scale engine, nor be a brakeman etc. etc. Regardless of how much I know about real world trains I am not qualified for running or controlling them.

 

I hope you see the analogy there that I am trying to make.

 

I also hope to get you to understand that if you take the responsibility of publishing data, you give credit to the developers of that material in some way and that you have a responsibility to check out priority of publication.

 

Perhaps I'm overly sensitive to this, but I have seen work of mine and of friends of mine quoted in articles by others with no references to the original patent or publication. It is discouraging to work 2 - 3 years on something, only to see it published in detail in a photo magazine taken right from the patent and see no credit given to the originator of the work. The writer of the articles in mind have higher reputations now in the amateur photo world than the originators of the work.

 

Best wishes.

 

Ron Mowrey

[patrick_gainer]

Ron, I take it by your reply that you have not read either Howard's article in the July/August Photo Techniques or my article using his data, with credit and compliments to him, in the September/October issue.

 

If you would like, I can snail mail you the graph that you had a hard time reading. I was limited by the size of what I could transmit. Or, you can see it in the article.

[ryuji_suzuki]

General engineering method is fine. I'm an engineer as well. But don't put your name on it until you have done very thorough search on literature and prior findings. Stealing someone's credit is a very bad thing in fields related to science, even if you are unaware of it. It just happens that most casual photographers reading your magazine may not notice or don't even bother to think about prior work, but people like Ron knows and that's why you see him getting very irritated when you write something.

 

Since you are actively publishing something, please keep in mind that a lot of "photographic science" in popular media contain a lot of misconceptions and we don't need to worsen this trend. We have solid knowledge in academic journals. It is true these journals are a bit beyond average photographers but as an author you should make every effort to make sure your writing is accurate and not stealing credits of prior work. On the other hand, if you study prior work and try to translate them to average photographers in a very accessible way, even if it requires some watering down, would be highly appreciated, even if you admit none of the knowledge presented is your work. I would still have that manuscript vetted by a few true experts before sending it to print, though. Lack of such a tutorial work may be one reason why popular photographic publications are so full of BS.

 

I made that page a long time ago with no relation to this discussion or your work. I certainly don't like that you now criticize me because my presentation of data is unsuitable to your generic engineering analysis.

[rowland_mowrey]

Patrick;

 

Thanks for your kind offer. I missed the issue with Bond's article, and IIRC, I missed the issue with yours as well.

 

I will pass on having you go to the personal expense of sending me a copy as the material is not very applicable to my work. I usually use color in night shots to pick up the varicolored lights, and using color with your method, as you and I both say above, is not warranted.

 

Regards.

 

Ron Mowrey

[patrick_gainer]

Come on, Ryuji. I stole nothing from anybody. The curve fitting that I did was from my personal experience. It is standard engineering practice. It has nothing to do with the essence of reciprocity relationships. Neither, I must say, does the work you presented at your web site. It doesn't help me in any way to know what comes next or give me a guide for reducing the number of tedious reciprocity experiments I must do to nail down an untested film.

 

If I make the observation that the films I tested are fitted by an equation of a certain form, it is because I saw that in the available data. If I surmise that one of the two arbitrary constants in the equation may be the same for all films, you may surmise that it is not, but I still maintain that there are only two and that two carefully performed experiments will make them known.

 

If I see a particular systematic variation in the available data by using common engineering methods, how much credit is due another person for doing the obvious? You cannot patent common sense.

 

I would still like to see the original data that went into your plots. I have no need to ridicule you for not presenting it. I would simply present the numbers verbatim in a different format to test my assertions. The exponential part of the equation I use applies only to the correction for reciprocal misbehavior.

 

I presume that you also did not read the data report by Howard Bond in the July/August issue of Photo Techniques nor the analysis I made in the September/October issue. You should know that by referring to his work, I also referred to the works that he referred to.

 

You have changed much since we used to correspond.

[ryuji_suzuki]

Right, my web site contains only a little new results. not much. when it comes to reciprocity failure, i know i can't add anything new to the fundamental understanding so it's a restatement of existing knowledge and i didn't hide that. do you know what kinds of things need to be measured? if you only need superficial phenomenological behaviour, you may only need sensitmetry, but in order to get deeper insights, people need to measure the lifetime of photoelectrons, dielectric loss with various microwave frequencies, all these at different temperatures, and a lot of stuff. way beyond me. i can barely interpret pieces of relevant results. all i did was to present the abbreviated mechanism and practical way to deal with LIRF in the field, on the same page. even that, i feel that the distance between theory and practice is a bit far, in taht particular case.

 

i am admitting (more like appreciating) that all fundamental works were done by those geniuses and i'm here a drunk who's sitting with a bottle of bourbon flipping through xerox of what peopl ewrote decades ago, thinking wow how they could figure all these out then when even the high end ph electrode was so crappy and amplifier was drifting like hell. believe it or not, capstaff didn't have a such a nice ph electrode, when he came up with d76, which we can buy for the price of a few parking tickets, and crabtree figured all (-most) out about fine grain develoeprs without using one. same for those people who came up with models for the formation of latent image centers without all sorts of measurement devices we have today (well, today in rochester, ashigara and leverkusomething, not in my livingroom).

[patrick_gainer]

Now you'rs talking. I don't have bourbon, but I do have some excellent Elderberry wine courtesy of a local vineyard and winery.

 

I do read as much as I can find. Generally, I go off on a problem because I saw that someone else has tha problem.

 

When you talk about "vetting" I presume you mean peer review. I suppose you know that NACA-NASA had, and I hope still has, a rigorous review process. Any paper had to go through the branch head and tha division chief, who actually did read the stuff, and if it passed them it went to peer review by a committee from the originating research center. I served on many of those committees, often as chairman. Simultaneously, it went to another research center, usually Ames for my stuff. After revision by their command, it went to the technical editing section where we often found that what seemed so clear to the researchers meant nothing to the editors.

 

I have no such network to keep me on the high road. I depend on these forums. If I leave out an important source, I would like to be told, even if it means everything goes in the circular file. However, in the case of reciprocity, I think I am entitled to view the available data and present it in a form that is more useful, if possible, to the photographer who does not live by theory alone.

 

I probably should go get a glass of that wine. They also make a fine Black Currant wine in addition to more elegant wines.

[ryuji_suzuki]

"I have no such network to keep me on the high road. I depend on these forums."

 

I doubt forum can be a substitute for that. It's simply so much crap. Reading someone's manuscript takes a lot of work especially if you expect detailed feedback. You don't expect that from som e random people.

 

"However, in the case of reciprocity, I think I am entitled to view the available data and present it in a form that is more useful, if possible, to the photographer who does not live by theory alone."

 

I frankly think what Kodak, AGFA or Fuji publishes adequate for practical photography and I don't know how an emprical fit benefit in the field...

[rowland_mowrey]

Patrick;

 

We used to measure at least 3 or 4 points on a curve to get a true value for reciprocity due to the fact that the entire curve shape changes as a function of LIRF and HIRF. This is one of the problems with any prediction.

 

In older films it was due to the polydispersity of the emulsions used, and in modern films which are more monodisperse, it is due to blending up to 3 different emulsions in one product (or in one layer in color for 9 emulsions).

 

So, as a result, curves can become straighter, more serpentine etc. and this does not show up in anything less than a family of curves for each film taken at the toe, midpoint and shoulder at the minimum. I higly recommend this to you and Mr. Bond in any future work you might contemplate.

 

As for vetting. I'm glad you agree that this forum can be used to help. I just hope that you and others do not become 'testy' when anyone points out prior art or minor errors, or for that matter - simply disagrees with you. I must say that several individuals have gotten quite upset with me when I dissagreed with them on some points even though I have far greater experience than they in Photographic science and engineering.

 

These individuals cite others. They have no authoritative work of their own to cite. In some cases, I can cite my own patents as references, in others, I can cite from my vast home library of chemistry and photography or I can call an associate on the phone or just walk next door for an authority. If you need help, I have hundreds of books on photography and chemistry sitting in the next room.

 

So, in this respect I agree with Ryuji. Please read the literature more. Read Mees or Mees and James. Read Haist. Read back issues of the SPSE journal. I have read all of these books cover to cover at one time or another and used to read the SPSE and SMPTE journals until they became mostly digital. I used to read JACS, JOC, and many foreign journals.

 

We both are experts but in different fields. Please heed my comments regarding my own area of expertise.

 

Thanks.

 

Ron Mowrey

[ryuji_suzuki]

I agree curve shape changes and setting one arbitrary speed point is a great simplification. Mostly toe in LIRF and shoulder in HIRF. At the same time, there is lens flare and other elements that overlap with sensitometric data in actual photography. Trying to be too exact on this issue for practical application is not very fruitful, especially because we have a few excellent products on market due to tabular grain technology and advanced dopants. T-MAX 100, Fuji Acros and several color negatives have almost no reciprocity failure within the range of landscape photography. I hope we will see another generation of b&w negative films using hole trapping dopants as well, which will increase speed and also greatly diminish the LIRF. With Acros which I use very frequently, even with a very long exposure, a bit of overexposure can take care of it. All these emulsions have very large overexposure latitude. The rest can be adjusted in darkroom when printing. This is the virtue of the negative-positive system.

[rowland_mowrey]

Ryuji;

 

That is a good point about flare.

 

My tests were done using a zero flare optical train to eliminate these effects and get at the pure characteristic curve. One should run two sets of tests. One on an low flare optics bench for exposures and one in-camera for a measure of flare which must be included. If one changes cameras and / or lenses, then the data becomes much less applicable to data derived with other equipment.

 

Patrick;

 

Do you have any comments / additions regarding flare?

 

Ron Mowrey

[patrick_gainer]

Do you have any comments / additions regarding flare?

 

Ron Mowrey

Avoid it like the plague.

 

You erudite people have judged me on no basis whatever. Sorry if I seem incensed. I am. You refuse to read what I published and you act as if you had. If you will read the article "Reciprocal Trade Disagreement" in Photo techniques Sept/Oct 2003 issue, then we can argue about whether my statements have meaning or not. If you cannot find it among your magnificent resources, I will send you a photocopy by snail mail.

[rowland_mowrey]

Patrick;

 

I did avoid flare. Does your data include or exclude flare? You are not specific on this in your posts.

 

I never said that your contributions were not of use. I said that they had been published before, as early as 1913, and were not new, and I also said that they were dangerous to use because of the large margin for error, and the fact that they were based on one point on a characterisctic curve.

 

I also added that you were an aeronautical engineer, while I was a chemist specializing in photographic chemistry and system engineering. I therefore felt that I had more depth in understanding the phenomena that you are reporting on having worked with them intensively for 32 years and having developed marketable products based on my work.

 

Because of my field of expertise I have accumulated a large library of photographic and chemistry texts and manuals. I don't believe this is a fault, nor is it wrong of me to quote them.

 

You can take all the offense you wish at my comments and you can get as upset as you like, but you cannot show that I'm wrong in the references that I cite. I am sorry that this causes you distress, nonetheless.

 

Regards.

 

Ron Mowrey

[patrick_gainer]

Ron, you can cite any reference you want, but you cannot compare it with what I did or said because you have not read what I did or said. If you insist on it, then you must follow your own wisdom and read the pertinent reference: Photo Techniques, September/October 2003: "Reciprocal Trade Disagreement". Until you do that, we have no common ground for discussion. Nothing either you or Ryuji have said has any connection with what I presented in that article.

[rowland_mowrey]

Patrick;

 

I have read all of the posts, and viewed the graph. I've seen the formula. That is all I commented on. My comments, restricted to that context, are entirely valid.

 

Ron Mowrey