Extreme minimal agitation - results of an unintentional comparison

j_landecker · Feb 13, 2006

I thought I'd share the results of my first try with extreme minimal agitation, as I was quite surprised with the outcome. I wasn't planning to do any sort of comparison at all - I just wanted to give one of the variations of semi-stand development a try, inspired by recent threads including Sandy King's guidelines in (there was a url link here which no longer exists) thread. It turned that the exposed negative I had was a duplicate of one I had developed in the Jobo a couple of weeks ago, so I couldn't resist a comparison.

The negatives were shot on 4x5 Arista EDU 200 (= J&C Classic 200 = Forte 200) at the end of the day, with an exposure of 6 sec. @ f22. Both were developed in Pyrocat HD @20 deg C, the Jobo negative for 8 minutes at a dilution of 1:1:100, and the EMA negative for 40 minutes at a dilution of 1.5:1:200, with 1.5 mins initial agitation, and then 10 seconds at the 10-, 20- and 30-minute marks. The EMA negative is overall denser looking than the Jobo one, but the amount of shadow detail is dramatically elevated in the compared to the Jobo negative (this is a simple visual comparison on the light table).

I cropped out small portions of the scans, which were done on an Epson 4990 @ 4800 dpi, inverted the raw 16-bit scans and adjusted the contrasts of the crops to match roughly, as I wanted to compare them as if I made 2 prints of equal contrast. The crops are unsharpened. The attached overall view is of an area about 1 1/8" high on the negative.

What I found was that the difference between the two in terms of detail and sharpness is subtle given the high magnification of the crops (if your monitor is 100 dpi, these will be about 48x magnification). The biggest surprise is that the EMA negative doesn't seem to show any of the edge effects which are supposed to result from minimal agitation. Why is that? Is it the subject contrast, developer or film? Is semi-stand development (no intermittent agitation) necessary to get edge effects? I thought I might see bigger differences in "micro-contrast" as mentioned in other threads.

I've simply labelled the crops from the 2 developing methods A and B, so see if you can guess which is which...

I know this comparison is very unscientific and will be poked full of holes, but I look forward to your comments regardless.

Jim

stormbytes · Feb 13, 2006

Just a hunch..

I'm told that only "certain developers" (Rodinal topping the bunch) bring out the "edge effects" and resulting perceived accutance. I don't know that this attribute is common to all developers.

The added shadow detail seems elementary as the high-density areas exaust the developer quicker, while the low-density (shadow) areas continue developing gradually - tonal differences should also be inherent to this process.

My $0.02

bobbysandstrom · Feb 13, 2006

Two observations:

1. the initial image seems soft
2. the noise in the scan will most likely greatly diminish any degree of sharpness/edge effects when viewed at 100%

I think you'd be better of wet printing both negs and comparing those.

fhovie · Feb 13, 2006

some emlusions are more responsive to producing Mackie lines than others. I know that TRI-X is will have a more dramatic increase than FP4 - in my experience. I have no experience with J&C 200. Also - my experience with hour long times in p'cat involve a dilution of 1:1:150. With aggitations every 15 minutes, an hour soak for TRI-X produced very noticable Mackie lines which made the negatives unusable for enlarging but suitable for contact printing. (pinhole photography/like unsharp masking) My best results are with the above concentrations for 36 minutes and aggitations every 7 minutes. It adds more accutance to an enlargable negative. Good for just about anything - especially for handheld roll film where motion or imprecise focusing are the case. Other benefits, as you listed include increased film speed (1/2 to 3/4 stop) and highlight control. The downsides include - higher base fog for outdated films and sometimes uneven development (more care needed to keep aggitations subtle - more important with FP4 than with TRI-X) What is really cool about the 36min-stand/7min-aggitation/1:1:150 p'cat techinque is that many emulsions will work with the same timing ... so I have successfully developed multiple emulsions together. Densities are typically about 1.2 DR and SBR for TRI-X is 9 or so and 7 for FP4. (approximately)

Tri-x using 36 minute p'cat - 4x5 neg scan SBR=9
(there was a url link here which no longer exists)

FP4 in same soup: - 4x5 neg scan SBR=5
(there was a url link here which no longer exists)

Photo Engineer · Feb 13, 2006

Fhovie is correct, it depends a lot on the film as well as developer.

A high iodide film will give more edge effect as agitation is decreased. But, this depends on developer. A high solvent developer can sometimes offset this and sometimes enhance it.

PE

avandesande · Feb 13, 2006

From my limited testing tri-x is king of edge effects, at least in rodinal.

j_landecker · Feb 13, 2006

Thanks for the comments so far.

Bobby, the negative is sharp - the first picture is a crop, ~1" high, from the 4x5 neg. It's been downsampled and is unsharpened, so it appears quite soft on screen. My feeling is that if I can't see any edge effect on a 48x scan, it probably won't be visible on an 11x14 or even 16x20 print.... but I'll see what wet prints from the 2 negs look like.

And maybe I'll have to try some Tri-x...

sanking · Feb 13, 2006

What I am going to say is opinion so please take it that way.

Your Epson 4990 is capable of effective resolution of probably no better than about 2900 dpi, which means that the very top resolution you can hope for is about 50 lppm, and even that is probably higher than what you are getting. I would estimate that in order to actually see the result of adjacency effects in a scan you woul need resolution capability of at least 200-250 lppm, which would require something on the order of an *effective* 10,000 dpi.

Yes, in extreme conditions adjacency lines can be so large as to be visible to the human eye. But by the time they get that large the pictorial effect will have long since disappeared from the image.

Sandy

j_landecker said:
Thanks for the comments so far.

Bobby, the negative is sharp - the first picture is a crop, ~1" high, from the 4x5 neg. It's been downsampled and is unsharpened, so it appears quite soft on screen. My feeling is that if I can't see any edge effect on a 48x scan, it probably won't be visible on an 11x14 or even 16x20 print.... but I'll see what wet prints from the 2 negs look like.

And maybe I'll have to try some Tri-x...

Steve Sherman · Feb 13, 2006

Jim,

Yes, most of my work is with J&C 200 and Pyrocat, but not limited to J&C. The file of the Manhattan Bridge, which I up loaded in a previous thread, was shot with out dated Tri-X and developed in Pyrocat.

Several things occur to me which haven't been mentioned.

I will assume that edge effects occur with most diluted / reduced agitation / pyrocat combinations. They certainly do in my methods.

The subject matter looks close enough so that depth of focus at F 22 could be in jeopardy, obviously reducing any evidence of adjancecy effects. Add to that the possibility of the leaves moving during a 6 sec. exposure doesn't make this a great candidate for adjancecy effects to be evaluated.

Now the part I have no idea about but nevertheless may have an impact. The file which I scanned and up loaded with the same Epson scanner (a friend's) is 250K file as you see it here (there was a url link here which no longer exists) Each of your files are a mere 52K, this may have a dramatic effect on clarity, others would know more than I about that.

Find a scene rich in texture without distracting sunlight, (overcast light) and make several identical negatives and develop them both ways for a more accurate comparison

j_landecker · Feb 14, 2006

Sandy, your point about the scanner resolution is well taken. If these adjacency lines are of the order of 200 lppm, will their effects ever be seen after passing through the enlarger lens to make a conventional enlargement of 4-6x? I guess I'm wondering whether, besides its usefulness in compressing the tonal scale, is there really a practical sharpness and acutance advantage to stand development (for me)? I could actually see slightly more detail in the scan of the Jobo negative, although the sharpness difference could be due to other factors such as film sagging, camera movement, etc.

Jim

sanking said:
What I am going to say is opinion so please take it that way.

Your Epson 4990 is capable of effective resolution of probably no better than about 2900 dpi, which means that the very top resolution you can hope for is about 50 lppm, and even that is probably higher than what you are getting. I would estimate that in order to actually see the result of adjacency effects in a scan you woul need resolution capability of at least 200-250 lppm, which would require something on the order of an *effective* 10,000 dpi.

Yes, in extreme conditions adjacency lines can be so large as to be visible to the human eye. But by the time they get that large the pictorial effect will have long since disappeared from the image.

Sandy

gainer · Feb 14, 2006

Steve, probably you already know this but maybe some do not. The ideal message file only contains information, in the technical sense. A clear sky contains little such information, no matter how beautiful it looks, because large areas of it have the same value. An efficient system needs only count the number of ocurrences of a certain value and the position of the first one, so instead of storing the amplitude and position of each point it stores something akin to "100000, 7" instead of 100000 entries of a Zone 7 level. JPEG files can store the actual information in a 5 megapixel image in much less than 5 megabytes. The storage space required will also depend on how accurately the information must be stored. The differential amplitude sensitivity of the eye is only about 2%. The 250 K of your image file, as you may have seen by printing it, may contain as much information as a digital camera of several megapixels would have had to put into it. The camera needs to sense what is there before it can decide what is and is not information in the theoretical sense. Much the same is true of a scanner that is providing the input for the program that sorts out what is true information, converts it into a JPEG file and stores it for reconstruction.

It is interesting to see and hear demonstrations of how redundant human languages are. There is a purpose to this inefficiency. I can misspell a lot of words here and you would still know exactly what I mean.

m_liddell · Feb 14, 2006

Your Epson 4990 is capable of effective resolution of probably no better than about 2900 dpi, which means that the very top resolution you can hope for is about 50 lppm, and even that is probably higher than what you are getting. I would estimate that in order to actually see the result of adjacency effects in a scan you woul need resolution capability of at least 200-250 lppm, which would require something on the order of an *effective* 10,000 dpi.

A lot of people with experience with the Epson 4990 think the top end of the resolution is about 2400ppi and can only resolve around 23-25lpm.

noseoil · Feb 14, 2006

Aside from all this banter about scanning resolution (gentlemen, please)..... It is possible to see the effects of minimal agitation when comparing an 8x10 print from 4x5 to the same print with "normal" or continuous agitation. With the naked eye only (ok, with my reading glasses in place) the depth, texture and shadow values all look more firm with minimal agitation. Will have to try stand or semi-stand and compare it to a "normal" agitation film. Trouble is, I just can't bring myself to go back to the "normal" agitation methods. Once bitten.... tim

P.S. Would gladly give an example of same, but with my scanner and software, I'm lucky to get an image to look at, let alone one which shows any real detail.

df cardwell · Feb 14, 2006

The implied differences between the scanned images are about what I expect to see: they indicate something is going on, and indeed, experience has shown me that printing will tell the story.

Most of the time, I shoot 35.

Working with silly little 35mm negs is like doing microscopy: you know the limits of the instrument's system ( condenser, objective, etc. ) but you have to magnify the image to the point that YOUR EYE can discern what the instrument has resolved. Routinely, working beyond the limit of magnification in order to see the results. Same problem with 35 mm. You must SEPARATE the the detail to the degree it can be discerned by my Viewer's optical system !

I can easily see acutance effects at a moderate magnification through my enlarger and grain magnifier. Seeing the effect in a print is easy, as well.

But when we scan, ( I have on good authority from my somewhat evil twin who resides in The Grey Area, who communicates with me only by e-mail... and his talking sock puppet Shorty ) the scanner really can't see too well and at some point becomes confused between what it sees and what it it imagines. Until we learn to do drum scanning at home, give up on actually resolving Tri X grain developed in Rodinal: Everyman's Enlarger Alignment Negative.

In short, Jim's demonstration is profound. Well done.

.

df cardwell · Feb 14, 2006

gainer said:
...snip...

It is interesting to see and hear demonstrations of how redundant human languages are. There is a purpose to this inefficiency. I can misspell a lot of words here and you would still know exactly what I mean.

GAINER

Gadget: words have discrete meanings, just like numbers. There is really no such thing as an synonym, two words with the same meaning.

While we accept that numerals 5 and 6 are similar, they may not be used interchangeably. We cheerfully confuse words of siimilar, but different meanings like Belligerent and Aggressive. While we need analytical competence to deal with numbers, intuitive proficiency is required for words. And intuition has been banished from our skill set since technicians began running the schools.

I blame it on The Enlightenment.

.

gainer · Feb 15, 2006

Information theory deals with words, not meanings. The same word can be transmitted several times with different characters missing or too noisy to read, but the word itself may be reconstructable. There are times when the transmitted characters might represent one of several words, but the redundancy of the language allows identification of the proper word when a considerable portion of the word is missing or garbled. The babbling of an insane or drugged person could be more difficult to compress and decompress accurately than the orderly speech I hope I am sending.

The kind of redundancy that allows data compression in pictures is important in art. A truly random group of grains or pixels, as the case may be, can get by as art in some galleries, but I wouldn't pay good money for it. I appreciate a well composed picture with leading lines and all that.

RGyori · Feb 15, 2006

I've been following this thread with great interest as I have plans to experiment with semi-stand development of 4 x 5 negatives (Tri-X). One question: With an eye to minimizing the possibility of uneven development, is it an advantage to pre-soak?

Thanks!

Bob

avandesande · Feb 15, 2006

another thing to think about too is what the visual effect is...In some cases it lends a 'graphic' effect to the photo that is not always a good thing.
In a way I think that much of the percieved sharpness of digital inkjet printing has to do with unsharp masking techniques that 'fool' the eye. not necessarily a road I want to go down.

sanking · Feb 15, 2006

jdef said:
Although King's methodology seems sound, and his reputation for accurate testing well established, his results are challenged by conflicting results reported by many, including Steve Sherman.

My point is that we shouldnt be surprised when our results run counter to others published results. There is still much to be learned about these intriguing techniques, and their effectiveness with the many film/developer combinations available, and the difficult-to-quantify nature of edge effects leaves room for a lot of inaccuracy, misestimating, and passionate reporting.

Jay

I think it fair to say that the range of adjacency effects that can be obtained varies greatly, from little or none to effects that are so pronounced that the resulting image has a decided un-pictorial quality. This of of courses depends on film, developer, dilution and type of agitation. And indeed, how much is enough also depends heavily on whether one is doing contact or projection printing, and if the former, whether exposing with point source or diffuse lights, of if the latter, how great a magnification is involved. And because of all these many factors we are often speaking different languages when we discuss these issues.

However, with regard to the comparison of my results with those of Steve Sherman, pleae note that he uses a semi-stand agitation protocol whereas I use minimal and extreme minimal procedures. With the same developer and dilution one should expect to see much more pronounced adjacency effects with semi-stand than with extreme minimal, and more again with pure stand.

Sandy

noseoil · Feb 15, 2006

Steve is not the only person to disagree with your results. Posters "in this very thread have reported that they see the most dramatic effects with Tri-X. My point is that there are a lot of variables in play here, and we shouldn't take one person's results as gospel, or be surprised when our own results don't match those reported by others. Jay"

In order to be valid, the tests must be repeated independently. I see no conflict in Steve & Sandy's data. They seem to be mutually complementary tests with similar results. The vagaries with respect to film, developer dilution, agitation techniques and temperature would lead me to believe that they seem to have a pretty firm grasp of reality. I have yet to see (certain) others specify, in concrete terms, the testing of these principals. tim

sanking · Feb 15, 2006

noseoil said:
In order to be valid, the tests must be repeated independently. I see no conflict in Steve & Sandy's data. They seem to be mutually complementary tests with similar results. The vagaries with respect to film, developer dilution, agitation techniques and temperature would lead me to believe that they seem to have a pretty firm grasp of reality. I have yet to see (certain) others specify, in concrete terms, the testing of these principals. tim

Thanks, that is pretty much how I see it as well.

Moreover, go back to what I said earlier about the different languages we are speaking. It is ridiculous to suggest that my results are in conflict with, or agree with, those of another person when we are using different methods of evaluating the results. What I reported was based on a particular methodology of evaluating the results, which was by looking at the boundary areas of high and low contrast lines. I always assumed that different people in actual practice were likekly to come to different results because looking at boundary lines at high magnification is not the same thing as looking at prints.

Sandy

sanking · Feb 15, 2006

You have made a leap in logic that is not justified by what I said. I did not say that the results were not related to print sharpness, only that people making observations based on different methodologies might come to different conclucions. This is because the two methods, analysis of boundary lines with a microscope, and evaluation of print sharpness by visual inspection, are both highly subjective, even as acts unto themselves.

Sandy

jdef said:
Sandy,

if the above is true, I wonder why you chose the methodology you did. If the effects you observed by microscopic evaluation are not related to print sharpness, what practical purpose do your results serve?

Jay

gainer · Feb 15, 2006

Any one who does the kind of testing involved in judging acutance knows to expect individual differences in response when the test is done properly. If you ask the right questions and ask the questions right, you get the most unbiased results, and sometimes the most scatter. If you ask the right questions in such a way that the subject knows the answer you hope to get, you'll get it from those who like you or want you to like them, and the opposite from those who don't like you. Don't ask a photographer to judge anything about print quality.

noseoil · Feb 15, 2006

"I don't see how the above could be read as anything other than a disagreement..."

Perhaps this is the problem, a subjective statement. Other people may disagree with you, but it is not a personal attack, meerly a difference in opinion. I have proved to myself that minimal agitation is superior to previous agitation methods I have tried. If others choose to agree or disagree, that is certainly a matter of opinion. Since I am my own best (and worst) critic, this is a valid statement for me to make about my own methodology. tim

sanking · Feb 15, 2006

jdef said:
When Sandy says he sees "only very slight adjacency effects with minimal and extreme minimal agitation." with J and C 200 compared to "very pronounced adjacency effects with extreme minimal agitation." with FP4+, and Steve says, "My impressions of the greatest sharpness gain have been with subject matter which has rich texture where adjancey effects become especially noticable, this with JandC 200 film in PyrocatHD."

I don't see how the above could be read as anything other than a disagreement, and my point is not that one is right and the other wrong, but merely that these effects are extremely difficult to test and quantify, even for experts, and we shouldn't read too much into the results of one test, or be surprised when our own results don't match those reported by others.

Jay

Again, if you see this as disagreement so be it. But I do not, because both the parameters and methodology of testing are entirely different. Just note, for example, that in the passage by Steve you quote the reference is to a subject matter which has "rich texture." I have never tested JandC 200 with this type of subject so it is hard for me to understand why you insist on seeing disagreement in our different comments abou this film.

I do agree that it is very difficult quantify and provide objective measuments to the results of edge effects.

However, just for the record I want to state that I do not normally use stand and semi-stand development. For my carbon printing minimal and extreme minimal agitation give me all contrast I need, since the appearance of sharpness is determined primarily by macro contrast that produces real dimensional differences in height at the bordre of areas of high and low contrast.

Sandy

Sandy

Extreme minimal agitation - results of an unintentional comparison

Attachments

Privacy & Transparency

Privacy & Transparency