Semi-scientific C41 temperature test ...

[dmr]

I decided to do a quickie sanity check to see for myself how much, if any, things like temperature and dilution affect my C41 processing, so I bought a pack of plain vanilla drugstore Kodak Gold 200 and shot three rolls (hold that thought) as identically as I easily could.

I took some shots of some test charts and some real-world shots, namely a neighbor's garden and the greenspace to the rear of the lot.

Below is a set of two test chart photos from the first two rolls. The upper two are cropped slightly and the bottom two are full-size crops from the scans. Both were done in the Jobo with pre-wet, develop, blix, wash 8 times, and stabilize. In each set, the top one was developed at 72F and the bottom one was developed at 102F and timed according to instructions. All were scanned at 3200dpi on the KM SD IV.

To me, anyway, the test shots processed at 72F look a bit "cleaner" and have less apparent grain.

At normal viewing distance on the monitor, however, it doesn't appear to be that much difference.

For the real-world shots, the change in apparent grain was not nearly as evident. The position of the camera and such had more effect on the look and feel of the images than did the temperature.

And for the third roll, I intend to process that to test the one-shot diluted solutions, but I have to wait (tomorrow or Saturday, probably) for more chemistry to come in. (You may now release that thought.)

I guess I could have diluted the stock solutions, but I want to do the test as a one-shot, rather than to re-use the solutions that already did two rolls.

Any thoughts?

 

[newcan1]

How long did you develop for at each temperature?

 

[tim_walls]

Interesting. Did the scanner do any sort of automatic adjustments to them (exposure/colour correction) or would you say these are an accurate reflection of the negs? (Hard to judge with colour negs I know... Bet interested to see the same experiment with E6 :-D.)

 

[dmr]

How long did you develop for at each temperature?

I have the sheet at home, so this may have errors, but it was something like 16 minutes at 72F and 3.5 minutes at 102F.

Interesting. Did the scanner do any sort of automatic adjustments to them (exposure/colour correction) or would you say these are an accurate reflection of the negs? (Hard to judge with colour negs I know...

What I did was set the white point using the white border and the black point at the darkest of the gradient squares. No other level/curve and no hue/saturation adjustment.

Bet interested to see the same experiment with E6 :-D.)

That may be in the future. I have, however, done one roll of cross-processed slide film (Retro Chrome) on the Jobo in C41.

 

[chassis]

Both tests with 100% freshly mixed chemicals?

 

[Photo Engineer]

I have the sheet at home, so this may have errors, but it was something like 16 minutes at 72F and 3.5 minutes at 102F.



What I did was set the white point using the white border and the black point at the darkest of the gradient squares. No other level/curve and no hue/saturation adjustment.



That may be in the future. I have, however, done one roll of cross-processed slide film (Retro Chrome) on the Jobo in C41.

Setting white point and black point can be done with a scanner, but not with color paper. What you have to do is scan with no corrections.

That is for starters.

Second, the blix is very slow to act to start with and at 75F it is quite slow. This too can distort the actual image.

PE

 

[Mr Bill]

Any thoughts?

Hi, as a note, the color wheel thing (to right of Macbeth chart in the enlarged shots) looks quite a bit different. The bottom image, the "correct" process, looks relatively horrible, including the amount of detail and graininess. (Do you have a plausible idea of why the grain difference?)

For the record, is the Macbeth chart real, or is the whole chart a printed facsimile, with a limited set of inks?

A note for your next roll of film: unless you somehow know for sure that your "standard" process was "in spec," whatever that means to you, I'd suggest having it done at a "known" lab, just for a reality check (maybe something is real "screwed up" in your process that you don't even suspect). If it were me, I'd have also included a "people" shot, but it really depends on what sort of work you personally do.

I personally wouldn't have posted this until I had resolved any possible issues - it's really hard to undo wrong perceptions once they get out there. Best of luck with continued tests.

Ps, A few notes about me and my thoughts on testing procedures: I've been involved in probably hundreds of film/paper tests with a large chain outfit. For the films, we always shot an assortment of actual humans, different complexions and hair colors (for paper tests, the previous set of negs are reused) over a wide range of exposures. These were the days of optical printing, so we hand-matched all the prints for final evaluation. We DID also shoot real Macbeth charts (they were unique for their specially made spectral characteristics), but the main evaluation was on visual reproduction of the skin tones, etc., the chart mainly gave some measurable points. When subtle color crossover occurred in the skin, the Macbeth doesn't typically reveal this. Something else we always did was to include some sort of "known" reference as a reality check - if something went haywire in the system the reference would be expected to show it. We also ran a process control strip with every test run, just to verify. Heck, we even did latent image shift tests BEFORE any of this just to know how long we had to hold exposed film prior to processing to get representative results. I don't suggest that you need to do these things - obviously our needs were different; I'm just pointing out some of the possible complications.

 

[bvy]

What you have to do is scan with no corrections.

Not only that, but scan both samples, side by side, in the same pass.

 

[Photo Engineer]

Obviously not Kodak chemistry then. The development time for C41 is 3' and 15" at 100F! Not what is given above.

PE

 

[pentaxuser]

For we trad analoguers, isn't the best and maybe the only test, prints of each neg and then reports from the tester as to whether there are differences and what they are. A scan of the prints for our viewing is ideal provided the tester can be sure that the scans replicate the prints authentically.

The problem as we have seen a lot of recently is that the tester doesn't have the facility to make prints but makes scans of a neg and assumes that because he/she thinks he/she has done the scan correctly that what appears in front of him/her must an exact replica of the negative.

A mini-lab print of the negs might be OK but it would seem that unless the commissioner of the print can be sure that lab can and will follow the instructions given to it we are still left with doubts.

A suggestion if this helps. Can the OP try to look at the negs through a 8-10 magnification loupe and report if what he and we are seeing on our screens reflect what he has reported as being factual evidence.

pentaxuser

 

[trendland]

I decided to do a quickie sanity check to see for myself how much, if any, things like temperature and dilution affect my C41 processing, so I bought a pack of plain vanilla drugstore Kodak Gold 200 and shot three rolls (hold that thought) as identically as I easily could.

I took some shots of some test charts and some real-world shots, namely a neighbor's garden and the greenspace to the rear of the lot.

Below is a set of two test chart photos from the first two rolls. The upper two are cropped slightly and the bottom two are full-size crops from the scans. Both were done in the Jobo with pre-wet, develop, blix, wash 8 times, and stabilize. In each set, the top one was developed at 72F and the bottom one was developed at 102F and timed according to instructions. All were scanned at 3200dpi on the KM SD IV.

To me, anyway, the test shots processed at 72F look a bit "cleaner" and have less apparent grain.

At normal viewing distance on the monitor, however, it doesn't appear to be that much difference.

For the real-world shots, the change in apparent grain was not nearly as evident. The position of the camera and such had more effect on the look and feel of the images than did the temperature.

And for the third roll, I intend to process that to test the one-shot diluted solutions, but I have to wait (tomorrow or Saturday, probably) for more chemistry to come in. (You may now release that thought.)

I guess I could have diluted the stock solutions, but I want to do the test as a one-shot, rather than to re-use the solutions that already did two rolls.

Any thoughts?

The scanning of films is to have it on a screen. In addition you may print a picture via digital workflow.
Thats OK .........I am no friend ofthis....?
To show a picture on Apug - there is no other way.
But to calibrate a homebrew experiment it is a real bad method - just belive me
dmr.
But you are on the right way dmr.
My respect is on your expertise - I give you just 6 month and you will become
THE expert to low temperature c41 process at the westcoast!
Your post showed quite clear : Now you
want to know what happened to films with different delutions and temperatur es.
comming -back soon........

 

[trendland]

Ok what is to do?
1) you will need a reference.
- develope one further film in standard c41 at the best comercial lab you know.
A second paralel film you should develope in your c41 brew with standard
time and delution just looking to PE. with
his last post!
THEN MAKE A REFERENCE PRINT IN DARKROOM WITH THE SAME COLOR PAPER. To each film.
Look if you can identify little differences.
In general you should use the same films,the same E.I., the same RA4 process/paper with most correctness you are able to handle.
Try to avoid heavy Filter corections while printing.Use the exact same light to your test chards.
Note you exposure settings/printing settings/filter settings!
AND THEN START WITH CHANGE OF TEMPERATURE AND DELUTION !
Look to differences and avoid much filtering - theoraticaly you should't change settings using for reference.
And you can compare results visualy.

with regards

 

[afriman]

The problem as we have seen a lot of recently is that the tester doesn't have the facility to make prints but makes scans of a neg and assumes that because he/she thinks he/she has done the scan correctly that what appears in front of him/her must an exact replica of the negative.

+1
I may well be wrong, but I am not even convinced that the greater coarseness of the second image really is evidence of increased graininess in the negative, and not a result of the scanning process. Could this not have been caused by a difference in density and/or contrast between the two negatives? How much of what we see is actual grain rather than digital noise?

 

[dmr]

Ok, I'm back in town for the better part of a week, so I'll have some time to play with this and maybe do some more tests.

I do appreciate the feedback.

In the meantime I tried a third test roll in a diluted one-shot batch of chemistry. I know this is against most advice, but I wanted to see what happened. My goal is to do onesy-twosey C41 rolls as conveniently as possible, and I would sure love to be able to do one roll at a time and not waste chemistry or worry about chemistry going bad if I did not do a second (third) roll within a week or so.

I've posted the two similar test shots from this roll at the end of this post. Meanwhile I want to respond to and discuss a few talking points.

Both tests with 100% freshly mixed chemicals?

Yes, 8oz of chemistry mixed from the bottles shortly before the 72F test roll. I used the same chemistry for both. I simply cranked up the Jobo temperature right after the first, went off and did some chores, and did the second roll.

Second, the blix is very slow to act to start with and at 75F it is quite slow. This too can distort the actual image.

Hmmmm ... I will take your word for this, deferring to your experience. I'm sure you've forgotten more about C41 than I will ever know.

The instructions are strange. My second kit is the Arista brand, my first was Cinestill, and they appear to be the same bottles, same solutions, but the instructions have some differences. The Cinestill TFM says 8 minutes in the blix anywhere from 75F to 104F. The Arista kit says 6.5 minutes in the blix at 102-104F, rising to 8 minutes at 90F and staying at 8 minutes all the way down to 72F.

The more I think about it, I do have the Jobo so I might as well use it to keep the solutions warm and scrap any idea of room temperature processing.

Hi, as a note, the color wheel thing (to right of Macbeth chart in the enlarged shots) looks quite a bit different. The bottom image, the "correct" process, looks relatively horrible, including the amount of detail and graininess. (Do you have a plausible idea of why the grain difference?)

I sure do not know. Scanning was identical. I looked again at the negatives, both appear equally dense and equally contrasty. Using a "loupe" (a spare M42 normal lens) I can't see any difference between the negatives. I actually got them confused and had to carefully check the registration of the frame numbers to be sure which was with which roll.

For the record, is the Macbeth chart real, or is the whole chart a printed facsimile, with a limited set of inks?

That's my makeshift sanity test chart. I got that from dpreview (I think) a while ago and keep it pinned to the bathroom wall to compare cameras and films. (LOL, it's turned into a conversation piece as well.) It was printed on the new Canon printer shortly after I got it.

I'd suggest having it done at a "known" lab, just for a reality check (maybe something is real "screwed up" in your process that you don't even suspect).

I do plan on repeating this with a little more attention to detail, like same exact position for each roll with the test chart. I'll probably send one of the rolls to Dwayne's.

I'm trying to think of how to do maybe 4 rolls with a couple of consistent people shots. I know I can get co-workers to pose, but four rolls will realistically mean four days, different outfits, and possibly different light (mixed office light and window light).

Obviously not Kodak chemistry then. The development time for C41 is 3' and 15" at 100F! Not what is given above.

No, Cinestill and Arista brands. I have not been able to find small volumes of the Kodak chemistry anywhere.

you are on the right way dmr.
My respect is on your expertise - I give you just 6 month and you will become
THE expert to low temperature c41 process at the westcoast!

Thanks. Of course I would have to move out west for that to truly happen.

Seriously, I'm becoming more sure and confident in doing this. It seems almost too easy. Maybe a bit of overconfidence in there? I've done quite a few rolls now and all came out very normal in appearance. No bizarre colors or anything. The grain in the one test roll is the only real issue I've had. I've even done one cross-processed roll of slide film.

I may well be wrong, but I am not even convinced that the greater coarseness of the second image really is evidence of increased graininess in the negative, and not a result of the scanning process. Could this not have been caused by a difference in density and/or contrast between the two negatives? How much of what we see is actual grain rather than digital noise?

I'm fairly confident that the scans are accurate. I did them all in a batch, same settings, did not even turn the scanner off. I batch scan 6 at a time, load six into the scan carrier, push it in, go off for a few minutes and do other things, then come back and change to another strip of six.

Anyway, here's the results of the third roll, using one-shot diluted chemistry. I diluted what would be 4 ounces of solution up to 6oz and increased all times 1.25x. I gave the blix an extra allowance, just to be sure, and because I was led to believe that it goes to completion and a modest overage will not hurt.

Using my eyes, they appear so close to the ones done at 72 F that I can't see any real difference.

The more I think about this, the more I think there may have been other factors involved with the processing of the second test roll. Maybe some chemistry cross-contamination? Maybe some timing or temp errors? Not sure.

Thanks again to all.

 

[Photo Engineer]

All of your examples have a neutral scale that goes from white to brown (reddish) to nearly black. This indicates that there is no crossover, but a serious color balance problem either in the negative or due to the scanning process itself. This assumes no effects from lighting or due to the nature of the chart.

Here is an example of the color chart done properly. Portra 160 at 160 as metered by my RZ67. Process, 100F in authentic C41 chemistry.

And, btw, I have done a series of tests from 75F to 100F with varying times and with and without water soaks to try and rebalance the 3 layers, but I always get problems such as I see in your examples. Sorry.

PE

 

[Sirius Glass]

Since you already knew that not using the correct temperature for processing C-41 or slides causes color shifts and cross over, please explain again exactly why are you looking to prove something that has already been proven.

 

[Mr Bill]

Here is an example of the color chart done properly.

Holy cow! You're a lot more casual about it than I am. When I was the QC manager at a very large chain outfit lab, if one of my people came in with that as a test (well, depending on what they intended to test for), they'd probably be going back to test again. I'd say, the chart is pretty close to the ground - look, that rock is reflecting more light onto the bottom part of the chart. Then, what else is in the foreground reflecting up? Is it significant? Why don't you turn the greyscale horizontally, so it's all equally affected by ground reflection? Etc., etc.

It really just depends on what one is trying to test for, I guess, but if you're gonna look for color crosses in that greyscale, especially with that brown gravel below, better to turn it horizontal. And figure that the green background, as light as it is, will likely put some measurable green flare into the darkest patches. Better to find a background in deep shadow, or even hang your own background. Again, it depends what you're testing for.

We did (almost) all studio work so that's the condition we tested under. We test shot all major complexions and hair colors over a wide exposure range (we routinely had studios botching up all the exposures so we want to know how forgiving the system is). Then hand-balanced all the prints within 1cc color of the "best" flesh tone from the "normal" exposure, and critically compare. (Now, by no means was that our production aim, we are mainly trying to see how wide is the path this we want to stay on.) Had we wanted to test an alternate processing condition (we wouldn't, though), we would have redone the whole shmear. Anyway, horses for courses, as they say.

 

[dmr]

All of your examples have a neutral scale that goes from white to brown (reddish) to nearly black. This indicates that there is no crossover, but a serious color balance problem either in the negative or due to the scanning process itself. This assumes no effects from lighting or due to the nature of the chart.

I kind of assumed the reddish shift in the fourth row was due to the lighting. It's mostly incandescent with some light from a curly-cue fluorescent.

Here is an example of the color chart done properly. Portra 160 at 160 as metered by my RZ67. Process, 100F in authentic C41 chemistry.

Okay, I'm convinced that I need to quit using that makeshift chart and use a real color checker chart in daylight. That's what those films are balanced for. That mixed indoor light (no windows nearby) is definitely warmish and probable very spiky with the fluorescent bulb overhead.

And, btw, I have done a series of tests from 75F to 100F with varying times and with and without water soaks to try and rebalance the 3 layers, but I always get problems such as I see in your examples. Sorry.

Don't be sorry! This is exactly the feedback I want. I see now that if I'm going to do any A-B tests I need to get rid of variables such as nonstandard light and also repeat any test that gives results that are definitely wonky, such as the excessive grain in the test above.

Since you already knew that not using the correct temperature for processing C-41 or slides causes color shifts and cross over, please explain again exactly why are you looking to prove something that has already been proven.

I wanted to see for myself just how much a difference there was between standard temperature and close to room-temperature development. Since I have the Jobo there's really no reason to do any lower-temperature stuff.

My goal for this is really to try to come up with a procedure and a workflow that will allow me to do one roll at a time, not waste chemistry, and not worry about storing once-used chemistry for up to a few weeks while I wait to get another roll shot.

I'm going for convenience, immediate gratification, and practical economy.

Here's my planned next test, which will probably have to wait until I get back in town for a couple of weeks, which will be the second week in August.

1. Test roll 1, color checker in daylight and some real-world shots that will be repeatable. Send this to Dwayne's. This will be kind of a "control" roll.
2. Test roll 2, same shots, processed in normal dilution as the first roll in a two-roll batch.
3. Test roll 3, same shots diluted chemistry, one-shot.

This should let me know how my chemistry and technique compares to a known lab, and secondly how a test of diluted one-shot chemistry compares to both the known lab, and the as-per-instructions dilution.

Thanks again everyone.

 

[Photo Engineer]

Holy cow! You're a lot more casual about it than I am. When I was the QC manager at a very large chain outfit lab, if one of my people came in with that as a test (well, depending on what they intended to test for), they'd probably be going back to test again. I'd say, the chart is pretty close to the ground - look, that rock is reflecting more light onto the bottom part of the chart. Then, what else is in the foreground reflecting up? Is it significant? Why don't you turn the greyscale horizontally, so it's all equally affected by ground reflection? Etc., etc.

It really just depends on what one is trying to test for, I guess, but if you're gonna look for color crosses in that greyscale, especially with that brown gravel below, better to turn it horizontal. And figure that the green background, as light as it is, will likely put some measurable green flare into the darkest patches. Better to find a background in deep shadow, or even hang your own background. Again, it depends what you're testing for.

We did (almost) all studio work so that's the condition we tested under. We test shot all major complexions and hair colors over a wide exposure range (we routinely had studios botching up all the exposures so we want to know how forgiving the system is). Then hand-balanced all the prints within 1cc color of the "best" flesh tone from the "normal" exposure, and critically compare. (Now, by no means was that our production aim, we are mainly trying to see how wide is the path this we want to stay on.) Had we wanted to test an alternate processing condition (we wouldn't, though), we would have redone the whole shmear. Anyway, horses for courses, as they say.

Wow Bill! What a commentary.

Well, in that photo, the grays are gray, the scale is neutral and the colors are quite accurate. The contrast is low, but that is the lighting (about 2 PM in July). I have the same scale on E6 and C41 products outdoors and in a studio, (I counted about 6 that I have posted here or elsewhere). The point is that I have compared it to my MacBeth chart and they show no crossover which was the point of that experiment and my point in this thread. If there is crossover, then the process or the film are defective. Clearly, that is not the case in my test above, but is in DMRs posts here. Something has caused her to have crossover. And despite the "faults" you find in my technique, there is no problem.

Thanks for your comments.

PE