Room temperature stand development -- for C-41?!

Donald Qualls said:

I plan to run a test soon: one film of a test subject, bracketed exposure, and cut in half, one strip in standard process, the other in stand, then scan both. That takes out a lot of variables -- same camera and lens, same light, same subject, same film (same factory cassette!) -- and after establishing the scanner settings with auto levels, change to manual so I can lock them for other frames and second strip.

Click to expand...

I am so looking forward to see this! Actually I hope that the results won't be the same. The thing about color is that everything goes, different films have different biases, and if stand development changes gamma relationship between channels, for example, creating a different color profile / look. This will basically mean hey, here's another film stock you can try!

Not to start a side discussion, but the v800/850 sensor has a native resolution of 3200ppi (in the middle), and has been measured at 2600ppi (when scanning at 4800ppi with SilverFast AI) according to the filmscanner.info website.

I don't know whey they scanned at 4800 ppi when the sensor is supposedly 3200ppi and the interpolated is 6400ppi, and I don't know why it managed better resolution with silverfast than epsonscan (2300ppi).. and neither do they.

Okay, controlled experiment has begun.

I set up a little still life -- subjects chosen more for a range of color and checkable detail than artistic value; a purple padded envelope (contains a black powder paper cartridge envelope cutting template that I haven't used yet), lime green Instax Mini 9, red-on-gray tube of Invisible Gloves (which I bought for the darkroom, but on reading the label is water soluble -- apparently only good for grease/oil), a roll of Verichrome Pan 828, an old Superia X-Tra 400 cassette with film hanging out, all on my speckle finish darkroom counter (with assorted dark colored junk in the background, because I'm not a neatnik).

Set up my #2 Kiev 4 (recently tested and found to expose accurately on speeds down to 25, not tested slower) with its Jupiter-8, loaded with expired appr. 2008 Superia X-Tra 400, on my cheap/light tripod (there's a space problem for using my big heavy one in the darkroom), found I was just barely able to back up enough to focus on the test scene. Good enough. Brought in a couple clamp lamps, one native LED (which was eliminated from the test due to a flicker slow enough to potentially change exposure between frames) and one reflector type with standard screw socket, with a daylight LED bulb installed (plenty of light); the incandescent room light was left on, but contributed little to the overall exposure). Metering with my Honeywell Pentax 1/21 spotmeter on the lit side of the purple envelope, and on the light surfaces of the Instax camera, both give f/8 at 1/25. Plan was to leave the camera set at 1/25 and adjust exposure for bracketing with the aperture. At minimum focus, there's every likelihood of losing sharpness on the wider open (f/5.6 and f/4) frames, but that's unimportant for this test.

Started shooting with the metered exposure, then one stop and two stops under (f/11, f/16) and one stop and two stops over (f/5.6 and f/4) -- and just about that time the shutter starting giving trouble: the speed dial wouldn't stay set when I advanced (speed setting is the advance knob on the Kiev, same as the Contax II/III). After the third time it messed up, I unloaded the camera (which, as you might expect, promptly started working right) and grabbed my Ricoh Singlex II with Super Takumar 50mm f/1.4; advanced 8 frames with the lens cap on, then resumed testing on the 9th, now with shutter at 1/30. I was able to get two more complete cycles of 5 exposures, and the metered and +1 for a (including the first with the Kiev) 4th before the 24 exposure roll ended.

After clearing up the lights, tripod, and setup, unloading and putting away the camera, I cut off most of the exposed leader, then went dark and pulled out approximately 45 cm of film (measured as two hand spans in the dark), clipped it off, loaded it into my single-reel Paterson tank, closed that up, then loaded the remainder (within a couple cm of the same length after cutting off the unusable tail) into a single reel in my two-reel Paterson.

Later today, I'll stand develop one half-roll at room temp, replenish 11ml, and put the chemicals into a tempering bath; once they're up to temperature, I'll process the other half roll in the standard process (drift-through temperature starting at 102F), replenish another 11ml -- and when the film is dry (tomorrow) I'll scan both strips for comparison of the results. My intention, assuming scanning software cooperates, is to establish color balance and exposure on an as-metered frame in the standard process strip, then lock exposure and set color settings to "manual" for all other frames. I haven't tried this before with this scanner, so I'm not certain how much I can override the "magic" between the actual negative and what I see on screen -- I may wind up having to manually scan as positives in order to avoid uncommanded corrections and then do the inversion manually (in Darktable?).

Planned comparison points: speed (hence the bracketing -- if I find a difference I may repeat with finer exposure increments to get a better speed correction figure), grain (hard to verify with my 10 year old flatbed scanner, but I'll report what I can or can't see), color correctness, color saturation, overall contrast. Based on the one roll I've processed in room temperature stand, I don't expect to see a lot of difference, but we'll all know by Monday.

Follow-up: Both strips are processed and drying. To the eye, in direct comparison, the strip from the stand development (today at about 18C) is a little darker, both images and rebates (suggesting this is a developer difference, not incomplete bleaching -- and the previous roll showed no signs of silver retention). I should be able to scan the strips for speed, grain, and saturation/contrast comparison tomorrow.

Donald, you may want to put both strips side by side on the scanner and scan them in one go with the same settings (not on a frame by frame basis). That allows a bit of a comparison to be made.

This is all very interesting, thanks for sharing!

koraks said:

Donald, you may want to put both strips side by side on the scanner and scan them in one go with the same settings (not on a frame by frame basis). That allows a bit of a comparison to be made.

Click to expand...

Agreed.

koraks said:

Donald, you may want to put both strips side by side on the scanner and scan them in one go with the same settings (not on a frame by frame basis). That allows a bit of a comparison to be made.

Click to expand...

Anon Ymous said:

Agreed.

Click to expand...

My 35mm negative carrier has four slots, which ought to accommodate both strips (cut to five frame lengths) -- that's a better suggestion than attempting to switch to manual settings and lock exposure, as I had intended to do after establishing settings on the "metered" exposure frame. Unfortunately, it looks like one strip has almost entirely exposures from my Kiev 4, before its shutter started acting up, while the ones from my Singlex II landed on the second strip, so exposure comparison is compromised, but both cameras have seemed accurate in the past, and the difference between 1/30 and 1/25 is smaller than individual shutter errors in most cases -- at most, that's about 1/3 stop, or half if the 1/25 runs a little slow and the 1/30 runs fast.

In question is the file size I'll get scanning nearly the entire carrier area at maximum resolution (for grain comparison) -- I "only" have 16 GB RAM, and everything slows down tremendously when the scan software starts having to store directly to a temporary file. I may find I have to interleave the strips or scan only a single 5-frame section of each (my carriers accept six frame strips, but either the carrier or scanning hardware clip part of a frame on one or both ends -- who designs this stuff?).

Minor update: I was afraid of that. Attempting to scan four strips of negatives in a single image (would have been a file larger than 5 GB) resulted in running out of space for the temporary file and crashed Vuescan so hard it lost its core settings (though it recovered after a second restart plus a scanner cold restart).

I was able to successfully scan a single 5-frame strip from each half of the roll (as a completely uncorrected "generic slide"), containing the "as metered" frame and at least one over and under frame in each case -- resulting file is 2.6 GB. Now I need to attempt to crop out comparison frames to be inverted and corrected. Might take much of the afternoon; I'm not exactly an expert with any of this software.

Uou could just use the film area guide and place/tape the strips onto the glass. Then make something like a 600dpi scan for comparing color and density, which won'tbe a ridiculously large file. Next scan some individual frames at high res for your grain comparison. Btw, I wouldn't attribute too much value to a grain assessment using a flatbed scanner.

Okay, here we go. My editing sucks, I tried to make the positives look alike, but without being able to adjust them while viewing side by side, I could literally spend all day at it. For those who want to look at it themselves, here are the appr. 31 MP raw negative scans in my Dropbox. Crop 1 -- stand Crop 2 -- normal These were cropped from a single 0.9 gigapixel scanned image covering two negative strips in the carrier. I don't have a storage location with enough space to even temporarily offer the full two-strip scan; it's 5.6 gigabytes. I probably won't even keep it long on my hard disk (in favor of scanning the frames as uncorrected raw slides, which will save two or three GB, and which I can and will offer for those interested once done).

And here are the as-metered frames inverted with negadoctor in Darktable.



Stand developed (Kiev 4, Jupiter-8 50mm f/2, Superia X-Tra 400, Flexicolor chemistry)



Normal process, 100F (Ricoh Singlex II, Super Takumar 50mm f/1.4, Super X-Tra 400, Flexicolor chemistry).

Both were exposed at f/8 1/25-1/30 under a reflector lamp with a 13W household LED bulb about 65 cm from the subjects, and the room light, a ceiling fixture with 60W incandescent close to two meters away (white textured ceiling) acting as fill.

The stand developed strip has higher density, both in the frames and in the rebate; this suggests some level of fog induced by the process. Color saturation and contrast looks a little higher in stand, though that could be due to a problem with my inversion -- Darktable automatically applied the same settings, but I had to adjust the exposures individually due to the different densities of the strips.

The normally processed film looks better IMHO. Looking at the table, the stand developed one has a strange transition from yellow to magenta, whereas the normal one looks... normal. The paper on the right also looks more reasonable.

Anon Ymous said:

Looking at the table, the stand developed one has a strange transition from yellow to magenta, whereas the normal one looks... normal.

Click to expand...

Yep.
A perfect example of crossover!

@Donald Qualls thanks for doing the hard work! I agree with others, the normal development shot looks better. Also, it just occurred to me that you could have saved yourself a lot of trouble by scanning at a low DPI setting, since we only care about colors here and the resolution supported by direct uploads on photrio is very low anyway.

I too should have thanked Donald for the hard work.
If I might make a suggestion or two in case you plan to do more tests: include a flesh tone, and make sure that the light on the flesh tone varies in intensity. It is best if it transitions from a highlight to a shadow.
It is also handy to include something that will yield an image that looks like a step tablet. The page from the Kodak Colour Darkroom Dataguide that contains a grey scale and colour patches would be excellent.

MattKing said:

I too should have thanked Donald for the hard work.
If I might make a suggestion or two in case you plan to do more tests: include a flesh tone, and make sure that the light on the flesh tone varies in intensity. It is best if it transitions from a highlight to a shadow.
It is also handy to include something that will yield an image that looks like a step tablet. The page from the Kodak Colour Darkroom Dataguide that contains a grey scale and colour patches would be excellent.

Click to expand...

Ah, yes, a big thumbs up to @Donald Qualls for taking the time to provide us some hard facts. I also agree about the usefulness of a grayscale in a test shot. Something like a colour checker would be ideal, but even an improvised one would still be useful.

Вormental said:

@Donald Qualls thanks for doing the hard work! I agree with others, the normal development shot looks better. Also, it just occurred to me that you could have saved yourself a lot of trouble by scanning at a low DPI setting, since we only care about colors here and the resolution supported by direct uploads on photrio is very low anyway.

Click to expand...

That's why I made the full resolution raw negative scans of those two frames available for anyone who wants to examine them -- they're about 183 MB (in 48-bit color, that's about 30.5 megapixels) and scanned as slides with color balance set to "none", so should allow for examination of grain to the limits of my scanner (claims 4800 ppi but many have disagreed, saying native resolution is more like 1600). That also lets others judge the colors without depending on my (installed yesterday) lack of expertise in Darktable.

MattKing said:

I too should have thanked Donald for the hard work.
If I might make a suggestion or two in case you plan to do more tests: include a flesh tone, and make sure that the light on the flesh tone varies in intensity. It is best if it transitions from a highlight to a shadow.
It is also handy to include something that will yield an image that looks like a step tablet. The page from the Kodak Colour Darkroom Dataguide that contains a grey scale and colour patches would be excellent.

Click to expand...

The only flesh tones I could reasonably include in roundly twenty exposures in a setup like I have would be a photo in a book -- and I'm not certain I have any photo books with color photos in them. Nor have I ever owned a copy of the Darkroom Data Guide -- but thanks for the suggestions; I'll see if I can't come up with a stand-in for a MacBeth chart if/when I ever repeat this test. I realized I missed a good yellow, other than part of the Verichrome Pan label (and it's pretty small); if I'd thought about it, I could have included a fresh box of Tri-X in 120. Better lighting would be an improvement, as well -- this was just a matter of what I could scare up over ten minutes or so.

OK I took the two "Raw" scans, sampled the tiny bit of rebate showing to do my normal inversion using: the mask divided layer removal process, inverted, clipped the white points in each of the three channels using warnings in curves and did no other color corrections. This process normally will get me very close to a correct looking color image if the lighting temp is correct for the film. As you can see, this done to the normal processed looks like it probably matches the colors in the scene fairly closely, the same inversion process done to the stand developed negative is a hot mess. I'm sure with enough editing you can get the colors to look better, but there is clearly crossover and other problems. I had hopes this would have done better as it would make processing color negs easier, but IMHO it's not really even usable.


This is the normal processed neg.



This is the stand developed neg



If anyone is interested in trying this inversion technique in photoshop, this is a repost I cut and pasted off this forum.

"The key step is pretty simple to do in Photoshop: sample the colour of the un-inverted negative rebate, make a new layer, fill the layer with the sampled colour, set blend mode to divide, flatten the layers, invert the image, clip RGB black & white points using warnings. Then fine colour adjustments & tonal balancing. The divide blending mode is essential - the mask is not a global colour - it's a mask that's formed inversely proportional to exposure & must be removed as such. If you do so, you're well on your way to manually matching how an optical print responds.

It takes considerably more time to describe than do! Main area of trouble people tend to have is judging how far to clip the individual black/ white points in each of the RGB channels in curves. Best solution I've found is to clip the black point till the rebate has a good black, and the white until just before it starts to clip in the image area. Other important thing is that black points must be set first. Far too often, the preset driven programmes are excessively aggressive with bp/ wp settings when compared to manual controls. In comparison, the Fuji Frontier (for example) tends to clip 'white' to outright white, then adjust the output back to an L of 95 amongst a whole series of other oddities that rather stifle the range of many films. It makes sense in the context of that sort of minilab, but if you want something more akin to what an optical print might deliver, I've found manual clipping to be significantly better."

@StepheKoontz I started to type that I'm not seeing the crossover (I don't have much experience looking for that), but now I realize that the greenish glow around the center grouping isn't scatter light from the Instax, it's green highlight against the magenta (even though more open) shadows around the film cassette and Instax.

Sigh. Apparently, auto-levels scanning in Vuescan does a pretty fair job of covering this up, or it mainly occurs at certain exposures (I may have selected the +1 stop frame by mistake on the stand example) -- my initial roll had very natural looking colors, but was mostly shot in "available light" (= barely enough light) conditions. Either way, it doesn't really save any time compared to my usual water bath prewarm time and 20+ minutes actual process. Maybe as a time saver I should concentrate instead on making a thermostatically controlled warming cabinet to keep my color chemicals at 102F all the time, so I can just load the film and pour the developer.

Donald Qualls said:

The only flesh tones I could reasonably include in roundly twenty exposures in a setup like I have would be a photo in a book -- and I'm not certain I have any photo books with color photos in them.

Click to expand...

You don't have hands? .
If you were on my side of the line, I would send you a Colour darkroom dataguide or the stand alone Kodak grey scales and colour patches I have. If you post/send me your address or zip code, I can check how expensive it would be to mail them.

A lot of scanning software tries to auto correct color and can cover up problem processing. This manual method I did, using the rebate color to remove the mask, shows all It's really a hot mess unfortunately. I had hoped it would work well myself.

Good job, Donald, this is already quite valuable. And indeed, the result doesn't look very encouraging for stand development. Of course a properly daylight-lit scene might have been a better starting point; the normally developed shot doesn't look quite right to me either, and led lighting may throw things off a bit depending on the properties of the light source. But the difference between normal and stand development is in any case very pronounced. I don't hold much hope for getting it to work well, based what we see here. You might try developing shorter, but I don't think it'll work very well.

MattKing said:

You don't have hands? .

Click to expand...

I do, but they were both busy at the time I was making these exposures.

StepheKoontz said:

A lot of scanning software tries to auto correct color and can cover up problem processing. This manual method I did, using the rebate color to remove the mask, shows all It's really a hot mess unfortunately. I had hoped it would work well myself.

Click to expand...

I'm grateful for you posting the method you use -- I'll give it a try (pretty sure GIMP has the same operations available) when I rescan the individual frames to check that the crossover wasn't exacerbated by selecting the wrong exposure. I also wonder about the darker rebates on the stand developed strip; that fog level might contribute to color problems.

koraks said:

Of course a properly daylight-lit scene might have been a better starting point; the normally developed shot doesn't look quite right to me either, and led lighting may throw things off a bit depending on the properties of the light source.

Click to expand...

I chose the LED because of the weather -- the sun's been going in and out a lot lately due to partial cloud cover, and I didn't want to have to meter the scene 20 times or change anything other than aperture. I probably will repeat this test once more, starting with a camera I can depend on to get through twenty exposures without jamming up (the Singlex II isn't that great -- it was unloaded when I put it away because sometimes the shutter cocking and film advance get out of sync and the shutter will fire during advance, instead of when I press the release -- and that second Kiev 4 has a pretty bad shutter hang at speeds slower than 1/25 as well as not liking stiff advance and being easy to misload in a way that causes stiff advance). Likely I'll transfer my Super Takumar to one of my other M42 bodies, or use my Petri 7S or Minolta AL.

Donald Qualls said:

I'm grateful for you posting the method you use -- I'll give it a try (pretty sure GIMP has the same operations available) when I rescan the individual frames to check that the crossover wasn't exacerbated by selecting the wrong exposure. I also wonder about the darker rebates on the stand developed strip; that fog level might contribute to color problems.

Click to expand...

It's possible that the darker mask, as seen on the rebate, is contributing to the problem as this darker mask is across the whole frame. That said, the mask is removed evenly by the divide layer process, so there are underlying crossover issues as well. It is possible that shorter timing on the stand development might produce better results? If I was going to experiment, I would next time exposed the film in normal daylight to eliminate the artificial, mixed lighting variable. Thanks again for sharing with us!

@StepheKoontz If the darker mask is fog, however, it won't have the same imagewise distribution as the orange mask normally does -- fog induced by development is generally uniformly distributed, and the mask is not (what I've read is that the mask forms a weak positive image in the yellow and magenta layers). Worth note in this context, Vuescan uses a film profile to remove the mask, and I selected the Vuescan film profile I use for Superia X-Tra 400 (the one marked for NHG 160) on the basis of appearance (since for some reason the authors of Vuescan don't provide profiles for current films -- must remember to e-mail and ask about that, or how to save a new film profile).

I'm going to try the mask removal using the mask from the "normal" development strip on the stand frames and see if it produces a result more like what I got from Vuescan on the previous test roll. Then, in a few weeks, I should be able to attempt optical prints. I expect they'll need significantly different filtration (the fog is probably not neutral gray), but that'll be the real, final test of crossover or no.