Grey negs...no midtone separation!

[nishad_joshi]

had this problem with the last couple of rolls of trix i developed at 400. in

most negs the highlights and darkest shadows seem to be fine, but the midtones

are excessively flat and without separation.

 

shot on a leica mp with 35 cron (these were exposed at a stop over the

recommended meter reading - but it doesn't make sense that overexposure is the

cause of this midtone issue)

 

developed for 9:45 min in d76 (1:1) at 20C.

 

scanned with a konica minolta dimage dual scan iv.

 

examples attached. any ideas what could be causing this?

 

thanks in advance for your help!<div></div>

[nishad_joshi]

the highlights and darkest shadows seem fine, but no midtone separation.<div></div>

[jukka_lehmus]

This is maybe a dumb question to ask, but are you using fresh D-76?

[wai_leong_lee]

The highlights and shadows do not look fine to me. The print is basically "muddy".

 

Two possibilities:

 

a. Your paper is fogged slightly or expired or there is light leak in your darkroom;

 

b. The compressed midtones are in your neg, which indicates that you probably didn't intend this effect.

 

If a, then the solution is easy.

 

If b, then you need to understand what's causing it, is the development wrong or the metering wrong or the scene wrong (basically no contrast).

 

Short term, the only thing you could so is probably use grade 4 or 5.

[nishad_joshi]

I should have clarified that this is a scan from the neg.

 

@ Jukka: It's about a month old, but been in a closed bottle..

 

@ Wai: This is a neg scan. I think the development is messed up somewhere, any theories?

[ronald_moravec1]

Weak developer. Did you dilute it 1:1 and try to save it? That does not work. Diluted developer needs to used within 1/2 hour.

 

Do not stir a lot of air into it when mixing. Watch the expiration dates on the package.

 

Leave some unexposed frames on the next roll and try with fresh developer. The blank frames should be clear base or the film was fogged.

 

Test using 6 exposures on the leader of a roll. Pull out 12 inches in the dark and develope just that much until you figure out what is wrong.

 

Some bottles leak oxygen like soda bottles. Glass bottles are best as you know they are impermeable. After mixing stock, store in small one time use bottles. Dilute just before use.

[nishad_joshi]

@ Ronald: Thanks for your advice, will try developing unexposed frames. Will be more careful with diluting and mixing..

[randall ellis]

These really dont look bad for negative scans. They are a little flat, but printing with a 2 1/2 or 3 filter will fix that. You have underexposed and perhaps underdeveloped, which leads to slightly flatter negative. Often this is a good thing as it results in better highlights and gives good shadow details. If you were shooting on a bright, sunny day, you exposed exactly as you should have and lucked into developing as you should have by having a slightly weaker developer.

 

I always adjust exposure and development to get a slightly flat negative in stronger light. I wouldn't worry about it until you try to print them. I think you'll find them easier to print than a negative with more contrast.

 

- Randy

[justin - httpjustinlow.]

I have the same problem with my DSIV if I use the Minolta software. I find that Vuescan helps a little, but ultimately, the negatives scan well only if they are not too dense or too thin.

[pc_b]

Hi Nishad,

 

your scans don't look too bad. I hope you don't expect these scans to look like your grade

2 or 3 paper prints! Even on my LCD I see quite a full scale of grays, lacking zones 2 and 1

though (normal for out-of-the-box scans).

Maybe you develop too long for Trix at 200? This naturally flattens the contrast of all gray

values above 18%.

I bet, a dark yellow filter (2x to 3x!) and using trix at 400ASA will get you results you like.

 

Cheers, Peter

[colin jackson]

I found Tri-X at 400 in D76 1:1 rather flat at first until I actually started following Kodaks

directions ie. agitating with 5 180 deg. snappy inversions in 5 sec. twice per min.and

using twice as much D76 1:1 volume as required to cover the reels (also in the small

print). 2 brisk inversions per 30 sec. works well for high contrast scenes.

 

Also I use a diffuser head which needs a little more contrasty neg. Your system will

probably give good results with high contrast shots. I see no shadows in your posted

shots so the lighting is flat and more agitation would perhaps help. They don't look at all

fogged to me.

[nishad_joshi]

All: thanks a lot for your responses.. will try some of the things mentioned, but wont worry too much until i have tried printing the negs i guess..

 

@ Peter: Could you explain how the neg would get flatter with overdevelopment? (I would have thought the opposite?)

 

@ Colin: Those are useful tips, will try them out..

 

thanks!

[refinder_refinder]

hi Colin and everyone,

 

sorry if this is off the topic, but did you get surge marks when you follow Kodak's method of extending your arms and twist your wrist vigorously?

 

- refinder

[lex_jenkins]

Have you had good results before this when scanning negatives using the same materials and techniques?

 

Even if the negatives are less than optimal I suspect the results can be improved with modified scanning and digital editing techniques, or using more conventional darkroom printing methods.

[josphy]

My negatives always look flat when scanned. A little tweak in Photoshop Levels and it's cool. If you look at your histogram when you bring up levels in Photoshop, for that first example, you have a lot of space before your black point and a lot of space after your white point.

 

Your developing is fine! Don't worry about that. Just do some reading on here or google about how to adjust your contrast by setting your white and black points.<div></div>

[josphy]

By the way, Nishad, I think this is the preferable way to scan! You start with a scan that is very flat -- has all the shadow and highlight data -- then you can use Photoshop (or whatever editing software you have) to decide where your blacks, whites and midtones go. Basically it's like starting out by printing on grade 0 paper to see all of the information that is on the negative. Then you decide how hard of contrast you want to go. Like I said, even if you don't have photoshop, just do a search about how to use the "levels" tool in Photoshop, and you will get an idea of what you need to do in terms of adjusting these.

[colin jackson]

Refinder: "sorry if this is off the topic, but did you get surge marks when you follow

Kodak's method of extending your arms and twist your wrist vigorously?"

 

No, never had the surge marks. BTW, I use small Nikor tanks for 1,2 or 4 reels. The

agitation is spelled out here:

 

http://www.kodak.com/global/en/professional/support/techPubs/f4017/

f4017.jhtml#small-tankprocessing(8-or16-ouncetank)

 

"Provide initial agitation of 5 to 7 inversion cycles in 5 seconds; i.e., extend your arm and

vigorously twist your wrist 180 degrees. Then repeat this agitation procedure at 30-

second intervals for the rest of the development time."

 

Cheers.

[aeiffel]

They don't look underexposed to me.<br>

You say you exposed one stop over and you dev time is the one recommended for the "new" Tri-x @400, so you should have some contrast.<br>

My guesses, in no particular order : developer is weak, insufficient agitation, overexposure under a flat light ?

[mitchell_kirschner]

I had a similar thing happen on the very first roll I ever developed. I'm pretty sure it was under-development (HC110 over-diluted due to VERY inaccurate cheap plastic measuring cups - took me a while to track that one down).

 

Anyway, it appears you do have something to work with (which wasn't the case with my first roll). I took the liberty of applying some aggressive brightness/gamma/curves adjustments to one of your examples.<div></div>

[pete_andrews10]

It's the scanning that's the problem IMHO. Conventional photographic printing applies a strong "gamma" curve to the print, which scanning, being linear, doesn't. That's unless the programmers of the scanner software are extremely savvy about the photographic process. However, in my experience, most writers of scanner software are total numbskulls.

 

You simply need to increase the midtone separation by applying an "S" shaped gamma curve to the scan. When you've got it right, save it as an action in Photoshop, or as a setting in the scanner aquisistion software.

[pete_andrews10]

PS to the above. Using the "image adjust - curves" option in Photoshop (or any other image program that supports the function), is the best way to get your scans looking anything like a real photographic print. Contrast and brightness controls won't help much, and "levels" is too coarse an adjustment, allowing only a 3 point setting (i.e. white point, black point and midtone pivot). With "curves" you can set as many "pivots" or anchor points to the transform as needed. I usually find that two midpoint-to-highlight anchors, and a shadow anchor are needed to get an approximation to the "S" shape of a photographic printing paper curve. That's sometimes in addition to an overall gamma adjustment.

 

Levels just don't cut it, I'm afraid. And as for the brightness and contrast controls, well, the less said about those the better.

[aeiffel]

The first example is really too flat, even for a raw scan. There's no midtone separation and a lack of highlight details when there should be a full range of tones.<br>

If you've to perform strong adjustments in PS to get the picture right then there must a problem with the neg. A moderate S curve should be sufficient to make it match a print look IMHO.

[pete_andrews10]

"If you've to perform strong adjustments in PS to get the picture right then there must a problem with the neg." - Not so. A properly exposed and developed B&W neg has a maximum image density range of about 1.8D and a bar gamma of around 0.6. Getting an excellent photographic print from such a neg would be totally straightforward.

 

OTOH a truly Raw scan of the above neg would theoretically occupy only 64 greyscale levels (a quarter of the available 256), AND you would still need to apply a gamma of around 1.7 just to get an overall gamma of one.

 

Since we actually need a gamma well in excess of 1, in order for a self-luminous image to look right, then severe adjustment most certainly IS required. That's before we've even considered compensating for the non-linear nature of the film curve.

 

It's also apparent that starting off with a raw 8 bit scan will result in a lot of missing levels after adjustment, so we absolutely need to start with as many bits as the hardware of our scanner will give us.

[aeiffel]

Yes, I shouldn't have said raw, as I didn't refer to the raw sensor data. Let's say "output of the scanning software after inversion and setting of the b&w points with appropriate gamma correction applied".<br>

During this step I try to get the full range of tones by playing with exposure and the limit points of the scale. AFAIC gamma correction and inversion are performed almost automatically by the software (vuescan). So I agree this isn't true raw at all. But this is the raw file in my workflow, and I assume it's the same for most people scanning (the pictures posted are no true raw either).