The right developer for scanning 35mm silver negs ?

[Bertram2]

Hi to all,

I recently read and article about scanning 35mm silver negs and that one can support a good scan result by the right developer.
Mainly that's about what grain and contrast you get and how dense the negs are.

For modern film scanners like a Minolta 5400 II with 4,8D max this does not play the role it plays for the older film scanners like Minolta Dualscan III or Nikon Coolscan III which don't have more than 3,2 D, same the flatbeds up to $300.

For scanning prints and MF negs I'll start with a flatbed soon, but maybe I can get it work for 35mm silver negs too, at least within certain limits. Would save me the CD from the lab. Their quality has fallen down to a level I don't want to accept any longer.

My Question:
Are there some experiences among the members with halfways decently working film-developer-flatbed combos ?

Thanks for all input !

bertram

 

[wlewisiii]

My best luck, by far, has come with the negs I've developed in Diafine. There is a flatness to the negative that makes scanning and photoshopping easier. In a way, when you will be scanning negs, you want ones that are like the old time newspaper shots because the scanning process looses information very similarily to the halftone process.

William

 

[kiev4a]

You don't want heavy negs for scanning. They should be as thin as possible without losing detail. My best scans from negs that probably look thin to some folks. Of course it tried to keep them pretty thin even in the darkroom days. I don't think the developer is as important as how you expose AND developing time.

 

[jlw]

I'd concur with the above posts that the key to good scanning is to avoid excessive highlight density. Compared to an enlarger, a scanner is much more limited in how much light it can punch through the areas of highest density (since scanners are primarily designed for scanning chromogenic films, which form their image out of dyes that aren't nearly as opaque as a silver-based emulsion.)

A scanner with a wider density range helps resolve more tones out of its AVAILABLE range, but the high and low limits of that range aren't necessarily any wider than a scanner with a lesser density-range spec.

So, you want to pick a developer, developing time, and metering index that will give you enough shadow density for the scanner to pick up details, while avoiding exceeding its highlight-density limit. In theory, that would argue for a developer with proportional action, such as Diafine has -- its two-part operation naturally limits how much the highlights can develop vs. the rest of the image.

However, I've found that I can get good results with a wide range of developers, simply by making sure I don't over-develop.

 

[kaiyen]

There was a good thread about developing for grain vs. acutance on PN a while ago. It can be found at http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=00D9XM

I agree that one should go for thinner negs. Get just enough exposure to get the shadow detail you want (I find that I can get a 1/3 stop speed increase as compared to the traditional ZS densitometer-based methods), and don't overdevelop. At the same time, I personally try not to way underdevelop, either. Dealing with excessively flat negatives makes it harder to work with the midtones, in my opinion. I try to get a nice, full histogram.

As for developers, I prefer to go for a bit more acutance than not, but try to avoid the really active ones like FX-1 or FX-2 (I haven't actually tried them, admittedly, but the comments about how grainy they are are dissuading me from trying). I would rather get the sharpness to begin with rather than going for low grain and trying to sharpen afterwards.

I'm in the process of puttign up sample images and 100% crops of grain results for my various combinations. Hopefully I'll have that URL soon.

allan

 

[kiev4a]

You are right. You have to walk a fine line because if you underdevelop too much things get pretty flat. Overall, however, I think I prefer that to overdevelopment because the highlights block up to the point where all you get is white-white.Everyone's thermometer is probably different but D-76 1:1 at 68 F on new Tri-X seems to work pretty well for me.

 

[vincentbenoit]

There was a good thread about developing for grain vs. acutance on PN a while ago. It can be found at http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=00D9XM

Thanks for the link.

I would rather get the sharpness to begin with rather than going for low grain and trying to sharpen afterwards.

Same here.

I'm in the process of puttign up sample images and 100% crops of grain results for my various combinations. Hopefully I'll have that URL soon.

I'd be VERY interested in this kind of first-hand data. What kind of film do you typically use?

Cheers
Vincent

 

[GeneW]

I certainly agree with what's already been said. Two things make a B&W negative tricky to scan: too much contrast and too much density.

I've seen recommendations to use the finest-grain developer possible on higher-speed films like Tri-X and HP5 because scanners exaggerate grain more than enlargers do.

I personally don't follow this advice. I use whatever I like -- anything from D-76 to Rodinal -- but I try to keep the density and contrast tamed.

Gene

 

[kaiyen]

I'd be VERY interested in this kind of first-hand data. What kind of film do you typically use?

It'll just be examples of what I'm using, provided I can find good, properly exposed and developed samples, of course. My films include:

Delta 100
FP4+
TXT
Delta 3200

I also have a bit of:
Efke 25, 50, 100
Fortepan 400
HP5+

These are in my usual devs:
Rodinal
D76
Microphen
Perceptol
FX-39 (only recently, and only for testing).

I think that's it. My goal is to put up a page with not just my own personal development info but also links to images and 100% crops for grain purposes. Of course, my results will have to be taken only for what they are worth - my results from my process, which very well might be flawed.

I'm going to start building it tonight. I hope to be done by next week with as much as I can.

allan

 

[kaiyen]

Overall, however, I think I prefer that to overdevelopment because the highlights block up to the point where all you get is white-white.Everyone's thermometer is probably different but D-76 1:1 at 68 F on new Tri-X seems to work pretty well for me.

Absolutely no question about that- definitely don't overdevelop. When I'm first trying something out, or if I was lazy and didn't measure the contrast range, I err on the side of caution and underdevelop.

My point was that I'd rather get closer to the correct development than underdevelop by too much. I scanned some Delta 100 in FX-39 this weekend where the histogram was waaaaay to the left. I overdid it a bit (or underdid it, I guess).

allan

 

[vincentbenoit]

My goal is to put up a page with not just my own personal development info but also links to images and 100% crops for grain purposes. Of course, my results will have to be taken only for what they are worth - my results from my process, which very well might be flawed.

Sure, but it'll certainly still be useful - to me, at least.

I'm going to start building it tonight. I hope to be done by next week with as much as I can.

Great! I look forward to it.

Cheers
Vincent

 

[Bertram2]

Gents,
thanks so far for all contributions, which were really helpful ! 🙂
Tho blown out highlights are really a serious fault which make a pic worthless for me those closed black shadow portions don't earn much more tolerance IMHO.
And I agree that, as some pointed out, it is no use to flatten the negs by the finest grain possible thus getting a boring mud of greys.

I don't expect any wonders by varying the developer, the basic 35mm product must simply have a certain quality to get scanned halfways decently even by a flatbed scanner.

Contrasty stuff works only good at film scanners, that's a fact. I considered to stay at C41 B&W as long as I do it all with a flatbed but my experience with C41 is that too dense negs push a cheap scanner over it's limits in the same way silver based negs do. The lack of grain may be helpful tho.

What nobody could tell me up 'til today is if scanning a (contrasty) print does not achieve a better result than scanning a neg when it comes to scanning 35mm with a flatbed.

I've seen scanned 6X6 prints which really blew me away but I haven't seen any examples of 35mm prints . Has somebody tried this already ?

Thanks again,

Bertram
Who now has to learn new all the dev stuff he knew already in the late 70s ! 🙂 Used HP4 and Microphen in those days AFAIR

 

[Tin]

Question : where does the fine-line lie when developing negatives both for scanning and for enlarger printing? I have an enlarger with a diffuser light source, which means that the normal procedure is to develop negatives to a higher contrast as compared to developing negatives for use with a condenser enlarger. Does that constitute overdevelopment in the context of film scanning? Should I follow the "old" procedure, or should I develop the negatives to a contrast level mid-way between diffuser and condenser printing?

Tin

 

[T_om]

Bertram,

You are imposing some serious limitations upon yourself by trying to do this with a flatbed. Is a good 35mm film scanner completely out of the budget? Some may be had quite reasonably now. The difference in scan quality is astounding.

Also, a less contrasty negative does not have to look like "mud" if your digital workflow is on track and the negative is properly processed after the scan. You will NEVER get the best quality image going directly from scan to print... it is just not possible in the digital environment.

Tom

 

[jlw]

Question : where does the fine-line lie when developing negatives both for scanning and for enlarger printing? I have an enlarger with a diffuser light source, which means that the normal procedure is to develop negatives to a higher contrast as compared to developing negatives for use with a condenser enlarger. Does that constitute overdevelopment in the context of film scanning? Should I follow the "old" procedure, or should I develop the negatives to a contrast level mid-way between diffuser and condenser printing?

Tin

I also have a diffusion light source on my enlarger, and I think you can develop negatives that will work well with both an enlarger and a scanner. It's just traditional good photography practice: expose enough that you get recordable shadow detail, and don't develop any more than necessary to get clean whites. You might have to go up half a grade on your normal printing paper, but beyond that you shouldn't need to do anything special.

 

[Bertram2]

Bertram,

You are imposing some serious limitations upon yourself by trying to do this with a flatbed.
Tom

A temporary solution only, as described.

 

[kiev4a]

No question a good dedicated film scanner has a leg up on a flatbed. But there are flatbeds available today that will produce results that aren't far off dedicated quality unless you are blowing something up to 16 x 20. And they are much better IMHO than the low end dedicated film scanners in the same general price range == plus they will handle larger film and reflective scans.

 

[derevaun]

Scanner software is important too--the Epson plug-in that works with my scanner won't let it scan in 16 bits, whereas VueScan does.

I respect but don't heed the advice to stick with one film, because at this point the surprises are more fun than the satisfaction of having what I expect to happen actually happen a lot. But I find that a compensating developer like Diafine or Rodinal with reduced agitation makes negatives that don't produce banding in the highlights. It takes some work with Curves to get a high contrast look (I can't get as much control with Levels).

 

[kaiyen]

Hi all,
So the beginnings of my film/dev page is up at this link.

It's taking me a lot longer than I expected.

Some things I have learned:
1 - while my prints look razor sharp, perhaps my scanner just isn't getting as much detail as I had thought. Many of the 100% crops don't show the grain that I was expecting, even if I used smart sharpening.

2 - however, at least a few of the rodinal examples have nice, sharp grain. I wonder if the solvent effect of ID-11, for instance, is a lot stronger than I thought.

3 - however^2, the grain with fp4 and rodinal looks good, but with d100 and fx-39 (supposedly higher acutance than rodinal) is more scanner noise than grain. I wonder if it takes a traditional emulsion with an acutance developer to really have tight, sharp grain.

allan

 

[Roman]

What nobody could tell me up 'til today is if scanning a (contrasty) print does not achieve a better result than scanning a neg when it comes to scanning 35mm with a flatbed.

Well, personally, I do get HUGELY SUPERIOR (if I knew how to write that in extra-bold type, I'd do it) results from scans of prints, compared to those from 35mm negs (and even 6x6cm negs), both done on the same 3200 dpi flat-bed scanner. Better resolution of fine detail, better shadow detail, better highlight definition. And let me state that again - the difference is GIGANTIC, my neg scans look like total crap, while (at least in my opinion) my scanned prints look at least accepetable, some people might even find them good... 😉
But then again, the reason might be that I consider myself quite good with darkroom work (hey, I've got a couple of years experience there), while my PhotoShop abilities are rather minimal...

Roman

PS: One exception: it is easier to get acceptable on-screen results with really thin, underexposed negs from scanning, even those that are absolutely unusable in a real darkroom often give at least slightly viewable results.