Negative Density for Scanning?

brian steinberger · May 4, 2011

It seems to me that scanners respond better to what a wet printer would call a "thin" negative. It seems you don't need as much exposure to reveal full shadow detail when scanning. Thoughts? And at the other end, does over-development block up highlight tones quicker when scanning?

glhs116 · May 4, 2011

On the contrary, my best scans tend to come from dense negatives. I routinely overexpose. The thinner the negative the more the small amount of remaining contrast needs to be expanded. This expansion of the picture "signal" is accompanied by expansion of any sensor deffects, emulsion disuniformity, or residue on the film itself. For an extreme case try scanning a blank frame and see how bad it looks.

The denser the neg, the more information your scanner has to work with. Even the densest negs are not as dense as a typical slide and are within the range of most scanners. It is slide film that typically has a density range that exceeds some of the more "affordable" scanners.

Just by way of example, this was shot three stops overexposed:
http://www.flickr.com/photos/samagnew/5489269833/

This was shot underexposed. The shadows are almost devoid of detail:
http://www.flickr.com/photos/samagnew/3378226943/

pellicle · May 5, 2011

Hi

brian steinberger said:
It seems to me that scanners respond better to what a wet printer would call a "thin" negative.

perhaps ...

good reading on a previous incarnation of this site is here

particularly Bruce Watson's post:

What I found for my 5x4 Tri-X work was that XTOL 1:3 gave me excellent results if I developed for a Zone VIII density of around 1.0. That's about N-1 in Zone System terms. I got very similar results from 5x4 TMY-2 when I optimized it for scanning last year.

How far is too far, or how thin is too thin, is almost certainly scanner dependent. Some scanners will want more density than others, so one has to optimize for the scanner one is going to use. OTOH, if one is *ever* going to use the film in the wet darkroom, one should certainly optimize for the wet darkroom -- it will scan just fine. But if one is *never* going to use the film in the wet darkroom, then optimizing for scanning can yield a small improvement. Not a lot, but noticeable and worth doing.

its always a balance between exposure and development for density and actuance. I did a bit of testing myself on this and ended up sticking with the middle ground ;-)

happy to chat more about this if you like

L Gebhardt · May 5, 2011

I have seen many people advocating thin negatives, but I have not had as good of luck as others. My best negatives for scanning have been the ones that print well on normal to low contrast paper. In other words if it prints well on grade 0 to 2 it will scan well on my either my Epson 4870 or my ScanMate 5000 drum scanner.

Tim Gray · May 5, 2011

I'm not a master print or scanner, but many of my negatives print just fine at grades 2-2.5 and they scan just fine on my Coolscan. I can't imagine what benefit I'd get by reducing the development over this. If it prints well, it should scan well.

keithwms · May 5, 2011

It's a bad idea to deliberately aim for thin negs for scanning. You should aim to use the full density range of your film- don't shove a lot of tones into too narrow a range of density.

Most scanners now achieve DMax close to 4, which is way thicker than any neg will ever be, so you certainly don't need to worry about a neg being too opaque. (velvia slides, that's another story...) On the other end of the tone scale: DMin can be an issue, and you want for your scanned details to be far enough above the base+fog or you'll just run into lots of noise when you autolevel and wind up with highlights that aren't well differentiated (unless you spend a lot of time curving).

My thinner negs typically turn out noisy and problematic, even with a lot of multiscanning. On the other hand, rather overexposed negs easily turn out right. And perfectly exposed and developed negs... scan perfectly! Every time.

A major issue to consider is whether you should be using a pyro developer. Some pyro negs may look a bit thin to the eye but will scan well. Bear in mind that when people speak of thin negs scanning well, they are almost certainly speaking of pyro negs.

Do the following experiment: bracket some identical shots on roll film and see for yourself what works best for you. I guarantee that it won't be the thinnest nor the thickest neg.

pellicle · May 5, 2011

keithwms said:
It's a bad idea to deliberately aim for thin negs for scanning. You should aim to use the full density range of your film- don't shove a lot of tones into too narrow a range of density.

Most scanners now achieve DMax close to 4, which is way thicker than any neg will ever be, so you certainly don't need to worry about a neg being too opaque.

I agree with this 100% (my emphasis on his post). I'll add one caveat, that of the case of C-41 negative. For black and white, I always try to maximize the tonal range recoreded. This makes it harder to print from, but (as Bruce observes) will make it scan better. Recall that in print exposure one tends to use methods to narrow the range (with exposure and development) to make it more compatible with printing. To my mind one does the opposite in digital.

Framed from that context, perhaps you may find this helpful

Using digital tools with film

Alan Klein · May 12, 2011

I've started to shoot medium format Velvia 100 (slide film) and Portra 160VC and 160NC (negative film) . Landscapes. I always bracket +1 and -1. Assuming my first shot is on the money, what would your recommendation be for the scan. To scan the money shot or one of the braketed shots? And does that change depending on whether slide or negative film? Tks Alan.

L Gebhardt · May 13, 2011

Alan Klein said:
I've started to shoot medium format Velvia 100 (slide film) and Portra 160VC and 160NC (negative film) . Landscapes. I always bracket +1 and -1. Assuming my first shot is on the money, what would your recommendation be for the scan. To scan the money shot or one of the braketed shots? And does that change depending on whether slide or negative film? Tks Alan.

For slides where you have lots of dark shadows I would scan the one with the most exposure that doesn't blow out any of the highlights (pure white, or very close to it). If it's a well exposed slide where your shadows aren't blocked up and the highlights aren't over exposed it should scan very well. Shadows on a slide are the hardest thing to get clean scan of.

In most cases the negative films shot at normal or +1 exposure will scan and print well. You won't ever (I've never seen it at least) get highlights on negative film that are as dense as the shadows on slides, so your scanner should do fine. For those I would just scan the +1 exposure, and effectively that's all I would also shoot in normal conditions.

I don't know what type of camera you are using, but on my Nikons with matrix metering it almost always get a good exposure on slides, and I find bracketing +/- 1/2 to have a better chance of producing a shot that is better than the base exposure. Usually a full stop over or under isn't useable. But with an averaging meter 1 stop brackets may be better.

Tim Gray · May 13, 2011

L Gebhardt said:
You won't ever (I've never seen it at least) get highlights on negative film that are as dense as the shadows on slides, so your scanner should do fine. For those I would just scan the +1 exposure, and effectively that's all I would also shoot in normal conditions.

I have some shots that are 8-12 stops overexposed that I had some problems getting scanned properly Of course, at that point, I don't think it's the density of the film that is causing the problem but the response of the film falling apart.

keithwms · May 13, 2011

Alan Klein said:
I've started to shoot medium format Velvia 100 (slide film) and Portra 160VC and 160NC (negative film) . Landscapes. I always bracket +1 and -1. Assuming my first shot is on the money, what would your recommendation be for the scan. To scan the money shot or one of the braketed shots? And does that change depending on whether slide or negative film? Tks Alan.

First of all, I'd recommend to shoot astia rather than velvia 100, if the destination of your slide will be a scanner. That's my honest assessment. Velvia is the hardest film to scan- whether 100, 100f, or 50 or the original 50. (N.b. I am not saying that you can't get good results from scanned velvia; of course you can. But very generally speaking, and particularly if you aren't willing to put the time in to optimize it....)

If you're bracketing then good for you: take the time to experiment and scan each frame see what works best. You'll probably only have to do it once to get your answer. But it has to be your answer. I am not trying to be evasive or difficult, but we (all) probably meter a bit differently, favoring highlights or shadows in how we make that decision. Unless we all scene meter in the same way and with the same kind of meter, and we value highlight and shadow detail the same way, and we use the same scanner and software and settings, and we print the same way, we probably won't ever agree 100% on the optimal exposure for the whole workflow.

Note that if you bracket then you can also easily do combined exposures after scanning each frame, should you need to. One of the really powerful things about digital. And before anybody jumps on me, I am not advocating crappy HDR.

One mistake you don't want to make with any neg film is for it to be too thin. Expose it perfectly or overexpose a tad. But don't underexpose!!! Scanners have a very hard time extracting detail from thin portions. You have a bright light source in your scanner and there has to be a transmission appreciably less than 100% over that thin portion of your neg for the corresponding signal to rise out of the noise....

2F/2F · May 13, 2011

Also keep in mind that if you are on a tripod, you can combine two exposures in Photoshop using automatic HDR programs or stacking and layer masking. If I am shooting transparencies on a tripod in a high brightness range scene, with the intent to scan, I will do this rather than overexposing and pulling the emulsion. I particularly use this trick with the high contrast transparency films. That way you still get their unique characteristics, but can combat their narrow dynamic range by taking an extra shot for the shadows (which you can also filter differently in camera while you are at it).

Alan Klein · May 13, 2011

Thanks fo all the replies. I'll have to try scanning the different exposures to see which scans best with my procedures. Regarding combining, will the +1 and -1 bracket give me enough to work with? Which program do you recommend to combine? I have Elements 8. Do you scan flat so the scanned files can be combined better or use some other method? tks Alan.

2F/2F · May 13, 2011

Hi,

Hi, I usually take one that is exposed normally for the high and mid tones, and two for the low tones one over and two over, though sometimes an extra exposure for the high tones is useful too (in extremely contrasty shots with black low tones and white high tones). If you make it too extreme it looks very fake. I scan on a Nikon scanner for 35mm and a flatbed with holders for larger formats. It is an Epson something or other. I can find out for sure on Monday. I use HDR Pro, but it usually stinks, so I end up doing it "by hand."

gmikol · May 13, 2011

keithwms said:
First of all, I'd recommend to shoot astia rather than velvia 100, if the destination of your slide will be a scanner.

That's all fine and good...but didn't Fuji discontinue Astia, at least in 35mm?

--Greg

keithwms · May 13, 2011

gmikol said:
That's all fine and good...but didn't Fuji discontinue Astia, at least in 35mm?

--Greg

35mm film? What's that?

Sorry I had to say it. OMG I am so busted on APUG now

Alan, regarding HDR, I would shoot the exposures at least 3 or 4 stops apart for HDR, if you are using neg film. Main thing is to make sure that one framer preserves highlight details and that the other preserves shadow details. With slide that might mean combining three exposures, whereas with neg film you are unlikely to need more than two.

Anyway, when it comes to combining exposures, when using film, I don't usually care that much about the dynamic range, I mostly do compositing to make panos and such. It is an incredibly powerful technique.

The historical purpose for combining exposures, of course, was not for HDR but simply for noise reduction in astrophotography.

glhs116 · May 13, 2011

Now that the discussion has moved to slide film we are really in a whole different scanning universe. Negative film generally has these properties:
1. More than enough range in the film to capture almost any scene
2. Very tolerant of "wrong" exposure
3. Very limited density range that is within the range, generally, of even the cheapest scanner

Slide film (particularly my old friend and favourite emulsion, Velvia) has these properties
1. Not a lot of range in the "sweet spot". Blows out almost as easily as digital (though not as harshly)
2. Intolerant of wrong exposure (though there is a lot in those Velvia shadows if your scanner can dig in there)
3. Capable of very high density which often still has information. Generally somewhat beyond the range of even very good scanners

Sam

Alan Klein · May 13, 2011

Keith: I love your B/W's - the tones are great. They remind me of mother-of-pearl. How do you get them? Can you point to one of your shots where you've combined film exposures?

Sam: I just got back into film and just shot a roll of Velvia 100 and Portra VC160 but haven't had a chance to scan them. I'm using an Epson V600 flat bed. My Velvia 50's taken 15 years ago seemed to scan pretty good. I selected the bracketed shots that appeared to be the correct exposures rather than using the +1 or -1. Flickr: alanklein100's stuff tagged with velvia However, I found that the negative films had a nicer mid-range to play with after the scan especially the Fujicolor. It's going to be interesting to see what Portra does. I’ve heard good things about it.

With my negs, I also selected the correct exposure rather than +1 or –1 to scan:
Agfa Optima:Flickr: alanklein100's stuff tagged with optima
Fujicolor SHG 100: Flickr: alanklein100's stuff tagged with fujicolor
Ektar 25: Flickr: alanklein100's stuff tagged with ektar

When I scanned 35mm Ektachromes, I notice the shadows are lot more opaque and hard to open up : Flickr: alanklein100's stuff tagged with ektachrome

The medium format film really does a lot better than 35mm, at least with a flat bed. I think I’ll take the suggestion to scan all the bracket shots this time in both color corrected scan and with the settings on flat scan to see what comes up on the other side.

glhs116 · May 14, 2011

Flatbeds will be kinder to medium format because you really need more resolution than flatbeds give for small formats like 35mm.

The point I was making about negative vs. positive is that positive film is like a final product. If it looks good (right exposure) it will scan well.
Negative film was always destined for printing. The final product from negative film always had a lot to do with the way it was printed (or scanned). Ansel Adams always called the negative the "score" and the print the "performance". I was just making the point that discussions of whether to deliberately under or over expose negative film for a certain final result are valid but the same discussion is not really valid for positive film.

Looks like you are getting some very nice results from that RB67. Your Velvia scans will have more open shadows on a more capable scanner but depending on the shot the "high contrast" look can actually work really well. By contrast you wouldn't really have any trouble with the range on a negative on the V600. There all a better scanner would do is give you smoother tones. It's hard to judge these things until you try scanning some of the same frames on different scanners.

Sam

keithwms · May 14, 2011

Alan Klein said:
Keith: I love your B/W's - the tones are great. They remind me of mother-of-pearl. How do you get them?

Thanks. My methods are laughably straightforward. I normally shoot traditional-grain films like fp4+ and hp5+, develop in ID11 1+1, and print them normally or scan.

What may be perhaps a bit out of the ordinary is that I favour shallow, texturing light. For example, in a still life with artificial light, I almost always have a shallow light as well as a broad light and I simply vary those until I get what I want. I suppose that tends to favor surface texture, iridescence etc., if that's what you mean by the mother-of-pearl look.

Alan Klein said:
Can you point to one of your shots where you've combined film exposures?

It's been a long while since I tried it with film, let me see what I can dig up. Last time I did it, it was an exercise to see if I could substantially reduce grain by placing two b&w film negs in an enlarger. That was a registry nightmare! But if instead you scan them and combine them digitally, then indeed you can kill grain (and imperfections) that way. It's been a long time since I did it though.

I've done a lot more combining of digital shots than film shots. Let's see, there is this 7 foot pano print....

http://keithwilliamsphoto.net/icelandpano.jpg

This was just a series of a half dozen handheld DSLR quickies which were then stitched in photoshop. Doesn't look like much unless you see the full print.

Alan Klein · May 14, 2011

Sam: I do like the darker shadows and more high contrast look in photos. I even drag the black Levels slider into the clipped zone a little once in a while to get an even darker look in the shadows. It helps make the pictures "pop". I think trying to open shadows too much creates a flat, uninteresting view. Many people seem to get excited about bringing out the shadows by pushing the Dynamic Range. But I often don't see the point. There's usually nothing interesting there plus you often wind up with a lot of noise or grain. PLus blacker portions in your picture forces your eye to the main lighter subject where you want the viewer to be looking. On the other hand, the 35mm Ektachromes really need more DR. There are just too many large areas that just look "muddy". Interesting observation you made about resolution. I was wondering why DR is more of a problem with the 35mm.

Keith: Yes it was your still lifes that had more of that look but your other pictures were done very well too. Have you tried scanning color and converting to B/W? Anyone have any experience with NIK B/W program? The pano is pretty amazing. I don't think my wife would let me mount a 7 foot picture on the walls!!

keithwms · May 14, 2011

Thanks again Alan.

Haha yeah mounting those pano prints was a major PITA. Had to get a 4'x8' slab of 1/10" aluminum, cut it, polish, mount the prints to that, then put them on the wall with mirror mounts. Quite a project. I only did it for the very dearest of friends.

One thing I have enjoyed doing is shooting slide and then making b&w negs from the slides. For that I've used tmax100. The resulting prints have a "special" look. Of course it's a bit hard to keep the highlights, but... especially now that there are so many tools to 'rescue' detail on the ragged edge of the tone scale, sometimes people spend too much time doing that rather than just letting a print be a print.

Alan Klein · May 15, 2011

Well I scanned both Portra 160vc and Velvia 100 both with and without auto color adjustments during the scan. I used the Epson program in the V600 scanner. The final files without the auto adjustment seemed very dark so I used the auto color adjustments and fine tuned in Elements afterwards. This is what they look like. The snowy scene is the Velvia the other the Portra. What do you think?
Recent Landscape Medium Format Film - a set on Flickr
Alan

Alan Klein · May 15, 2011

I added a third (Branch in Water) that was also Velvia 100 but I scanned totally flat. It was very dark and needed a lot of level changes in Elements.

keithwms · May 15, 2011

Those look good to me Alan, the snowy one is a slightly tricky exposure and it looks like you nailed it. The highlight whites seem to be bang on.

Negative Density for Scanning?

Privacy & Transparency

Privacy & Transparency