Thin or dense negative better to scan?

[holggger]

Hello everyone,

something has been bugging my mind for months now in regards to dense negatives and scanning.
Generally I tend to the dense negative side when it comes to shooting black and white (and also C41, actually) film, believing that the film contains more tonal differentiation especially in the shadows and is easier to print in the darkroom, and considering that film is definitely more forgiving to overexposure than underexposure.

When it comes to scanning these dense negatives, what I was wondering is the following: I am sure scanners have some kind of dynamic range in which they can capture tones. It would seem to me that if the negative is very dense you have all the tones on the dark end of the scanner's dynamic range – which makes me wonder if the scanner would be able to capture all these tonal differences, especially considering that it can theoretically capture more tonal differences on the highlight end of its dynamic range due to the nature of digital sensors (keyword "exposing to the right" – see https://en.wikipedia.org/wiki/Exposing_to_the_right for technical background)

As an example, when I scan negatives with Hasselblad's/Imacon's Flextight scanner, I get a histogram in the preview scan of the negative that shows the scanner's dynamic range and the tones of the negative fall somewhere inbetween. For very thin negatives the histogram is shifted towards the right/bright side, for very dense negatives it is shifted to the left/dark side (as it should be), without ever clipping or touching the borders (the scanner's dynamic range is larger than the tonal range that can be on the negative). Considering the nature of digital sensors, it would seem like the scanner would catch most tonal information if the histogram of the negative is spread out over as wide the dynamic range of the scanner as possible, or if it peaks towards the bright side of the histogram. A very dense negative's tones are spread out over only a very small part of the scanner's dynamic range/histogram, which on top of that is the dark part that contains less tonal separation - which seems like a bad idea.

There's probably some flaw in my way of thinking and it is driving me crazy that I can't wrap my head around it! – maybe you know where I am wrong? Is a dense negative bad for scanning?

P.S: Bonus question to those who know more about the workings of the Flextight/Imacon scanner in particular (could not find an answer in their manual): Does it make a qualitative difference if I scan a negative using the whole dynamic range the scanner provides me (setting white and black point to 0 and 255 during scanning, respectively) and then setting the white and black point I actually want for my image in post-processing, compared to setting that white/black point BEFORE scanning in the scanner software?
It seems to me that all that setting the white and black point before scanning does is, technically, making the scanner scan at full dynamic range anyway and then clipping/remapping the tones according to my settings in post - so basically the same as above.

 

[Frank53]

I’m do not know very much about the technical background, but I’m using an older (precision2) Imacon scanner for a few years now.
My idea is that exposing to the right is as much true for scanning as it is for dig. photograpy.
So too dense a negative is not a good idea.
Then imo it is very important to set white and black points before every scan. If you do that you always get the best possible scan from any negative.
Regards,
Frank

 

[Sirius Glass]

It is easier to avoid thin or dense negatives by avoiding the problem than to fix in the darkroom or with scanning.

First of all to avoid thin or dense negatives, shoot box speed. Second stop taking light reading that include the sky. That will solve most of your problems. The latitude of print film is more than wide enough to capture the shadow detail. The overexposing is a thing of the past and is rarely needed for the shadows. Shoot box speed and use burning, dodging and bleach.

Thin negatives can be the result of shooting a light subject on a dark background or a dark subject on a light background. There are two solutions: take a reflective reading of only the subject or take an incident reading.

Printing thin negatives can be frustrating and more difficult than dense negatives.

 

[Bob Carnie]

Thin negatives always come up BETTER on scanner in my workflow than dense negatives

 

[Prof_Pixel]

Image silver scatters light so that dense negatives won't scan as well as correctly exposed negatives (not a problem with color dye images).

 

[Alan Edward Klein]

I bracket B/W negatives +1 and -1 from my meter reading using box speeds. When I scan with my Epson V600 flat bed, I scan flat and don't apply any pre-scan settings. I find that the scans that are more to the right bring out more shadow details. So I would say that less dense is better. Why don't you try bracketing and trying the same with your scanner and see the results? Then let us know what you find. With color slides (Velvia 50) I also bracket +1 and -1 and select the slide that looks natural and correct to my eye, not that one that is more or less dense. Colors and contrast shift if you're not at the correct exposure setting, something you're not concerned with when shooting B/W negatives.

I happen to agree with you that setting levels (black and white points) before the scan produces no difference than scanning flat and doing the settings in post. At least I haven'y seen any difference although some people claim there's more pixels if you set before the scan. Frankly, I think the scan programs alway scan flat and apply settings afterward to the resultant file much like a camera converts RAW data to a jpeg photo.

 

[jim10219]

I think a lot will depend on your scanner, scanning software, and scanning techniques. For me, a properly exposed negative is easiest to scan. After that, I find denser negatives harder to scan, but when properly scanned, produce better images than thinner negatives. However, I'm able to set separate points for the R, G, and B sensors, which helps in establishing a wide dynamic range. Since figuring out how to do that, and how to use the multi-exposure option, scanning isn't as hard as it used to be. Usually, if I can wet print it with decent results, I can scan it with better.

 

[shutterfinger]

A negative that prints well at paper black in standard wet print process is easiest to scan.

 

[removed account4]

Thin negatives always come up BETTER on scanner in my workflow than dense negatives

+1 !
thinner negatives are a dream to work wtih.

i have 20+ years worth negatives on the thin side
10+years of stuff in the middle and
10+ years of negatives so dense you can't see through them
(contact)printing the super dense negatives on paper is a blast
and while scanning and printing inbetweenies isn't hard if you know what you are doing
thin ones are so much easier to scan ..

 

[Ted Baker]

There's probably some flaw in my way of thinking and it is driving me crazy that I can't wrap my head around it! –

Yep, remember a scan is an exposure as well, what your talking about with exposing to the right applies to the scan, not you negative.

 

[Mainecoonmaniac]

How about doing multiple scans? One for the dense highlights one for the middle tones and one for the shadows. You then can merge the image into an HDR image. You can tone map it to your liking.

 

[Ko.Fe.]

I scanned all kids of negatives with three different scanners and four different applications. If scanner and software is good it works fine with different negatives. Some negatives where so thin I can't see what on them, but if scanner and software is good, it will read it. Most of the dense negatives would come out OK as well.
But if negative is really dark this is where problems comes most. But doesn't it make sense? Scanner sends light to read....

 

[Alan Edward Klein]

How about doing multiple scans? One for the dense highlights one for the middle tones and one for the shadows. You then can merge the image into an HDR image. You can tone map it to your liking.

Multiple scans will not matter since the scanner's single pass will not clip at either end. Remember, the film is limited to a few stops. That's unlike a digital camera trying to capture a let's say 15 stop range in a landscape shot with bright sky and dark shadows where setting different exposures and tone mapping might work. You only need to capture 5-7 stops on film. That range has already been condensed on the film from the 15 stops of natural light it was exposed too. What limits the results is the scanner's dMax which is it's ability to scan through the darker areas to get details That limitation is designed into the scanner and is set to maximum by the manufacturer. Even of you could tweak it, you could only reduce the light output or amplification of the sensors which will distort the results or give less details. The limits are set by the scanner hardware as designed and built. You can;t get blood from a turnip.

 

[Patrick Robert James]

Thinner negs get you a better signal to noise ratio which makes the files look cleaner. In the darkroom though, thick is better.

 

[Ted Baker]

I am not sure about the imacon, but you may find it is already has automatic exposure control, that does a good job of solving the problem. Even the epson scanners have exposure control, and can do multiexposure if necessary. Keep in mind you may find that with even with your densest negatives, that they are still comfortably within your scanners capabilities, such that you are not able to discern any difference. Other than the effects of having a denser negative will have in terms of light scattering, grain etc.

 

[mdarnton]

I think there are a lot of unconsidered variables in this problem. For instance, MY thin negs scan nicely because they are underdeveloped but appropriately exposed for that, not because they are underexposed and lacking shadow detail which would be impossible to regain in scanning. It's not because of the range of the scanner, which easily covers a B&W negative and much more. To fill the histogram, you'd have to not expose more (which would just slide your insufficient-contrast film left and right on the histogram) but develop more, for greater contrast. This has its own problems, including additional grain from additional developing, which would basically make the results worse. However, my natural-contrasty xray film photos scan just fine, because they're 8x10 film and grain isn't an issue, no matter what

 

[Mainecoonmaniac]

You can;t get blood from a turnip.

Thanks Alan for the info. So my guess is to buy the scanner with the best dynamic range one can afford.

 

[Chan Tran]

I have no problem with dense negative when scanning. I don't like thin negative.

 

[NortheastPhotographic]

I can tell you that when we scan on our Frontier, very dense negatives scan oftentimes better than a negative that would print well in an RA-4 darkroom. Most of our clients over expose Portra films by at least 1 stop, and Fuji 400H by at least 2 stops. Each scanner is somewhat different however, so as usual it's best to test, test, and re-test. This of course does not generally apply to B&W or chromes, just C41 based films.

 

[holggger]

I am not sure about the imacon, but you may find it is already has automatic exposure control, that does a good job of solving the problem. Even the epson scanners have exposure control, and can do multiexposure if necessary. Keep in mind you may find that with even with your densest negatives, that they are still comfortably within your scanners capabilities, such that you are not able to discern any difference. Other than the effects of having a denser negative will have in terms of light scattering, grain etc.

This is exactly what I was wondering about in my PS question and couldn't find an answer to. If the scanner controls exposure by, say, setting a larger aperture or longer scan time in those instances when the tonal range I want to scan is on the dark side, then it'd make sense to set white and black point before scanning, I guess. If aperture and scanning time are fixed – which I have the impression it is on the Imacon – it wouldn't make much sense technically to set the points before. Except maybe if the conversion from the RAW image captured by the scanner to the TIF file leads to better results than me setting white and black point on the already converted TIF file.

 

[holggger]

Thank you all for your answers – seems like many share my impression that thin negatives are easier to scan (lead to better results).
Maybe dense negative can only be scanned "better" on scanners that technically support exposure adjustments (on a hardware level, by change of aperture, speed, ...)?

 

[Alan Edward Klein]

Thanks Alan for the info. So my guess is to buy the scanner with the best dynamic range one can afford.

DMax is more important than range. The higher the number, the better the scanner can pull out details from denser portions. The dMin at the other end is usually captured by all scanners. Film only has 5-7 stops.

 

[Ted Baker]

If aperture and scanning time are fixed – which I have the impression it is on the Imacon

I doubt that the exposure time is fixed, I suspect it is automatic (which may well be just as good for single pass). However it is wrong to assume that setting the white point or black point has anything to do with the exposure time.

very dense negatives scan oftentimes better than a negative that would print well in an RA-4 darkroom.

Exactly and there are a few variables that are either being ignored or misunderstood. It's possible your imacon is not up to the job, or perhaps it IS, I would be wary of making a sweeping statement that it is not.

 

[Chan Tran]

DMax is more important than range. The higher the number, the better the scanner can pull out details from denser portions. The dMin at the other end is usually captured by all scanners. Film only has 5-7 stops.

If I remember correctly most color negative only get to about 2.0 density and I think some scanner can handle 4.0 density

 

[holggger]

I doubt that the exposure time is fixed, I suspect it is automatic (which may well be just as good for single pass). However it is wrong to assume that setting the white point or black point has anything to do with the exposure time.

Exactly and there are a few variables that are either being ignored or misunderstood. It's possible your imacon is not up to the job, or perhaps it IS, I would be wary of making a sweeping statement that it is not.

True, it is wrong to assume that setting black and white point before scanning has anything to do with exposure time – but it would make sense in a scenario in which setting these points before scanning leads to better results (debatable).

What variables that are being ignored or misunderstood are you referring to?