Impossible to scan B/W without clipping?

CNNY - as to the color tab thing - I don't even know what I have it set on now. I've tried various settings over the years since I wrote that up and have found very little difference in the final product. If I do have Vuescan clipping data, it's set up so it only clips out 'empty' data, which I don't have a problem with. I'll check tonight - I'll fire up Vuescan once I have a scanner hooked up to my computer.

I actually don't bother with the lock exposure thing with B&W negs either.
 
I'll have to look at my settings when I get home and pull out the Coolscan V. I've not had any problems getting full-range captures with Vuescan, however.

G
 
Have you read following thread http://www.rangefinderforum.com/forums/showthread.php?t=15304&highlight=vuescan where fellow rff member schmoozit's scanning technique in Vuescan is described ? Long story but in one of the threads about Vuescan I mentioned this technique I had found and that I no longer knew the source. Anyway, for me Tri-x and Diafine worked very well, but the key is to create flat looking negatives. Are you maybe " overexposing" too much for a scanning workflow ? Problem with Diafine is that there really is not much room to manipulate density with dilution/temperature/time as you have with other developers. The only variable is how you expose the film in the camera.
 
Tim Gray said:
CNNY - as to the color tab thing - I don't even know what I have it set on now. I've tried various settings over the years since I wrote that up and have found very little difference in the final product. If I do have Vuescan clipping data, it's set up so it only clips out 'empty' data, which I don't have a problem with. I'll check tonight - I'll fire up Vuescan once I have a scanner hooked up to my computer.

I actually don't bother with the lock exposure thing with B&W negs either.
Click to expand...

There is no penalty in clipping out empty data. I tend to find having low contrast negatives is a more frequent problem, so then there is emptiness to clip out.
When you scan, Vuescan puts the data in a 16bit file, which you then manipulate once in Vuescan. The resulting 16bit file then gets manipulated once more in PS or LR. I think it makes more sense skipping one step and doing a single final manipulation to the original data.
The habit of doing manipulation in scan programs is a left-over from before Photoshop could handle 16bit data, nowadays there is no need.
Spending less time staring at Vuescan's crappy previews is also a huge time saver.
 
Agreed. That's why (I think) I have it set to Auto, but clip 0% of the pixels. But I might not - I'll check when I get home. I don't even look at the previews or ever touch the controls in Vuescan. The file comes into PS in a consistent manner where all the adjustments are done. Most of the adjustments are pretty much exactly the same file to file, same curve and everything. Obviously pictures taken in different lighting sometimes call for a different curve.
 
I scan BW with Vuescan all the time with no clipping.

I use the Nikon LS-8000, which is very similar to the 5000, just with the added ability to do medium format.

Here's how I set things:

http://chriscrawfordphoto.com/technical/scanning.php

If you still get clipping with those settings, you're film is probably overdeveloped. Scanning as a transparency rather than a BW Negative does capture a longer tonal range, so you can try that if my normal settings do not work for your films. Scan in grayscale, there is ZERO advantage to scanning a BW Neg in RGB, it just makes a file 3 times as large. I've tested this extensively.
 
problem with B+W is that the scale on the negative is so variable depending on how you have exposed and developed it. No one scanner setting will get it right for every neg and auto makes to many assumptions which might work reasonably well for colour but not for B+W which is too variable.

Go in manual mode in grayscale or transparency mode. Then after prescan goto the histogram and slide the low bar just to lower end of histogram and the highlight bar just to higher end of histogram. Set the low output value to 5 and the high output value to 250 and you are good to go. You will get everything in the scan which is likely to be a little soft but can easily be corrected afterwards in PS or whichever software you are using by adjusting contrast and/or levels. It ain't difficult.
 
Holy cow! What is going on here? My tipps:
1. Get the negative right.
2. Get the negative right.
Explanation: If you blow out your highlights in the negative, no scanning-software or scanner will help. Same counts for shadows.
3. How to get the negative right?
Long story. Basic idea: Adjust exposure and development to capture whole tonal range of the scene in the negative, depends on film, developer and what not. While scanning: Adjust blackpoint, whitepoint and contrast to liking. Mostly I try to get the negative right in such a way, that I do not have to adjust any curves during scanning. Look at the negative. Low contrast development gives more leeway for scanning, but I am not sure if this is a good strategy if one wants to wet print.

4. If it is too complicated, try digital shooting - even with a cell phone - and post-processing software, look at the histogram, maybe this is enough for your purposes. Quite similar principles apply to both, RAW files and negatives. However, with film the adjustment of the curves is implicitly done during shooting, and development -> difficult. There is still some room in the wet printing stage, but I do not do that, usually.

5. Have fun!
 
Speedfreak said:
Please post a sample.

A b/w negative potentially holds a wider dynamic range than a scanner could capture in single pass, but it may or may not matter for a specific image.
Click to expand...

sorry chum but you are talking out the top of your head. A properly exposed and developed B+W neg should have a DMax of between 1 and 1.8 at the most. A B+W neg overexposed by two stops may even have a DMax of 2. But either way, that is easily within the capability of just about every consumer grade scanner ever made.
A transparency on the other hand may approach a DMax of 4.0 and that is beyond the limits of most consumer grade scanners. That doesn't mean you can't scan transparencies. It means that some will have shadows so dense they get blocked or show a lot of noise. But B+W should be no problem as far as density range goes. Unless you're scanning ortho film but I don't think we're talking about that.
 
and to the OP. never try and scan B+W film using a colour neg setting. Colour neg has an orange mask in it (for wet printing) and the scanner tries to compensate for it which is not a good thing if it ain't there. Hence the advice to use grayscale or transparency. But always specify B+W neg as input if there is an option for it.
 
135format said:
But either way, that is easily within the capability of just about every consumer grade scanner ever made.
Click to expand...
I do not know what 135format is meaning with dmax, but I agree, my consumer grade scanner is well capable of scanning the whole tonal range of the negative. That means, I do not see much of a difference between the eyeballed neg and the scan and given that I did the development right I have quite some room in the shadows and in the highlights.
 
I struggle with highlight clipping on my Nikon too when the neg is high contrast too... I rent an Imacon every few months to scan my high-contrast negs, and anything I'm printing large. The Imacon can see the grain where the Nikon can't - albeit I'm no scanning expert.
I'm reading everything said here closely, I would love to learn to get my Nikon in tune with my negs, if in fact that is possible.
 
Most color neg or B&W neg shouldn't have high enough densities to pose much of a problem for decent consumer scanners - particularly the Coolscans. If you are getting clipping, it's probably a function of the software.

I started using Vuescan with my Coolscan because Nikonscan did always seem to clip my negatives.
 
gho said:
I do not know what 135format is meaning with dmax, but I agree, my consumer grade scanner is well capable of scanning the whole tonal range of the negative. That means, I do not see much of a difference between the eyeballed neg and the scan and given that I did the development right I have quite some room in the shadows and in the highlights.
Click to expand...

DMax is just the abbreviation for Density Maximum or Maximum Density.
Scanners have a claimed Maximum Density (Dmax) that they can scan and typically this is between 3.0 and 4.0 with 4.0 usually being a bit optimistic on the part of the manufacturer.

The values are the log( base 10) of the negative density. You require a densitometer to read neg/pos densities which are usually shown in log(base 10) values.

B+W negs typically should be between 1.0 and 1.8
Colour transparency between 3.0 and 4.0 but can be less or more.
Colour neg I don't know but less than colour transparency.

So the only thing that may cause a consumer grade scanner problems with density range are colour transparencies everything else should scan full density range no problems.
 
Been scanning B&W with Nikon scanners and a Mac for almost 15 years now and this is what I've learned. High contrast negatives are going to give you clipping. My original Coolscan III clipped B&W negs pretty badly, negatives that I could print in the darkroom with no problem, clipped with the Coolscan III.

A Nikon tech gave me a good piece of advice: Everytime I open NikonScan, I first trash the .plist Nikon Maid preference and start over. He said there is this glitch in NikonScan where after a certain number of scans of B&W the software/scanner combination starts to go very high contrast. Which produces alot of clipping.

I then acquired a Coolscan 5000, and although it was better than the Coolscan III for B&W, it still clipped quite a bit if the negatives weren't lower contrast. Still kept using the "trash preferences" thing each time before I opened NikonScan.

Then I finally got a Coolscan 9000, and the difference between it and the 5000 was quite noticeable. Like the difference between a condenser enlarger (the 5000) and a diffusion enlarger (the 9000). I still trash the preferences before launching NikonScan each time, and I have a computer that will be stuck running Snow Leopard, because as noted above, Lion won't run NikonScan and Nikon refuses to update it. But I now get glorious B&W scans from my negatives and am quite happy with the set up.

Tried Vuescan a number of different times and never saw an improvement over NikonScan.

Best,
-Tim
 
Color neg has about the same dmax as B&W - 1.8 to 2ish. Of course, that's on the straight part of the characteristic curve. If you have some real high exposure up on the shoulder and/or you've over developed your stuff, you could end up with some higher densities. But not as high as slide.
 
My experience is that in some negs I get clipping in the shadows using Vuescan.
I tried a lot of settings, exposure lock etc but couldn't make it right.
Using a scanhancer makes it worse. not using one makes the scratch removal/ dusting a nightmare on some frames (I have to improve on keeping my negs clean)
Nikon software yields better images, and since it is much simpler, I feel there are less chances of something going wrong as it sometimes happens with vuescan.
It's a shame because I love the batch work (scan from preview) of vuescan.

[edit] here: http://www.rangefinderforum.com/forums/showthread.php?p=1722788
 
You must log in or register to reply here.
Back
Top