Nikon Coolscan 8000 + NikonScan4 = Clipped Highlights BW Noise
Old 10-27-2015   #1
BLKRCAT
99% Film
 
BLKRCAT's Avatar
 
BLKRCAT is offline
Join Date: Aug 2012
Location: Toronto
Posts: 1,775
Nikon Coolscan 8000 + NikonScan4 = Clipped Highlights BW Noise

So I traded my Mamiya6 for a Nikon 8000. I've just been playing on and off this week and I see potential but I'm not wow'd yet.

I'm still trying some things but I thought I'd post here since what I'm searching on google doesn't reveal much info. So either I think theres something Im doing wrong workflow wise or something else maybe?

Overall I'm finding the scans noisy. It doesn't matter color vs bw, 120 vs 35mm the noise looks consistent. I know this scanner is capable of resolving grain so I want to do a test with some 25iso to see if what I'm seeing is in fact grain or noise but first impressions are that it's noise.

Color scans look great as far as color info and accuracy. I'm not seeing any weird shifts. Dynamic range looks good.

Black and white Im finding I always have to set a black point and my white point is always clipped. I can't pull back the highlights without going into the analog gain option, but I find pulling back the highlights with this option does weird things to tones and contrast.

I'm going to give vuescan a shot and see if I can get better control and files from that. Hoping that the issue is software related and not hardware related.

Some samples of what I'm seeing below.

This is a scan of Plus-X @ 100, F4 1/1000 from an M4, Zeiss Biogon 35mm F2. Developed in HC110B, ~5:00 (cant remember off the top of my head) 20*c Scanned at 4000dpi with multi sample 16x and Superfine option enabled. Files were scanned raw to NEF (thinking I could pull back highlights more from what I was seeing in preview) and processed in Photoshop. Results straight from the scanner with corrections in nikonscan even at 14 bit are even worse than this in some instances.



Notice the clipping in the highlights?




Also apart from the clipping, whats with the uneven size of the grain/noise in the OOF areas?


Any 8000 or 9000 users with some tips or tricks?
__________________
Tumblr Youtube
  Reply With Quote

Old 10-27-2015   #2
mfogiel
Registered User
 
mfogiel's Avatar
 
mfogiel is offline
Join Date: May 2007
Location: Monaco
Posts: 4,658
Multisampling is superfluous in case of B&W. When you prepare your scan, first make a preview and control the histogram. In Vuescan I set the B&W vendor to Kodak, film to TMAX 400 and CI to 0.4, if your histogram overflows on the right, even when the framing is accurate and does not contain blank film or worse, empty space, then try setting the input to colour negative, and if the problem does not go away, to slide. Then invert in PS.
In B&W there is no "noise" you resolve everything as is. You will see "film noise" only where the negative is underexposed or clear, like in night shots.
As to "uneven size of grain" , it is caused by lack of flatness in your negs. If you want tack sharp grain, use a glass holder. Overall, to scan with ease, do not overdevelop your negatives, this will avoid excessive highlight blocking and will shorten the scanning process.
  Reply With Quote

Old 10-27-2015   #3
Ron (Netherlands)
Registered User
 
Ron (Netherlands)'s Avatar
 
Ron (Netherlands) is offline
Join Date: Jul 2007
Location: Netherlands
Posts: 1,915
Always had great results with color. Unfortunately the 8000 and 9000 are not good @ black & white
__________________
__________________
When day is done......

Leica: IIa synch conversion, IIIb, M6 TTL Millenium, 2x Rolleicord Vb, 2x Rolleiflex 3.5F white face, Rolleiflex 2.8A, 5x ICA Tropica

My Flickr
  Reply With Quote

Old 10-27-2015   #4
BLKRCAT
99% Film
 
BLKRCAT's Avatar
 
BLKRCAT is offline
Join Date: Aug 2012
Location: Toronto
Posts: 1,775
Thanks mfogiel, I know the development of the neg is good. I've been using this recipe with my V700 and no issues.

You are totally right in saying that the histogram overflows to the right. It's almost like everything is shifted when Im setting everything from the preview.

I'm going to give vuescan a shot tonight. I will also try scanning as color and positive/invert to see what I get.

Is the level of noise/grain acceptable or consistent with what others are getting from their scanners?
__________________
Tumblr Youtube
  Reply With Quote

Old 10-27-2015   #5
mfogiel
Registered User
 
mfogiel's Avatar
 
mfogiel is offline
Join Date: May 2007
Location: Monaco
Posts: 4,658
From what I see, there are two issues:
- the grain appears so mushy, that it either has been caused by a developer like Diafine, or at least in part the film was not flat - typically you get sharp grain at 100% magnification
- there is some blockage in the highlights, although you can only see this when looking at small detail, which could mean, that the neg was a tiny bit too dense. When you increase the scanners Dmax by switching to colour neg or slide film, you will have everything inside the histogram, BUT the histogram distribution will be shorter. This way you get all the tonal extremes in at the expense of some separation in the midtones. V700 does not resolve the grain, but typically produces a more hazy and tonally acceptable scan, which you then need to sharpen up. In CS 8000/9000 you should avoid edge sharpening as much as possible, as it will alias the grain - use some local contrast enhancement instead.
  Reply With Quote

Old 10-27-2015   #6
Chriscrawfordphoto
Real Men Shoot Film.
 
Chriscrawfordphoto's Avatar
 
Chriscrawfordphoto is offline
Join Date: Apr 2007
Location: Fort Wayne, Indiana
Age: 43
Posts: 8,889
I've used A Nikon 8000Ed and both Nikon Scan and Vuescan for over a decade now for BW neg scanning. Here's the deal:

- Nikon Scan sucks for BW scanning. Use Vuescan. I have a written tutorial and a video tutorial showing how I set Vuescan with my Nikon scanner to get the results you see me post here all the time:

http://crawfordphotoschool.com/digital/scanning.php

- Grain: You're getting typical results from a film scanner. As Marek said, do not sharpen the scans, that makes the grain worse.

- Grain sharper in parts of the photo than in other parts: The Nikon 8000 requires the film to be PERFECTLY flat. The lens in it is very sharp but has NO depth of field, so if the neg is even the tiniest bit unflat, you'll see areas lose critical sharpness in the grain. Get a glass carrier; it is a requirement to get good results with this scanner, even for 35mm film.
  Reply With Quote

Old 10-27-2015   #7
BLKRCAT
99% Film
 
BLKRCAT's Avatar
 
BLKRCAT is offline
Join Date: Aug 2012
Location: Toronto
Posts: 1,775
Fair enough, thank you both. I'll report my findings tonight!
__________________
Tumblr Youtube
  Reply With Quote

Old 10-27-2015   #8
Timmyjoe
Registered User
 
Timmyjoe's Avatar
 
Timmyjoe is offline
Join Date: Jan 2009
Location: Chicago, IL
Posts: 2,918
Totally agree with Chris on this. The Nikon Coolscan 8000/9000 are great scanners for B&W work, but Nikon Scan just doesn't get the most out of the scanner.

I use Vuescan exclusively for B&W work, and remember it also has the focus point, cross hair. That tells the scanner which part of the frame you want in sharp focus. Chris is right, if you want everything in sharp focus, use one of the Newton Glass film holders, but I like using the 35mm film holder and picking the point I want most sharp, because it imitates what I do in the darkroom (don't use Newton glass film holders with my enlarger), and imitating the darkroom is my whole purpose for film scanning.

Good luck.

Best,
-Tim
__________________
http://www.timcarrollphotography.com

New Photo Books
Sturgis Stories
& Scenes From Sturgis
now available
  Reply With Quote

Old 10-27-2015   #9
Oren Grad
Registered User
 
Oren Grad is offline
Join Date: Jan 2012
Posts: 267
Quote:
Originally Posted by BLKRCAT View Post
I know this scanner is capable of resolving grain...
What you're seeing is an aliased pseudograin that's bigger and uglier than the film's actual grain. The subjective effect most of the time is to make the image more harsh and gritty than it is in the negative.

That can't be avoided. Probably something like 7000-8000 dpi would be needed to faithfully image PX/TX grain, more for finer-grained films. That's drum-scanner territory.

I have a 9000. To capture the full density range of B&W silver negatives, I've been using SilverFast and scanning as transparency, which allows me to set black and white points with maneuvering room at both ends of the histogram. The goal is to get a low-contrast, full-information scan. All other processing, including inversion to a positive, is done later in whatever image editor I'm using.

Finally, I use a glass carrier for medium format scans in my 9000, and for any 35mm scans intended for critical use.

Good luck!

Last edited by Oren Grad : 10-27-2015 at 15:44. Reason: Removed tangential material to focus more clearly on OP's questions
  Reply With Quote

Old 10-27-2015   #10
newsgrunt
Registered User
 
newsgrunt's Avatar
 
newsgrunt is offline
Join Date: Jul 2007
Location: Canada
Posts: 1,152
Interesting topic. I've been considering getting a legacy Mac just so I can use Nikon Scan lol. I find Vuescan not bad but I absolutely hate the UI. Will see if my old Silverfast will offer a cheaper upgrade path as I use curves and histograms for all my scanning and unless I'm missing it, VS doesn't have that feature.
  Reply With Quote

Old 10-27-2015   #11
BLKRCAT
99% Film
 
BLKRCAT's Avatar
 
BLKRCAT is offline
Join Date: Aug 2012
Location: Toronto
Posts: 1,775
Looks like vuescan and Chris' how to was the ticket. Being that I'm a photoshop guy and not so much a lightroom guy the workflow is very familiar.

Now it looks like Im going to have to start paying more attention to development


"Class"
__________________
Tumblr Youtube
  Reply With Quote

Old 10-27-2015   #12
Noll
Registered User
 
Noll's Avatar
 
Noll is offline
Join Date: Oct 2010
Location: Wisconsin
Posts: 493
Quote:
Originally Posted by Oren Grad View Post
What you're seeing is an aliased pseudograin that's bigger and uglier than the film's actual grain. The subjective effect most of the time is to make the image more harsh and gritty than it is in the negative.

That can't be avoided. Probably something like 7000-8000 dpi would be needed to faithfully image PX/TX grain, more for finer-grained films. That's drum-scanner territory.
+1 to this. Actually resolving grain is very difficult, and it doesn't just smooth itself away when you scan at lower resolutions. But, whether or not it actually matters depends on your needs! Glad to see you're getting better results from it.
  Reply With Quote

Old 10-27-2015   #13
Chriscrawfordphoto
Real Men Shoot Film.
 
Chriscrawfordphoto's Avatar
 
Chriscrawfordphoto is offline
Join Date: Apr 2007
Location: Fort Wayne, Indiana
Age: 43
Posts: 8,889
Quote:
Originally Posted by BLKRCAT View Post
Looks like vuescan and Chris' how to was the ticket. Being that I'm a photoshop guy and not so much a lightroom guy the workflow is very familiar.

Now it looks like Im going to have to start paying more attention to development


"Class"

I'm glad I could help. Your new example looks nice. This stuff took me a long time of practice and trial and error to learn because I got the scanner when it first came out and no one was using them or writing about them online yet. Can't believe its been so long! I got mine in 2001 if I remember right. Might have been 2002.
  Reply With Quote
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -8. The time now is 02:12.


vBulletin skin developed by: eXtremepixels
Powered by vBulletin® Version 3.6.8
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.

All content on this site is Copyright Protected and owned by its respective owner. You may link to content on this site but you may not reproduce any of it in whole or part without written consent from its owner.