Scanning woes - Coolscan 8000ED

[Terao]

Have spent the whole evening trying to get decent scans from Acros 100 out of my Coolscan 8000ED.

Scanning as B&W negs and I seem to have two issues:

1) the scan from the auto-exposure is lacking blacks - when I look at the histogram the start of its ramp is offset from the left by about 20

If I leave the levels as defualt (0-255) and scan I get a flat neg with muddy greys. Applying curves adjustment in Photoshop results in posterisation/noisy negs, particularly in shadows

If I adjust the curves before scanning I get less posterisation in post but its still there

2) So, I tried Vuescan. Bit fiddly to use but it generates great scans to work on in Photoshop - full range, nice open shadows, very little posterisation

Problem is of course it seems incapable of locating frames in the 120 holder. Seems to be a well-know issue from a quick Google but the solution if any is less clear. Anyone else having problems with Vuescan and have any tips for getting around them?

So, an evening of frustration - I either need to get Nikon Scan under control or find a way of controlling frame selection in Vuescan :bang:

Any help much appreciated!!

EDIT: Should add that this is the first time I've scanned B&W negs, have plenty of experience scanning colour slide/neg & even Scala and not had problems like this before

 

[Johnmcd]

I have a Minolta Scan Multi Pro and using the Minolta software, I find it best to scan as a colour positive then invert and desaturate. By doing this I believe it utilises more dynamic range of the scanner (or so I have been told). The histogram normally appears fairly narrow but adjusts well for better contrast. The resultant image is much better than scanning as a B/W neg. Maybe it would be worth trying this with your Nikon?

I could not get better scans by using the trial version of Vuescan (then again I didn't persevere too long).

The images do show a lot of grain normally but print nicely. BTW colour slide film scans much better.

All the images are scanned this way. HP5 and R3A with 40mm 1.4 Nokton

Hope this helps.

 

[Terao]

Thanks John, certainly scanning as a colour transparency is helping. PC is creaking under the strain of 500mb 12-bit TIFFs though!
Nice shots as well, not always been convinced by the Nokton as a portrait lens but those shots say otherwise.

 

[Johnmcd]

Wow! How big are you printing? When I scan some of my old 120 negs I normally match the pixel count width to approx. that of my 8 meg 20D as I am happy to print those files to A3 and the resultant files are manageable.

Though perhaps there is some fault to my logic in doing so?

I like the Nokton and normally shoot wide open. BTW, got serious RD1 GAS at present and saving according

Cheers,

John

 

[mfogiel]

I have CS 9000 - for the frames setting the offset for the first and second to -40 (minus 40) and to -60 for the third, works well. For scanning itself, scan in B&W 16 bit with the Ilford XP2 profile and see what you get. Make sure to adjust the white and black points after preview and before actual scan, and try not to include the black frames in the scan, this way your actual tonal range will distribute better in the file.

 

[Chriscrawfordphoto]

I have the 8000ED. Nikon scan is your problem. It has a well known bug when scanning negatives of any kind (color or black n white) that causes the tonal offset on the dark end of the histogram. When I switched to Viewscan that soldved my tonality problems. Nikon Scan does good quality transparency scans but Viewscan is just as good for transparencies and is a LOT faster.

I use the glass rotating carrier and have discovered the correct offsets to use for it in viewscan for 120 film. I do not know what numbers to use for the normal glassless carrier or the non-rotating glass one....experiment and you'll find them.

 

[Terao]

Is there a guide to how to set the per-frame offsets in Vuescan? Its a bit complicated! I'm generally shooting 6x7 so am only ever scanning two frames.
Scanning as transparencies in Nikon Scan is working but I'd like to minimise the workflow a little, particularly when scanning to TIFF

 

[Terao]

Wow! How big are you printing? When I scan some of my old 120 negs I normally match the pixel count width to approx. that of my 8 meg 20D as I am happy to print those files to A3 and the resultant files are manageable.

Though perhaps there is some fault to my logic in doing so?

I like the Nokton and normally shoot wide open. BTW, got serious RD1 GAS at present and saving according

Cheers,

John

I'm still getting to grips with the whole medium format workflow and also checking my technique with the Mamiya (had rangefinder alignment issue, and also skill issues ), will need to work on the file sizes. PC is about as fast as you get and its still struggling.

I've had my R-D1 for over a year now, for convenience it can't be beat but since I started developing my own B&W its sat on the shelf. I just can't get it to produce a B&W file that I like the look of. I'm sure I'll go back to it when the novelty of film wears off but I've never been a fan of digital, film is where its at for me. Overall the time spent when shooting RAW is not a great deal different to the time spent on dev & scan for 35mm.

 

[Chriscrawfordphoto]

Is there a guide to how to set the per-frame offsets in Vuescan? Its a bit complicated! I'm generally shooting 6x7 so am only ever scanning two frames.
Scanning as transparencies in Nikon Scan is working but I'd like to minimise the workflow a little, particularly when scanning to TIFF

no, unfortunately not. You have to trial and error it. what I would do is choose the frame number (one is closest to the front of the carrier, two is the next, etc) and do a prescan with no offset. see if it got the right frame, and that the whole frame shows in the prescan area. If the frame is shifted to the right, dial in some negative offset (eg: -25). If that helped but still wasnt quite enough, try again with a higher negative offset. If it was too much, try a lower number. If the frame was too far to the left in your initial scan try some positive offset (eg: 25) and just keep playing with the numbers till you get it right. then write it down! It will be the same or very close for the next time you do that frame. I suspect it will take a different offset for different frames so you may have to do this for each of the possible frames. i shoot 645, so I have 4 possible frames in a strip. can't remember how many you get from 6x7. Two maybe? That makes it easier than doing this for 4 frames! Anyway once you figure out the offset for the first and second frame, it should be the same for all subsequent scans....it is for my 645 scans using the glass carrier.

 

[Terao]

Thanks - fortunately only 2 frames for 7x6. Will experiment

 

[Johnmcd]

I've had my R-D1 for over a year now, for convenience it can't be beat but since I started developing my own B&W its sat on the shelf.

If you ever want to sell?

I can understand your love of film. With the extra time spent on film I feel there is more of 'me' in the image - and more of an investment in the art.

I just wish that my B/W scans were as good as slide film.

Cheers,

John

 

[Terao]

Yeah, I just like the idea of having the chunk of film that was exposed at that precise point in time. Bit of a hippy in this regard

For example, I have all my Dad's slides from the '60s and the actual camera/lens that took them. That's a real connection with a point in time that I just don't expect my kids will have in 40 years, even if the EXIF will tell them precisely when (and in some cases where to within a few metres) the shot was taken, and Flickr tags might tell them who the people were, and who liked it.

For someone who works with cutting-edge tech all day this is all very luddite

 

[Terao]

One week on and I'm much happier with my scanner workflow, John's tip to scan as a colour positive works really well. It also forces you to post-process which is a good thing, you get to "print" each shot rather than apply some average batch adjustment.
Only takes a few minutes in Photoshop - invert, desaturate, crop, sort out the curve and you're done.

I'm really loving being back shooting film

 

[rich815]

I understand the theory behind the whole scan as a positive and then invert method and I've seen a few articles on it, but honestly I've tried numerous times with numerous scanners through the years (Minolta Dual Scan II, Nikon LS-IV, LS-4000 and LS-9000, and Epson 1200U, 2450 and 4990) and I cannot for the life of me find any benefit vs. what I get scanning as a B&W neg.

 

[stevew]

It depends on your scanner and the type of negative density range. On my Nikon CS V there are some negs with highlight details right at the shoulder that need the positive scanning method to not burn out.

 

[rich815]

It depends on your scanner and the type of negative density range. On my Nikon CS V there are some negs with highlight details right at the shoulder that need the positive scanning method to not burn out.

Ok, I can see that. I guess I don't run into many problems because I tend to slighlty under-develop my negs for scanning (development time 10-20% less than what might be stated in the charts). Nothing more challenging for scanning than an overly contrasty or over-developed neg I've found.

 

[stevew]

Rich,

There's one shot in my gallery of an old warehouse in the LA Harbor "Steel Work" where I shot light reflecting off the building. To hold any of the lettering in the upper part of the building, scanning as positive was the only way.

 

[rich815]

Rich,

There's one shot in my gallery of an old warehouse in the LA Harbor "Steel Work" where I shot light reflecting off the building. To hold any of the lettering in the upper part of the building, scanning as positive was the only way.

I have no doubt scanning that way may have indeed helped. My point was, with all due respect, that if highlights necessitate that then perhaps the development was a little too long.

My own personal conclusion, after years of different films, developers and about 6-7 scanners and using both Vuescan and the scanner's software, that normal development times on typical development charts are targeted towards using the negative in a traditional darkroom environment and that following such can have a tendancy to produce an overly contrast or dense neg for scanning, and as a result can often blow out highlights or have it difficult to get them right in a scan. I've seen many a neg from people who have complained about difficulty in scanning and they are usually way too dense. That coupled with the fact that many people seem scared to under-develop their negs and often give a little more time "just to be sure" are making matters even worse.

 

[stevew]

I understand you point, but development for 35 mm is a compromise. 4x5 is easy to develop individual negs for optimum. 35 mm I use hc-110 dil H to help control highlights, most negs are no problem, but some images push the limits.