B&W Scanning

[ianders1]

Through trial and error, I've found that the best way to scan in B&W film is to scan in 16 Bit HDR Greyscale mode. This is assuming one does not own HDR, because I imagine that would be the best way to go.

After the scan, if it's a negative, go to Image>Adjust>Invert otherwise skip to the next step. You don't have to worry about the "orange mask" with B&W negs, so this works fine. After that, go to Image>Adjust>Curves, and make your final black point, white point, contrast (S Curve), and midtone adjustments. Then you can edit the image if necessary in 16-bit mode. Check out Ian Lyon's excellent tutorial on that here:

http://www.rgbnet.co.uk/ilyons/16bit/16bit_1.htm

When finished, convert it back to 8-bits for printing, etc.

I know this isn't some great new tip, but I've been experimenting a lot lately, and this seems to yield much better B&W scans for those of us without HDR.

-Ian A.

[ilyons]

It might be worth trying with SilverFast set for 48bit Colour rather than black and white. This can help overcome some of the grain aliasing effects that some users report. Once the image appears in Photoshop convert to greyscale and carry on as Ian A describes.

Ian L

[Guest]

Color negatives are superb. I don't understand why a perfect B&W negative prescan doesn't carry through on the final scan to photoshop. I've read all the pdf manuals.I've tried every combination, every setting but can't make it come out. If this is operator error, I would appreciate the secret. Nikon LS-40 can't do ice on B&W either...

[BB]

Nikon scanners can't do ICE on B&W film because B&W film is opaque to the infrared light used by ICE to detect dust and scratches (the same applies to Ektachrome, I think).

[Bill Wood]

Nikon ICE won't work on Kodachrome due to residual silver in the image. Ektachrome works fine as does any E6 processed film.

Bill

[munchello]

I've been scanning Agfa Scala (BW positives) in 48-bit HDR color and using the channel mixer with monochrome checked as my first step in optimizing Hi-bit grayscale images. Choosing to use mostly that channel with the least grain aliasing and noise is a good first step. I then proceed with the 8-bit duping and selection-as-channel-mask technique that Ian Lyons so generously and succinctly described in his tutorial.

Of course, one could apply the channel masks and curve adjustments before using the channel mixer command, in order to have a better view of the final channel mixed output. Whatever floats yer boat.

Happy New Year!

[grischa]

Has anybody of you experience scanning B/W negs in Color mode instead of Grayscale mode?

[ianders1]

Mow that AI outputs 16-it grayscale, and they've added better B&W NegaFix profiles, I've changed, simplified, and improved my workflow.

Now I:

1. Use Negafix (with the great new Ilford profiles)

2. Adjust Exposure, then Midtones and Black and White Point.

2. Scan into 16-bit Grayscale.

3. Use PhotoShop to make final adjustments with levels, and rubber-stamp tool.

4. Save for Archive.

5. Convert to 8-bit and print.

[simoncs]

Ian,

Why do you convert to 8 bit to print.
All the Epson printers that I knoe of can print in 16 bit mode, at least in Photoshop 6.

Regards,

Simon

[ilyons]

On 2002-02-25 19:21, simoncs wrote:
Ian,

Why do you convert to 8 bit to print.
All the Epson printers that I knoe of can print in 16 bit mode, at least in Photoshop 6.

Regards,

Simon

Simon,

NO Epson driver can print using 16 bit image data. Yes you can send 16bit to the Photoshop print dialog, but once it sees the data is not 8bit and it also knows that the driver is GDI (Windows) Quickdraw Mac) it immediately converts on-the-fly to 8bit.

That said if there is no GOOD reason to convert to 8bit I to send 16bit data to the printer and let Photoshop handle the rest.

IanL

<font size=-1>[ This Message was edited by: ilyons on 2002-02-25 19:38 ]</font>

[ianders1]

Ian,

I've never even tried to send my Epson a 16-bit file from PhotoShop. I guess I just ASSumed that you had to convert it. Silly me. Thanks for the tip. Also, checked out your new review of iCorrect EditLab - great job as always.

-Ian A.

[ilyons]

On 2002-02-25 20:41, ianders1 wrote:
Ian,

Also, checked out your new review of iCorrect EditLab - great job as always.

-Ian A.

Yes it works very well. As noted in the review I don't like the saturation control - but I can't cry too loudly since it was me who got it put in

One really cool feature is the ability to create a selection Mask in Photoshop (16bit or 8 bit mode) open the filter and apply corrections to that selection only. Then apply a new selection mask and apply more edits to that in Editlab. Since the data remains in 16mode the losses are minimal.

Other than the selection mask facility users of Silverfast HDR aren't going to see much advantage in using EditLab.

IanL

[ilyons]

Ian,

read the private message section for my thoughts.


IanL

[Caleb Clapp]

Mow that AI outputs 16-it grayscale, and they've added better B&W NegaFix profiles, I've changed, simplified, and improved my workflow.

Now I:

1. Use Negafix (with the great new Ilford profiles)

2. Adjust Exposure, then Midtones and Black and White Point.

2. Scan into 16-bit Grayscale.

3. Use PhotoShop to make final adjustments with levels, and rubber-stamp tool.

4. Save for Archive.

5. Convert to 8-bit and print.

...why would one not use 48bit HDR setting in AI? Would'nt this still give the best possible RAW scan for post processing in HDR? If not, why not?

Thanks

--Caleb

[RP]

Q1:As mentioned above, is Ian Lyon's excellent tutorial :

http://www.rgbnet.co.uk/ilyons/16bit/16bit_1.htm

still visible somewhere?

Q2: Is Multiscan option possible on Microtek FilmScan 1800?

Q3: And finaly: Is better to scan BW neg in 48 bit color or in 16 bit Grayscale?


Thanks for your opinions.
Peter rus