scanning b&w -- grayscale or color?

[jano]

Hi 🙂

I'm still using nikonscan with my CS V. The last several times I have forgotten to switch the grayscale flag in the software, but, honestly, the results are often a sort of brownish tone caused by the light through the various shades of the film's base hue (some slightly gray, some slightly purple, etc).

Are there others who allow this tone through, or do you prefer to just notch the grayscale (or let the color go through and then do your own conversion to grayscale in something like PS)?

Jano

 

[skipc]

With C41 film I scan in RGB then convert to B/W in APS Channel Mixer to remove the color mask...skip

 

[DougK]

I usually scan in color unless I'm dealing with true black-and-white film, then desaturate later.

 

[jano]

Oops.. meant to say standard b&w, not c41 or desaturation of color shots.

I'm actually asking to see if anyone leaves the "toned" images from the scanner without changing to grayscale.

 

[laptoprob]

I followed Vuescan advice on scanning orange toned BW negatives as color. Appearantly true grey negatives are better of in grayscale.

No experience there yet.

Rob.

 

[peter_n]

Oops.. meant to say standard b&w, not c41 or desaturation of color shots.

I'm actually asking to see if anyone leaves the "toned" images from the scanner without changing to grayscale.

I recently saw some B&W shots that were posted in RGB and they looked quite nice. I am tempted to do this myself with a particular neg but I haven't actually done it yet.

I scan in color then desaturate in PS.

 

[smileyguy]

I always scan in colour and then use any of the myriad of methods for getting to b&w: channel mixer, straight conversion to grayscale, desaturate, etc. etc. It will provide you with more control that way.

As for actually leaving it as is once scanned in RGB that would be completely up to you--there is no right answer. If you like it as is, perfect. If you fiddle with it and like it more, great. Try a couple of different versions and do side by side comparisons and see what you end up with.

 

[skipc]

Grayscale = 8bits and should only be used for line art, text etc. Why would you want an 8bit file for a photo? RGB = 8bits "per channel", or 24 bits plus 8bits grayscale., providing more information for photographic images. I scan Tri-X, HP-5 etc in Color RGB. I never convert a photographic image to Grayscale. I convert to "Monochrome " using Channel Mixer as default, or numerous other schemes partialy suggested above for specific conversion requirements not meet by Channel Mixer. My favorite from Rusell Brown link below...skip

http://av.adobe.com/russellbrown/ColortoBW.mov

 

[hiwatt]

I always scan in RGB and then desaturate in PS. Tried greyscale but had awful results.
Sometimes I get a mild warm tone on BW negs, and, if I like it, I just desaturate less than completely. Otherwise you can get better results with actions that try to emulate platinum prints (much warmer color in darker areas) and many other...

 

[wilt]

Tried scanning in RGB, but results did not really differ from scanning as 16 bit gray. I use Tri-X and Efke KB 100 in Rodinal 1+50, Vuescan, Scan Dual IV.

 

[derevaun]

I scan in 16bit BW unless I plan to tone the image or do some special WRT color with it. 24 bit color doesn't add any more tonality to a monochrome image; it adds more colors. And, it more than doubles the size and processing time.

I know that scanning in color gives you more information to work with, but in most instances I'm going to throw away at least that much extra information. I usually don't find it worth the extra work when I compare the final images. But I do think it's worth learning how to do it.

 

[x-ray]

I have a Fuji Finescan 5000 prepress scanner and have to scan through the Fuji Colourkit software which will only scan as color. I scan 16bit transparency with all my B&W negs and have found the profiles for RDP to be more linear than any other profile partiucularly the color neg settings. I like the linear nature of this method and add my own curves for the look I want. I take it into photoshop convert to gray scale and invert. I feel I'm getting everything thats in the neg. All of my images in my gallery and website were done this way and I'm quite pleased with the results. Some of my negs are very difficult to print in the darkroom but work well scanned this way. I feel it's a superior way compared to the direct B&W scans that i did with my Imacon.

http://www.rangefinderforum.com/photopost/showgallery.php?cat=5045

www.x-rayarts.com

 

[jano]

Hi and thanks everyone for your time. I read my original post and yeah, I was not very clear in what I was looking for.. smileyguy seems to have understood, so thanks 🙂

Basically, I seem to like the "warmer" images of b&w film (standard B&W like acros, trix, hp5, etc, NOT C41 or E6!!) produced from the scanner when I leave the flag set to basic negative scan (i.e. color neg). Yeah, the photos are supposed to be done to the artist's taste, but I'm not sure if maybe I'm not seeing things the right way (untrained eye).

Examples follow. See the "warmer" image? That's a straight scan of fuji acros. And next to it is b&w is #2 -- I dunno what exactly my issue is, but I think I often prefer the warmer ones.. why? (p.s. to skipc: when I said "grayscale" I meant conversion to b&w using one of the zillions of methods, not the 8bit stuff).

Ignore the image content.. just some tests with long exposures at night 😛

 

[hiwatt]

Here's a TMX400 scanned as slide and only inverted in PS...the tonal effect can change even on the same roll of film. In this case it reminded me of the results you get from developing a paper print in exhaust lith developer...so I kept it.

 

[hth]

Skip, I do not understand, would like to learn. As I understand it RGB is for example 3 channels of 8-bit color, that is 24 -bits, where do the extra 8-bit grayscale come from?

When converted to grayscale, the color information is thrown away, which sort of leaves 8-bit in the end. (I say sort of becuase technically I would assume that it could be possible to stretch it a bit using the channel mixer, but I would assume that it is not done?)

I output to a monochrome printer, with the possibility of mixing in some toning. Sadly, I only have an old PS Elements 2.0 that cannot take 16-bit monochrome files, so I am left to do 8-bit monochrome. I have tried a little bit (by accident) using color but have not seen any difference. I use traditional B&W film.

I guess I should update to a more capable PS at some point...

/Håkan

Grayscale = 8bits and should only be used for line art, text etc. Why would you want an 8bit file for a photo? RGB = 8bits "per channel", or 24 bits plus 8bits grayscale.,