Coolscan V Newbie - 8 bit?

[dwaoka]

hi,

This friday I received CS V from my friend, I also have pro version of vuescan. I already applied locking base technique to my b/w negs and the results are fantastic! There's only one thing that I cannot change, when I scan to DNG and open it in PS, it's always 8-bit. I have 48 bits per pixel selected in Input, and 48-bit TIFF file type in output. I'm also not sure which color space should I use.

Thanks,
Peter.

 

[Bob Michaels]

I have used Vuescan for about 7 years but you are using a more complex method than I ever did. I also scan b&w but have never bothered with the locking base thing or scanning to DNG. I simply scan to a 16 bit greyscale TIF file. That has always worked well for me.

It sounds to me like maybe you are using some sophisticated color workflow for b&w. Are you using some RGB color space instead of 16 bit greyscale for b&w? I find that totally unnecessary as all three color channels will contain essentially the same information since you started with monochrome.

But, everyone has their favorite way of scanning that works for them. I am not one to contend there is a some "right way" and the others are wrong. I am just one of those that believes a simple workflow that delivers the same result is better than some complex one.

 

[dwaoka]

I will choose B/W negative and make two scans 8 bits per pixel and 16 bits, see if I can tell difference. be back in few minutes 😉

Ok, I see no difference between them, but I will stick to 16 bit grayscale.

I will be back with MORE questions 😉

 

[dng88]

I thought DNG is always 8 bit. Can you just use 16 bit Tiff Raw ...

BTW, the scan base color is to scan a section of unexposed negative and then lock base colour over that, is it? Something like this http://www.flickr.com/groups/ishootfilm/discuss/72157608204093047/

(I do have issue about setting of the 0.1 / 0.1 for curve low/high and not sure why the histogram seemed cannot be used directly to set the low/high point like other software)

Good luck and want to hear more about the experiment.

 

[dwaoka]

You have the right link to flickr,

Locking film base gives me very good results with almost no fiddling in PS.

 

[amateriat]

I scan for 16 bit greyscale. The difference isn't in what you can see on the screen, but in how well the file stands up to assorted editing tweaks in PS or the like: a 16 bit file, due to data redundancy, suffers less destruction from manipulation than an 8 bit file. Unretouched, you can't tell the difference between an 8 bit and 16 bit file. 16 bit TIFF is best for either color or b/w scans (unless you're into RAW files, but that's a whole 'nother thread...I only deal in RAW files when shooting purely digital).


- Barrett

 

[Tim Gray]

In ACR in photoshop, on the bottom of the screen will be a line with something like "Adobe RGB (1998); 8 bit; resolution; dpi". Click on it and change the settings - these are the settings Photoshop opens up your DNG after conversion with.

 

[MCTuomey]

in my experience, when looking at prints (not a monitor), 14/16 bit scans make a difference: better tone gradations, richer blacks, highlight details. i believe the 8 bit files don't respond as well to typical PS processing esp levels and curves.

 

[dwaoka]

amateriat: I scan to DNG, and then open them in PS, the DNG files go to DVD straight away after I inspect them. I really like to open them in ACR and fiddle a little bit.

Tim: I found this yesterday, but assumed wrogly it's for a single file, not for default setting, I have that changed know, thanks!

thanks guys for a helping hand... I'll post some files later on.

 

[Bob Michaels]

in my experience, when looking at prints (not a monitor), 14/16 bit scans make a difference: better tone gradations, richer blacks, highlight details. i believe the 8 bit files don't respond as well to typical PS processing esp levels and curves.

I believe you are partially right. If you do much with curves in Photoshop, a 16 bit file will hold up better.
The 16 bit precision yields higher gradations of data that minimizes gaps (honeycombed histogram) when the file is manipulated.

But, other than that, there is no difference printing a 8 bit vs. 16 bit file. Remember that all printer drivers (except QTR) convert the file to 8 bits before it sends it to the printer.

 

[dwaoka]

few quick changes in ACR (a bit of "punch" added mostly), crop, resize, no USM, quite good for 2 day experience I think 🙂