PS "Techniques" for Chromogenic film scans - Ideas wanted

C

ch1

Guest
Lately I've been shooting some chromogenic film, both KodaK TCN 400 and Ilford XP-2.

After scanning with my Nikon Coolscan 5000D in NEF (RAW) mode I bring the pics into PS (CS2).

Upon initial viewing they seem to have a slight reddish "tint" to them.

Here's what I then do:

1) Auto Contrast

2) Desaturate

3) Minor manual adjustments to contrast/highlights

4) Sharpen total image (optional - not everytime).

5) For posting to RFF Gallery resize to 800dpi on longest side and save in JPEG Fine

That's about it.

A couple of questions:

A) Does the order of #1 and #2 matter? I sometimes "reverse" these steps.

B) While I prefer minimal PS processing - are there any other general steps you would recommend?

Note: Obviously if there is a nasty scratch etc. I try to "fix" that too.
 
What I'd do is to avoid auto contrast entirely.

First, instead of desaturating, considering using the channel mixer. You don't _nec_ have to do more than just click the "monochrome" button, but it's nice to have the added flexibility of picking one channel over the other, or even mixing them in different amounts.

I recommend using a curve to get your contrast right. Often just a simple S curve is sufficient.

allan
 
I don’t think the order of one and two matters very much. Another method to bring out contrast is to set a black point, white point, and grey point on a levels layer. Open a new levels adjustment layer. Click on the black eyedropper icon in the adjustment window, then click on the part of the image that you want to set as black. Do the same with the middle eyedropper (grey) and the rightmost one (white). You can tweak the amounts with the slider bars for more accuracy.
 
Almost invariably the first thing I do is get rid of all the dust, spots, etc. Then proceed with the rest. Curves, channel mixer, etc.

Drew
 
i scan as a colour positive so the first thing i do is invert.

then levels with a bit of contrast/brightness fiddling sometimes, but not auto contrast.

i clean the neg/file if need be and then a 'touch' of unsharp.

i need to learn curves so i don't use it yet.
i have played with it but i just make things worse so far.

i also want to learn how to burn & dodge but so far i haven't a clue.

so much program so little brain...:)
 
Copake_ I almost always use 400CN, and scan on a Nikon 9000 scanner using Nikonscan 4.02 software. I set digital ICE to Normal, (but not Fine!). This takes care of dustbusting for the most part. I scan as 16 bit grayscale. I see very little point in scanning in RGB mode if you're going to end up with a B&W image.

Ed
 
back alley said:
i scan as a colour positive so the first thing i do is invert.

then levels with a bit of contrast/brightness fiddling sometimes, but not auto contrast.

i clean the neg/file if need be and then a 'touch' of unsharp.

i need to learn curves so i don't use it yet.
i have played with it but i just make things worse so far.

i also want to learn how to burn & dodge but so far i haven't a clue.

so much program so little brain...:)
Click to expand...

Hey Joe,

I'm "baby steps" with PS too.

I tried the "curves" tonight and realize its just the histogram approach to doing the manual highlights/contrast with the "draw bars". Guess it's one's preference for numbers or graphs. Either is cool.

But I will wean myself away from the "Autocontrast"

Which scanner do you use? I have the option of scanning color pos or monochrome and have been doing the latter. Maybe I should try the CP since the film is "chromogenic".

Damn - I gotta take a course in PS.

Oh, as "kaiilen"(?) said up above - the channels thing is probably the way to go - but I forgot how to turn on the TV! :eek:

P.S. doubt you could come back to NYC now - you're spelling "color" the wrong way! :D
 
i scan all my b&w film as colour positve, not just the c41 stuff.
i read it somewhere on the net, it offers more information then when scanning as monochrome.
my scanner is an old canon (of course) fs 2710.

joe
 
I think the "more info" may be mostly empty "info packets"-that's what bulks up the file size. I don't think you're getting more detail out of the film, though.
 
aad said:
I think the "more info" may be mostly empty "info packets"-that's what bulks up the file size. I don't think you're getting more detail out of the film, though.
Click to expand...

You see, here's an interesting counterpoint.

If a film image is still at least equal, and preferably better, than an equivelent digital image (i.e . same ISO/shutter/apeture - shot in RAW) then a digital scan in RAW should provide equivlelent data.

Maybe I need more info on how scanner "pixelates"?
 
Instead of hitting grayscale, I convert the profile to sGray, which gives me grayscale but drastically reduces the file size (of tiffs). I believe sGray gives the gray palette of sRGB (which is what you use if you put stuff up on the web). Anyway sGray does just fine for me.
 
Instead the use of auto contrast, I like more the results of Auto Levels.

This tool removes almost all color tint. If you want for really grey pictures.

I like the Konica Monochrome VX400 film, it give good pictures.

The following pictures are made:
Auto Levels
Grayscale mode with flattening
RGB mode
and sharppenig.
 
I scan b&w as 16 bit greyscale in Vuescan - the images are 1/3rd the size of 48 bit RGB. I use Vuescan with the black and white points set to 0.01 to 'stretch' the number of bits used in the actual image.

I never use any of the 'auto' adjustments.

I'll then use an adjustment layer to errr adjust either the levels or curves. Output sharpening is via the Photokit sharpeners. I may also use Neat Image if there's more noise on the image than I want.

One other thing, I always keep the original untouched scan. I save any images I alter in another folder called rather grandly 'Masters'.

Steve
 
"kaiilen"? That's a new one.

Scanning in RGB does provide more info - you will actually pick up slightly differing amounts of detail in the red channel vs the green vs the blue. By getting all of that data then converting via channel mixer, you can get the most out of the negative. After you go to monochrome, then you can go greyscale, going down to 1 channel and cutting your file size by 2/3.

I choose to scan in RGB also because Photokit Sharpener requires that, as does my black & white printing workflow.

Also you should scan in 16 bit if you can handle the file sizes. You will get more information that way with which to work as you manipulate the file.

allan
 
back alley said:
i scan as a colour positive so the first thing i do is invert.

then levels with a bit of contrast/brightness fiddling sometimes, but not auto contrast.

i clean the neg/file if need be and then a 'touch' of unsharp.

i need to learn curves so i don't use it yet.
i have played with it but i just make things worse so far.

i also want to learn how to burn & dodge but so far i haven't a clue.

so much program so little brain...:)
Click to expand...
The best method for dodging and burning and the least destructive is the gray layer method.
Make a new layer(PC shft>Ctrl>N)
on the popup set layer style to overlay
check the fill with gray box and hit ok
now get a big soft brush
make sure your Fgrnd/Bkgrnd color is B&W(PC hit D)
Lower the brush opacity to about 20% to begin with.
Use black to burn and white to dodge.
You may have to make multiple pasases or play with the brush opacity
When done just flatten
If you don't like the results just delete the layer
 
You must log in or register to reply here.
Back
Top Bottom