Help, please, I could not find clear answers fo these, sorry

Dear Professionals,

 

I have few questions being new to the business.

 

Setup: I scan negs and slides using Nikon 5000, and want to use vuescan as an

afforadble soft.

 

1. Negs. I learned that I have to take an unexposed partion and scan it in order to

determine the exposure and base color (super-advanced workflow by Erik Krause

is a bit doubtful for me from the mathematical point of view, anyway, it is not

related to the question directly).

 

How do I know that the exposure got this way is the correct one?

 

2. I want to make raw files. Does "lock film base" color affect raws?

If not, I hopefully can scan them (raws) and set film base color later.

 

What does it mean "scanner color space" and "device RGB" output when I scan

from a raw file? Does raw file has any info what scanner was used and what is its

RGB?

 

3. Slides: How do I know what exposure should be set to scan rolls and calibrate

(I'll use velvia and provia targets). I mena, I inderstand I have to use the same

exposure for calibration and scanning. But how to set it correctly?

 

Thanks for your help and the time spent,

Alex

Don`t know a thing about Krause. But I will take you back to my color darkroom and

perhaps it may clarify things a little.

 

Color prints need to be color balanced so they do not appear to have a color cast.

Every emulsion batch needed a different balance so one time saver was to purchase

many rolls of the same batch. One color balance would print them all eliminating

testing. After many years, I was able to determine that it was not the emulsion

coating that varied, but the masked color layer. I could put my color meter on the

blank area between frames, neutralize that with the enlarger filtration and print with

success until I changed type of film. The big variable batch to batch of a certain kind

of film was the color of the base.. I could also color balance on a black or white with

equal success.

 

When scanning for printing, the orange mask base is unnecessary. I believe Rollie

scan film eliminated it so it is back to the original color film, no orange mask.

 

Evidently what he is trying to do is electronically eliminate the base color the and

just work with the color layers.

 

When you scan the color base, it is eliminated if the scan color is black and not color

casted. If it goes to grey, then there is too little exposure. The overlaying color when

you scan the whole neg will intercept some of the light going to the scanner sensor

and make it less than black. This will be no different than making a contact sheet

and seeing black rebates on the film.

 

Evidently he has some fancy mathmatics to determine the base is not too black, but

just black enough.

 

I have been a seat of the pants darkroom worker so long I don`t use anything and it

has transfered to digital. I look at the print under the proper light and it is either right

or wrong and I fix it. Once I calibrate the screen to my printer using small prints, I can

print anything that looks good on my screen. Now if I sent files to a publishing

house, I would make an effort to do better, but as long as I can control the final

output, if it looks right, it is right. My pro lab send back perfect prints from my files.

 

I think all the fancy software to scan is way to complicated. I use the Minolta 5400

software that came with the scanner, I scan on manual. Get the density and color

right on a prescan, save those settings for the next neg, and do a final scan.

Photoshop does all the rest better, easier, and faster and makes scanning go faster.

"2. I want to make raw files."

 

Please be aware the the RAW files produced by the scanner is not the same RAW that is produced by Nikon cameras. What software are you going to use to process the RAW files?

I use film camera and then scan films using Nikon Coolscan 5000. Raw files, I mean raw files, produced by vuescan including infrared channel.

 

My questions are specific for vuescan, I would say. Other soft is either expensive or useless for me, I guess. My wish is to make raw files and then, in some time, say in few months process them through some workflow which I do not know now. I will dig in.

 

I'm not so stupid to understand that scanner exposure does affect raw.

However, it is a mystery for me how scanner adjust exposure scanning just a gap between frames and why the resulting value of exposure is correct?

 

Basic question about negatives: does "lock film base color" affect raw produced by vuescan or not? If not, this simplifies the current task which is scan of physical media.

 

Sorry, if I put questions not clear from the beginning

"2. I want to make raw files. Does "lock film base" color affect raws? If not, I hopefully can scan them (raws) and set film base color later. "

 

No- film base color is a software correction. Eric's method just has you set per channel analog gain to approximate this in hardware. I'm curious as to why you don't think that's better than doing it in software. The "gap between frames" or a blank frame gives you a reading for the orange mask which needs to be removed one way or another. If you lock film base color then even if Vuescan can't find any blank space around the frame you should still get a decent result- it won't try subracting the film holder color from your image or otherwise screw up!

 

"What does it mean "scanner color space" and "device RGB" output when I scan from a raw file? Does raw file has any info what scanner was used and what is its RGB?"

 

No- a raw file has no color space assigned to it. Now the next step is to take the Vuescan raw file and do something with it. If you stay in Vuescan you need to output TIFFs. At that stage the color space you set becomes important. Slides only- you can profile linear raw TIFFs or gamma corrected TIFFs with output color space set to "device RGB" (basically no color space. If you're just shooting negs output to a reasonable working space like AdobeRGB or ProPhoto.

 

"3. Slides: How do I know what exposure should be set to scan rolls and calibrate (I'll use velvia and provia targets). I mena, I inderstand I have to use the same exposure for calibration and scanning. But how to set it correctly?"

 

Try to expose the raw histogram for the IT8 targets as far to the right as you can without clipping the whites.

Roger, I think, you opened my eyes. Just to repeat that I have got the point:

 

For negs: I scan the gap, this adjusts the exposure. Then I make a number of raws. Then, working with raws I assign as output my favorable profile, say AdobeRGB. Correct? (or alternatively use device RGB and then convert in PS)

 

May be I'm really stupid but I have not understood yet why the exposure, somehow set by scanner on the gap is correct? My worry is the following:

How can I be sure that the exposure set on the empty frame will not be too low (and will still give details in dark areas, which are light in negs) and will not be to high (and will still give details in highleighted areas, which are dark in negs)?

or may be the best way is a scan of a frame with highlights and the gap at once, so that scanner will find the best value.

I mean, empty space is just transparent. Suppose I shot on sunny sea with very bright water in the direction of sun. Most likely I'll have almost white close to the sun. It will be extremly ddark on negs. How I can be sure that exposure, set by transparent area will be enough to distinguish hues in the bright area?

 

Erik's method involves increasing of the gain which means increasing of noise. Also I do not see why changing the gain per channel in order to neutralize the base will not affect (at least slightly) main colors? Anyway, I have to write few formulae myself to see what is going on.

 

Slides: Is it correct to calibrate with targets, then assign the resultant profile as the scanner profile, set output to device colorspace and eventually convert it to the workflowprofile in PS?

"May be I'm really stupid but I have not understood yet why the exposure, somehow set by scanner on the gap is correct? "

 

You're just confusing yourself.

 

If you are scanning color negative film *one exposure will work for every single roll of the same type of film*. You should also be able to reuse your film base color settings (or analog gain based on Eric's workflow)- with a competent lab there shouldn't be big variations in processing.

 

By setting the exposure on the clear part of the film you will not blow out the shadows on neg film. Your scanner has a good enough dynamic range that the highlights can fall where they may and you shouldn't have any visible scanner noise. Remember that color neg film has a very small dynamic range and does not test your scanner hardware at all.

 

"Erik's method involves increasing of the gain which means increasing of noise"

I think that's totally wrong. I don't have a Coolscan but try testing it with a stopwatch. I believe it increases the amount of time the LEDs light up your film on a per channel basis. Exposure on my scanner directly correlates to the amount of time the film gets exposed by the scanner (Twice the exposure= twice the time). His method will not increase nosie.

 

Slides- figure out proper exposure (analog gain) with your IT8 targets, save the settings, scan all future slides of that type with the same exposure settings. I recommend outputting from Vuescan as device RGB and then "assign" the right scanner profile in Photoshop and then convert to your workingspace (AdobeRGB, ProPhoto, etc).

Yeh! Now I feel myself educated. Thanks a lot, Roger! Thanks everybody for your efforts and time.