Recommended Cezanne Elite settings for scanning old negs

[jyarrow]

I am working with a company that is changing from analog to CTP. Many jobs they of course have in digital form which we are wrangling in to the Platemaker. However, they also have a substantial number of jobs for which they only have negs.

We are using a Cezanne Elite flatbed scanner to bring these into digital files and then lay them out and create new electronic masters to plate. I am curious about what the optimal resolution to scan at would be for the negs. It seems that we max out at about 1800 dpi, and so far I've been scanning 1800 RGB (3 channels), picking the best channel and chucking the rest.

Am I scanning at way to high of a DPI for what I need? Responses from those who have been down this path would be very much appreciated.

Jim Y.

[jyarrow]

This is for a shop that has an inventory of mostly line art negs. They do virtually no 4c (process) work. So that means that most of the negs being scanned are composed of rules, type, and the occasional screened background.

Jim Y.

[gordo]

I am working with a company that is changing from analog to CTP. Many jobs they of course have in digital form which we are wrangling in to the Platemaker. However, they also have a substantial number of jobs for which they only have negs.

We are using a Cezanne Elite flatbed scanner to bring these into digital files and then lay them out and create new electronic masters to plate. I am curious about what the optimal resolution to scan at would be for the negs. It seems that we max out at about 1800 dpi, and so far I've been scanning 1800 RGB (3 channels), picking the best channel and chucking the rest.

Am I scanning at way to high of a DPI for what I need? Responses from those who have been down this path would be very much appreciated.

Jim Y.

I've been somewhat down that road, many years ago.

You should scan at as high an optical resolution as you can that is an even divisor of the dpi of the output device that you will be printing to. This is to avoid artifacts caused by a mis-match of resolutions. So, if the CtP is 2400 dpi, then scanning at 2400, 1200, 600, 300 dpi are all even divisors. But if the CtP is 2400 dpi and you send it a bitmap that's 1800 then 2400/1800 = 1.333333 The .333 is uneven and will likely cause moiré.

Scan in greyscale or bitmap/line mode not RGB. At the least you should calibrate your scanner's settings so that you can scan a stepwedge gradient that contains tones from 1%, 2% 3%, 5%, 10, 15, 20, 30, 40 50, 60, 70, 80, 90, 93, 95, 97, 100% If you don't have such a stepwedge then make one and output it to film. Set up your scanner, PShop, so that you capture as much of that tone scale as accurately as possible. The final image should be a bilevel bitmap - no grays. If there are any grays then your RIP will screen them and you will get fuzzy text and lines.

A simple and practical way to test whether you have captured the tones accurately is to bring the stepwedge into PShop and apply a Gaussian blur just sufficient to create an even tone in a patch. Then use the eyedropper tool to measure the resulting grey.

One you've got your scanner and PShop setting defined for the stepwedge then those are the settings that you use for all the negatives.

best, gordo

[jyarrow]

I've been somewhat down that road, many years ago.

You should scan at as high an optical resolution as you can that is an even divisor of the dpi of the output device that you will be printing to. This is to avoid artifacts caused by a mis-match of resolutions. So, if the CtP is 2400 dpi, then scanning at 2400, 1200, 600, 300 dpi are all even divisors. But if the CtP is 2400 dpi and you send it a bitmap that's 1800 then 2400/1800 = 1.333333 The .333 is uneven and will likely cause moiré.

Scan in greyscale or bitmap/line mode not RGB. At the least you should calibrate your scanner's settings so that you can scan a stepwedge gradient that contains tones from 1%, 2% 3%, 5%, 10, 15, 20, 30, 40 50, 60, 70, 80, 90, 93, 95, 97, 100% If you don't have such a stepwedge then make one and output it to film. Set up your scanner, PShop, so that you capture as much of that tone scale as accurately as possible. The final image should be a bilevel bitmap - no grays. If there are any grays then your RIP will screen them and you will get fuzzy text and lines.

A simple and practical way to test whether you have captured the tones accurately is to bring the stepwedge into PShop and apply a Gaussian blur just sufficient to create an even tone in a patch. Then use the eyedropper tool to measure the resulting grey.

One you've got your scanner and PShop setting defined for the stepwedge then those are the settings that you use for all the negatives.

best, gordo

I can't even express how grateful I am for your detailed and precise response. Thanks for providing me with clear directions. I will work on setting up the wedge and calibrating our scanner.

Again, thank you so much.

Jim Y.

[jyarrow]

Gordo:

I did as you suggested, created my Stepwise Wedge, and am getting very strange readings across the board.

These are some of the readings I'm seeing:

1%=90%K
2%=86%
3%=83%
5%=78%
10%=71%
20%=58%
30%=58%
40%=40% !!!
50%=33%
60%=25%
70%=20%
80%=15%
90%=12%
93%=10%
95%=9%
97%=7%
100%=8%

This is pretty consistent against all 3 resolutions scans (300, 600, 1200). Is it possible some sort of calibration curve is (or needs to be applied) or is it possible that the lamp isn't outputting properly?

Thanks so much for all of your help.

Jim Y.

[gordo]

Gordo:

I did as you suggested, created my Stepwise Wedge, and am getting very strange readings across the board.

These are some of the readings I'm seeing:

1%=90%K
2%=86%
3%=83%
5%=78%
10%=71%
20%=58%
30%=58%
40%=40% !!!
50%=33%
60%=25%
70%=20%
80%=15%
90%=12%
93%=10%
95%=9%
97%=7%
100%=8%

This is pretty consistent against all 3 resolutions scans (300, 600, 1200). Is it possible some sort of calibration curve is (or needs to be applied) or is it possible that the lamp isn't outputting properly?

Thanks so much for all of your help.

Jim Y.

Yes it is possible (likely) that a curve will need to be applied. Also, are you scanning the film as a transparency or as reflective copy? Most dedicated copydot scanners that I've had some familiarity with scan the film as reflective copy not as a transparency. This is because when light passes through film the halftone dots act like little camera apertures and distort tones.

best, gordo

[jyarrow]

Thanks again for the response Gordo. I have more information so I can run another test and see if I can get what I'm looking for using the more correct settings (it seems I was using the wrong film settings). I'll test it again and get back to you.

Jim Y.

[jyarrow]

Somewhat counter-intuitively, one cannot use the Negative or Black Separation on the Cezanne to scan black negatives. One must use the Mechanical TRN setting. Once you do that, it scans as a 1 bit scan (thus no grayscale). Mystery solved. Thank you all for your kind input.