PDA

View Full Version : Scanning Workflow



neil poulsen
8-Feb-2006, 03:41
I've read through many of the archived threads on scanning, but still have the following question.

Looking at Silverfast AI or Epson Scan for the 4870, there are multiple image adjustment capabilities, including curves, levels, unsharp mask, color adjustments and the like, that are also available in Photoshop. To obtain the best scans, which of these should be applied in the scanning software, and which should be applied in Photoshop. Or, does it matter?

It's my understanding that the adjustments in scanning software for scanners like the Epson 4870 or 4990 flatbeds do not affect gains, they only numerically affect the resulting digital file. So, if I were to follow my instinct on scanning workflow, I would adjust only those that relate to scanning directly, like resolution, whether to invoke digital ice, bit depth, and the type of film using the scanner software. I would adjust the other attributes like levels, curves, sharpening, color balance, etc., post-scanning in Photoshop which has greater flexibility in these tools.

Is this strategy correct? More advanced scanners do directly affect gains (as I undersand it), so the optimum scanning workflow would be different. But, my interest is in the medium priced flatbed scanners like the 4870 or 4990, since this is what I and many others own.

ronald moravec
8-Feb-2006, 05:31
I try to get it right from the scanner as far as color, contrast, curve shape and density. Save as tiff. Use PS for dodge, burn, filter additions, size for web, local color changes.

I don`t use use ICE or anything except a clean negative. One would not print a dirty neg, why scan one.

Ted Harris
8-Feb-2006, 06:43
Neil, the quick answer is that you want to make those adjustments that impact the amount of information that will be present in the file produced by the scan and at a minimum they include setting the black and white points and doing some curve adjustment. Resolution, if by tht you mean doing any sharpening should not be done s part of the scanning (all the algorithms used by scanners are way too simplistic) but left for Photoshop. BTW, I am using the term 'curves' here a bit generically because any adjustments that need to be made in the scan to insure tha tthe file includes all the information you want available for further manipulation in Photoshop needs to be considered, generally that involves the information at the far ends of the spectrum, e.g. the blacks and whites (think in Zone terms of you seet exxposure and development that way). Color values also should not be ignored if there are strong color casts which are easier to remove when you scan than in Photoshop.

For a step-by-step approach take a look at the third of the articles Michael Mutmansky and I did for View Camera on scanning. For a much longer discussion you should get "Real World Scanning and Halftones" (3rd edition), it is the only book I know of that focuses on the scanning process specifically and it does it very well.

Kirk Gittings
8-Feb-2006, 10:30
" I don`t use use ICE or anything except a clean negative. One would not print a dirty neg, why scan one.'

Because if you are scanning some of your vintage chromes that have been all over the world to magazines to be scanned, they are in pretty bad shape.

Ted Harris
8-Feb-2006, 11:05
Kirk's comment is well worth noting. "pretty bad shape" is even an understatement. I just finished scannning a 50+ year old 8x10 chrome that had been in and out of the hands of numerous magazine editors, etc. and it needed a lot of attention. FYI, I decided NOT to use any of the automatic digital solutions, such as ICE but went through the digital equivelent of spotting and dodging and burning and touching up. Took hours but was worht it.