No announcement yet.

Review:Epson: EPSON Perfection 3200 PRO Color Scanner

  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    So Doug - are the larger resolutions available on the 3200 (above 300 dpi) primarily for scanning negatives rather than prints? I have beenb scanning my negatives at hardware maximum without interpolation but prints at a much lower resolution.


    • #17

      Just to make sure I have this right, to scan a 35mm slide (or negative) with enough resolution to print a 13 by 19 inch photo, I would need to scan at about 4000 ppi to get about a 16 Mpixel image (I have rounded off the conversions a bit). At full color and TIF that will be one big file.

      Of course all of the above assumes the original image was of high enough quality to support such a scan. Oh, and all bets are off if you want to crop that baby.


      After reading your review again, I see you have scanned 35mm at 3200ppi. So for the .945 inch by 1.427 inch image area on the 35 mm slide that is 3024 by 4534 pixels or 13.71 Mega pixels. At 300 dpi that is enough pixels for a 10 by 13.4 print. Hmmm, just rambling and doing the math.
      Last edited by catia; 01-07-2004, 12:44 PM.


      • #18

        Yes. Negatives, slides, 3d objects, etc.
        The one exception would be lineart (ie: logos, etc.) on smooth, glossy stock. Scanning up to 1200ppi can often be worthwhile there, especially if it's going to be converted to vectors.


        Yep, and add 16bit to the mix and they get even bigger. But for best quality, hundred-meg files aren't unusual. Send them out for a drumscan and they'll be even larger. And imagine slides larger than 35mm! Slides will support about 8000ppi scanning, but few of us can afford to scan at that level.
        Learn by teaching
        Take responsibility for learning


        • #19
          I'm going to try to understand all the various mathematics above...

          But I'd like to pop in here and say that with the encouragement of people on this site I bought a 3200.

          After a few rushed scans, just barely getting to know how to power up the scanner, how to find and minimally use Photoshop Elements, and Silverfast SE, I committed to an entire wedding.

          The wedding had been shot on 200 ASA Fuji. Processed, fouled, crossed over and scratched by a one hour lab.

          I was able with bonehead trial and error to produce a very credible set of prints, including some highly productive use of sharpening and some substantial color/contrast correction.

          In subsequent several months I've learned more.

          Most impressively, I've scanned badly fogged 35mm color negatives and printed them B&W, correcting for the fogging with the lasso.

          I'm a highly experienced (old timer) color / b&w photolab guy...I'm convinced that these B&W prints from virtually destroyed color negatives are as exciting and sharp as could have been accomplished at 11X14 magnification with a good enlarger and, in most cases, original B&W film..and I'd used a little unsharpen because I'd exaggerated a few details too far with increased contrast.

          3200 plus Silverfast SE plus Photoshop Elements plus 2200printer is capable of a LOT more than any traditional lab runs into, with the exception of the slooow workflow, the virtual impossibility of producing a nice fat stack of 4X6 proofs quickly.


          • #20
            I've just been printing 13X19 using an Epson 3200 and 2200. Museum Digital paper (

            I scanned 35mm slides at 800dpi, printed at 720 and (1200?). The output is beautiful, the difference between the two outputs is almost insignificant but can be discerned if you look closely. I have no doubt that 35mm can be printed far larger with higher-dpi original scans, not to mention Genuine Fractals. At 13X19 the only drawback Vs large C prints via 4X5 internegs (with which I'm very experienced) is that the scan is sharper but less "smooth" looking...a different look that might be minimized if someone wanted to get sophisticated with a little diffusion.