Scan quality from Epsons?

[Kirk Gittings]

Michael, All the improvements I have been refering to of the V750 vs. the 4990 are really incremental not dramatic. They are still nowhere near a drum scan or even an Imacon and the 4x barrier for flatbeds is intact for my tastes anyway. I primarily use flatbed for commercial work a small FA prints. Otherwise I go to the Imacon 949 or spring for a drum scan.

A well meaning guy sent me a sample of why I was wrong about that 4x barrier last week. He said his 30x40 prints were much more than acceptable but "first class" fine art prints form his 4x5 film and Epson 3200 scanner. Sorry, I would have been embarrassed to show those prints with mushy grain and oversharpened. I found an acceptable viewing distance at about 12-15 feet. That is not why I shoot LF. To each his own............

[Michael Mutmansky]

Kirk,

Oh my... that's 8x. I agree that 4x is pretty safe with the consumer flatbeds, but interestingly, 4x should be well within the range where differences between a flatbed and a drum scan are visible on some (if not many) images. Whether the differences are worth stepping up to a scanner is a point of debate, but the difference will be there for sure. That's the subject for another discussion.


---Michael

[Jay W]

I've often read posts talking about oversharpening, but I'm curious about what types of numbers people are talking about. I'm pretty green in digital printing and the number probably depend on subject, but a common number for me in USM is 50% and 2.0. Last night was a first for me in that I tried a gausian blur on a sky and I think it was 1.7. It looked pretty good in a 6x9.

Comments?

Jay

[Ed Richards]

MM - I wonder if the biggest gain with drum scans is with the worst negatives. I had good drum scans made of two excellent negatives and found little difference with 4x prints. In retrospect, I probably should have sent great images with problematic negatives to really see the difference.


Jay - the numbers on sharpening depend on a lot of variables. They change with resolution, grain, and fine detail, and, most importantly, with the final print size. There is really no value in comparing them. Took me a while to figure that out. You also want to look at selective sharpening. I often sharpen the foreground and do not sharpen the sky - use the mask and paint out the sky with black. Sometimes I blur the sky and blend that with the sharpened and masked layer.

[Kirk Gittings]

Also. There is some time improvements with Digital Ice (plus Digital Ice comes in two strenghts, lite and full now). I am getting scans in Silverfast with full DI at less than half the time I was getting with the 4990, same computer. I suspect it is a software improvement by SF, but on a 40 MB file I went from 16+ minutes to 6-8 minutes with the 750.