While we're on the subject of scanners, here is a comparison I made of the Minolta Dimage Multi Pro Scanner versus the Microtek ArtixScan 2500f at very similar settings. No sharpening. A 6x7 slide. What a difference between a dedicated film scanner and a flatbed. The dMax of each is comparable, but the difference in resolution is telling.
If I had it all to do over again, I would have kept the Minolta and just gotten an Epson 3200 for the 4x5. I may still do that. !
The Canon still wins in terms of resolution, but it's becoming harder to see the difference.
It's easiest to compare images of the same size. So if you upsize the lower-res scan, and place it next to the the high-res, the differences become more obvious. For example, see here. The two images were scanned at 1250 and 2500 ppi, but the 1250 ppi image has been upsized to match the larger image. A little sharpening helps the comparison too.
The newest file is a 3200 scan with the clipping problem manually corrected.
A poster in a Usenet group gave a heads-up the other day about Amazon listing these at $327 (free shipping, too). I was considering one and impulsively bought one at that price when I checked out his information. It has since had the price raised but it might bear watching if you're considering the purchase of a 9950F. Mine is back-ordered to 11/3 at this time.
This reviewer is not very impressed
"...soft but good for refrigerator prints or Web work." ...?
http://www.cnet.com.au/accessories/scanners/0,39026179,40001986,00.htm
Scanning 8x10 -
The Scangear software won't let me input 8x10 for a film size. It seems to want to find an image from the standard adapters. I downloaded a test version of Vuescan but it hung in the preview mode. So I don't have an answer for 8x10. It does scan a 7x9 area when doing 35mm strip film.
I have Silverfast software for the Epson 3200 but I believe it is scanner specific. I haven't used Canon scanning software before, but I like it better than Epson or Silverfast. The Canon scanner with Canon software produced a reasonably accurate scan right out of the box with no profiling. (Although I'm sure it can be improved with profiling, which I will do if I use the scanner for more than play).
I'm not going to try and test 8x10 again , but I will take another run through the Canon software to see if I'm missing something.
Tonight I'll find some slides of dark man made objects and maybe we can get a feel for real dmax and noise.
I'm seeing better gradiation of tonal values with the Canon. I'm not sure if this is just a function of higher true resolution, or if the Canon captures more real values than the Epson.
I'm seriously considering one of these scanners to scan b+w negs for a book project.
They will be 4x5 tmx 100 negs, and will be enlarged up to about 3X.
Most of the negs have a relatively low D-max, but there are some exceptions.
I need the scans to be 720dpi at final file size.
If anyone has any thoughts on how this scanner might do for the project, please let me know.
My frame of reference is the epson 1640xl that we have where I work. The results from this scanner are fine, but If I can get significantly better results from the canon, I'll buy it just so I can work on the project at home. How does the 1640XL compare with the 3200?
Thanks!
Paul
Also, no one has mentioned the epson 4870.
Thoughts on this vs. the 9950f?
"They will be 4x5 tmx 100 negs, and will be enlarged up to about 3X. Most of the negs have a relatively low D-max, but there are some exceptions. I need the scans to be 720dpi at final file size. "
Given that the 1640xl actually delivers around 60% of the advertised rate, it really only gives around 1000 ppi. If you want to create output at 720 dpi, then you can enlarge by around 13% , and still retain critical sharpness - a far cry from 300%.
If you want output at 720 dpi, that means you need to scan at 2160 ppi. (That way, when you enlarge and thus divide 2160 by 3, you will still end up with 720 dot per inch). If you use a scanner like the Epson 3200, which actually delivers around 60% of 3200, you will be fine, and you will get really sharp images at the size you desire.
Thanks, Ken ... I'm also interested in a more subjective evaluation that might include actual edge sharpness, tonal separation at the extremes of the scale, evenness of illumination, and while we're at it, which company's magic dust and scratch removal technology works better, if at all.
Posting Permissions
Reply With Quote
Bookmarks