View Full Version : Scanner comparison re-animated
Two more scanners have been added to the comparison page ('http://largeformatphotography.info/scan-comparison/'). They are:
<li>Microtek ArtixScan 1800f, thanks to Ted Harris.
<li>Canon CanoScan 9950f, thanks to Ed Richards who turned the slide around in record time.
As usual, there was no attempt to achieve colour consistency beyond a grey-point click on the lightmeter case.
The 1800f scan was ressed up to 2400ppi. Ted tells that me the 1800f scan was controlled by Silverfast.
There appears to be a halo effect in the darkest areas, which may be
due to over-zealous recovery of shadow detail by the scanner software / driver.
Ted also gained an i800 scan, but it appears to have major technical problems. A replacement scan is sought.
And for those who have waited patiently (or otherwise) for these two scanners, I hope forbearance was justified.
This comparison relies on the largesse of busy people, as I am reminded by the presence of 600 emails in my 'Scan-around' folder.
The slide became delayed in UK & Europe for a few months last year, but, despite a few scratches and bruises, is still in languid circulation.
I am not sure I got the best results from the Canon 9950 - I mostly scan b&w and downsample it, but I do not have enough memory in my computer to do that with color. That said, take a look at the shadow detail and noise comparison - this is a simple scan, no post processing for noise removal or shadow recovery, just straight out of Vuescan.
My observations on these comparisons as I have owned both.... There is both a greater difference in sharpness and a greater difference in noise levels than I would have expected from my tests. That is the MT is much sharper but the Cannon has much less noise. I would have expected both results but not nearly as dramatically as shown.
On the MicroTec, where did you place the film. In the holder or on the glass?
The downsampling trick with Vuescan works with most all scanners no? To have done it here would have invalidated the comparison unless Ted had done it also with the MT.
what is the difference with your earlier posting where you did scan color and downsample in Vuescan? I don't understand the difference. You said in this thread:
"This is a two pass scan to allow 4800 dpi scanning of a color negative. The first pass is used to generate a raw (unprocessed) scanner file that is down sampled to 2400 dpi. This is done because the full size file is more than 2 gigs, which causes memory issues in Windows because processing the file can push past the 4 gig limit. Vuescan is then used to load the raw file for processing."
The transparency was in the glassless holder. I agree that the noise is surprising and I didn't see it in my original scan. Leigh and I have been talking about possible causes one of which might have been that I had the auto AACO feature in Studio turned ON as opposed which was my intent and usual practice .... can't believe i made a stupid goof like tht but hey, I have made many worse ones in my time and there is no way now for me to reconstruct what happened. When the transparency comes back to me again, as it will to do some different scans, I will redo the the 1800f and we can then see if there was some significant 'opeator error in the now posted scan.
> what is the difference with your earlier posting where you did scan color and downsample in Vuescan?
In the earlier posting I was able to get the downsampling to work. At the moment the Windows memory management demons are giving me fits and I cannot get it to work. I do not think this is a Vuescan problem, but a software interaction problem from something else I have loaded which screwed up my virtual memory management.
You are right that it it would invalidate the comparison had I done it, but I think it helps the sharpness. (Maybe that is just my imagination.)
I am as surprised by the low noise as you are. It is about as good as any of the scanners. Given that I am not making huge enlargements, it may have more to do with why the prints look good than the resolution.
One other difference - I increased the spacing between the film and the bed of the scanner since I posted the first example to reduce the Newton's rings on some of my bowed negatives. That might have pushed me beyond optimum focus.
Ed that sounds very plausible. It looks to me like the edges are softer than the center which is what would happen in that scenario. I actually use the Canon 4x5 film holder in my Epson 4990. It is a little better made. I adapted the calibration slot to fit the 4990 and I added a few layers of black masking tape in the film track to "grab" the film so it doesn't sag.
The masking tape sounds like a good idea. A holder that would tension the film would really help. I am not willing to go to the wet mount kludges for the 9950, but I eagerly awaiting more info on the Epson 750 and its wet mounting support.
I am hoping that the initial reviews of the Epson 700/750 are confirmed. It would give us a clear and relatively affordable choice.
I will second that about the new Epson's. I look forward to the possibility of having a single adequate scanner instead of the three I have now.
I will add my hope's to those of others regarding the new Epsons. The improvements are going to have to come in the optics and the spacing/alignment of the film with the scanning array/optics and other. less obvious features such as stepping motors rather than the array itself which is the same basic design as that currently being used.
Powered by vBulletin® Version 4.2.0 Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.