Ralf Schmode wrote: > looks as if my brand new Scan Elite II has got at least three problems > rather than one: Please look at > > http://schmode.net/banding_channels.jpg
You could try locking the exposure, putting some opaque card in the slide holder, doing a scan and making exactly the same gamma adjustments. If it shows exactly the same problems (and I'm sure it will) it cannot be to do with any foreign matter in the light path. As far as the individual lines are concerned, my view is that this is a blackpoint calibration issue, especially as it only affects the deeps shadows. If you select the offending row in Photoshop and use Select/Color Range to select just the shadows, adjusting the black point and gamma makes it identical to the other rows nearby. I'm sure I recall reading that there are keystrokes for the Dual II that can be used within the Minolta software to force the unit to recalibrate itself. If there is a similar function in the Elite II software, it might be worth trying that. Was this scan done in 16bit mode? It might also be worth doing a raw 16bit scan and seeing how it looks using Vuescan. It's a pity that Vuescan doesn't support the Elite II yet as Vuescan's own calibration process might show different results. Interestingly, my Elite suffered from similar issues as your Elite II, but not as badly - there were only a couple of tracks in the green channel and no banding in the red channel. You may just have a rogue unit. Presumably, no Scan Multi Pro users are seeing any similar problems - everything I've read about it so far has been very good. (Similarly the older Multi models seemed fine in this respect.) This would seem to suggest that there is something very different in the design, component selection or manufacturing process for the Dual II/Elite II and for the Multi Pro. Any comments from Multi Pro users on CCD issues or dynamic range? I await the arrival of my Elite II with considerable interest! Al Bond ---------------------------------------------------------------------------------------- Unsubscribe by mail to [EMAIL PROTECTED], with 'unsubscribe' in the title or body
