Re: Epson 10000XL scanner and i1 profiler
Re: Epson 10000XL scanner and i1 profiler
- Subject: Re: Epson 10000XL scanner and i1 profiler
- From: Florian Höch <email@hidden>
- Date: Mon, 01 Sep 2014 19:13:55 +0200
Am 01.09.2014 um 16:04 schrieb MARK SEGAL:
> Hi Florian,
>
> Wasn't worried and I got the spirit of your comment. I was technically
> interested, as a person with quite some years of scanning experience
> under my belt (Scanning Workflows
> <http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml>
> ), to examine your comment in the context of both this thread and user
> experience. I agree with you that it would be interesting to see
> rigourous comparisons of those discontinued high-end behemoths, versus
> the current and recent crop of the better prosumer models. I have either
> worked with or tested about half a dozen of the latter - the better ones
> - and didn't experience banding or unevenness of illumination.
To be honest, I don't really have experience with actual recent prosumer
models, so I'm not really in a position to asses whether they have
improved over what I've used some seven years ago ;)
The problems I had back then were mostly related to reflective scans
(the Minolta film scanner on the other hand produced imo quite good film
scans), e.g. technical drawings on cardboard-like material at around A4
or letter size. We always had to go into Photoshop and "fix" the
backgrounds because the scanner produced visible gradients (lightness
and also to a lesser extent color shifts) over what should have been
uniform, a problem we never had with the Nexscan (but of course that one
also cost and weighed a multiple of any prosumer scanner). Also,
occasional slight bands (or rather, narrow strips) that were either
slightly lighter or slightly darker than the scanned original were also
a recurring problem.
> My tests, however, revealed without exception that tested resolving power falls
> quite short of advertised resolution. This was not a revelation to an
> industry specialist with whom I peer-reviewed my findings. I also
> considered that perhaps the testing technology itself is defective, but
> that seems questionable. The line pairs printed into the target are
> clearly visible at all resolution levels seen through a 30X microscope,
> so unless the viewing and printing of these scanner readings are
> impaired at the level of an NEC PA271 display or an Epson 4900 printer
> (I doubt), advertised resolution is usually below revealed resolving
> power. The scanner sensors may allow for the advertised resolution, but
> save for the qualifier about the testing above mentioned, these scanners
> are probably lens-constrained. That said, the DR and the resolution of
> the best prosumer scanners these days are high enough to make good
> quality prints from 35mm film media at least to A3 size, provided the
> user knows how to use software to advantage. Much also depends of course
> on the technical quality of what's on the media and the media itself.
It's also my experience that prosumer models (atleast the flatbeds I
used, the film scanner came close to its advertised resolution) used to
have much lower effective resolution than what was advertised.
> All that is why I had reservations that Peter's problem is
> hardware-related. My first thoughts turned to software settings and my
> second thoughts - en passant - to the brightness or reflective character
> of the media - on the latter of which Ernst has explored in some depth;
> on reflection (pun intended), it would appear that the character of the
> media may well be the real culprit crushing the highlight detail; if so
> it could be that this is best overcome using a DSLR with a good macro
> lens and an appropriate lighting set-up to catch the paper texture.
Makes sense. I wonder if the particular model Peter is using does allow
for some sort of "raw" mode? It would be interesting to know if the
scanner can resolve the paper detail at all, or if it's indeed a
software-related problem (I'm inclined to think it's the latter rather
than the former).
--
Florian Höch
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden