TonySleep@halftone.co.uk wrote:
>
> On Fri, 08 Feb 2002 10:09:42 -0800 ThomasH (herthh@attbi.com) wrote:
>
> > With every new version the
> > declining speed of Vuescan is becoming a significant problem.
>
> I told you so! :) You can't have every feature under the sun *and*
> speed.
>
> With my workflow - do most of it @48bit in PS - I'm still using
> 7.1.something for real work. I think the time has come where VS needs
> to diverge into Vuescan Hair Shirt and Vuescan Kitchen Sink versions.
Yep, I also still have the 7.1.24 and 7.2.7, the last version
before the exposure locks were introduced and the development
shifted toward the new user interface.
According to Ed the newer versions of the IR cleaning and the
grain removal attribute to the slow down of the process. This
"tyranny of complexity" shows though even more clearly the
excellent quality of algorithmics done at AFS: Their ICE and
GEM processing is now dramatically faster compared to the
present Vuescan algorithms, by a magnitude of 4:1 up to 8:1.
So far though I was not very concerned about the speed because
I do photography for my pleasure and for my friends only. I put
a roll into the SA-30 late evening and the next morning the roll
was scanned. The 8min per image were already a lot (approx. 5
hours) but the night is long enough.
Now I face the hard reality that the SA-30 hardware spills out
the film in the process. The Vuescan 7.5.0 is at least for the
SA-30 off limits!
And since the 7.4 version was released by mistake without the
film offset adjustment and never corrected as (say) 7.4.3, the
latest version which I can use is 7.3.13 and I like it actually.
It has still the old single preview and "Scan"--"Scan Mem" scheme,
which was juuuuust fine.
Vuescan's grain removal in 7.5 has improved indeed. The LS-4000 has
enormous problems with Ektachrome. The scans appear very grainy.
This is not necessary the film grain, I have no explanation for
this effect. Maybe its a property of the Kodak celluloid breaking
up the LED light used by Nikon? NikonScan uses ASF GEM, which works
incredible wonder on the scan but the color rendition of the "Nikon
Color Management" is so often unacceptable. This grain removal
improvement in Vuescan is thus a very valuable improvement but it
comes at a price which is not acceptable. Nikon's hardware
implementation in the roll film adapter SA-30 ejects the film after
some undocumented amount of minutes. A scan at 4000dpi, size
reduction 5 and medium grain removal takes now 11-12 minutes on a
333Mhz Pentium with 384Mbytes and the film gets ejected after every
single scan...
Thomas.
>
> Regards
>
> Tony Sleep
> - Online portfolio & exhibit; + film scanner info
> & comparisons
----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or
body
