Re: filmscanners:Vuescan bugs isolated

2001-12-18 Thread jimhayes 

Ed,

I was just testing this very thing. I tried changing three things: 1) focus
from always to on preview; 2) Changing both monitor and file space from
Apple RGB to Adobe RGB and back (remember I have a bw image); and 3)
checking/unchecking the release memory box. I normally leave this unchecked,
since Pshop can run with vuescan up.

That's the variable that changes things Ed...release memory reflips, lightens,
and compresses to 1/2 the contrast range in pshop. When I uncheck it I get nice,
full histograms, correctly flipped image.

I can also tell you that on my windows 98SE system, both the displayed image in
Vuescan and the bw image in Pshop are a little darker with Adobe RGB selected
vs Apple RGB. It is also reflected in the pshop histogram. I hope that helps
solve your color space problem.

[EMAIL PROTECTED] wrote:

 In a message dated 12/18/2001 4:16:28 PM EST, [EMAIL PROTECTED] writes:

  Ed it looks as if the SS 4000 pulls the film holder in and out a few times
 or
  more
   during scan.

 Do you perhaps have the Prefs|Release memory option turned on?

 This option causes this kind of problem.  It's fixed in 7.3.6, which
 I'll release in a day or so.

 Regards,
 Ed Hamrick

--
Jim Hayes

Re: filmscanners:Vuescan bugs isolated

2001-12-18 Thread EdHamrick 

In a message dated 12/18/2001 5:35:10 PM EST, [EMAIL PROTECTED] writes:

 That's the variable that changes things Ed...release memory reflips, 
lightens,
  and compresses to 1/2 the contrast range in pshop. When I uncheck it I get 
 nice, full histograms, correctly flipped image.

Yes, I've found the problem and fixed it.  The problem was that it
has to do two passes over the memory buffer in 7.3.5, once
with the Monitor color space and another time with the File
color space.  The memory buffer was getting cleared after
the first pass, so the passes for the file writing were going all
the way to the scanner.

  I hope that helps solve your color space problem.

It's fixed in VueScan 7.3.6 - thanks to all the people on 
this mailing list for helping solve this.

Regards,
Ed Hamrick