>> I made a test slide where I took a black slide
>>(unexposed) and punched many needle holes into it and scanned it. So they
>>are plain holes now in the film. You can see the result at
>Harry, I tried the needle hole test with my Minolta Scan Dual
II/VueScan and got no trace of the ghost imaging you are seeing.
OK, Harry's been taking some ribbing here on a genuine problem, but his scan
(above, and compressed & attached here as StellartcropOrig--excuse spelling,
please) shows it clearly.
The light is definitely bleeding out from the horizontal center, as you can
see from the top and bottom "stars." Rob is right that the effect has a
name, but it escapes me, too. I think "halation" is not quite it, but that's
Anyway, using Harry's pin-prick method with a piece of black neg leader, I
did the same thing Roger did with my Acer Scanwit at 2700dpi (Stellartest1).
No ghosts, no bleeding. Actually, I expected quite a bit of noise, and got
some, but it adjusted right out with the curve tool in MiraPhoto.
But since Harry had used Vuescan, I tried that program, too, giving it
"full-tilt" at 48-bits. VSstellarscan is the Raw scan, unadjusted. As you
can see, Vuescan somehow "pumps" a lot more light through (notice the larger
size of the same holes, and how it burns through the leader). For me, this
is *great* and now I'm going to revisit some of my impossibly-dark slides
with VS 7.x!
For the last test, I adjusted the color controls in Vuescan to get a black
background--which meant a gamma of 1.8 and a brightness of .3! You can see
in VSstellarcrop that the "stars" are clean, with no ghosting--although at
72dpi and JPEGed they're not as sharp as the original tests.
So I dunno, guys--is it the lens or the light-source? Is something wrong
with the scanner? Seems like.
FREE! The World's Best Email Address @email.com
Reserve your name now at http://www.email.com