On 8/17/05, Moose <olymoose@xxxxxxxxxxxxxxx> wrote:
> I know this is true for the FS4000, and I believe it is generally true.
> With only R, G and B sensors, the R has to do double duty for IR and it
> can't separate the visible from invisible in one pass. Maybe it isn't
> true for some others?
I suspect you're right but wasn't sure if this was always the case so
that's why I said it's how it works with my scanner.
> I believe I remember you said multi-scanning does make a difference.
> After much reading before buying the 2720, I decided I needed VueScan to
> do multi scanning to get decent results. After some testing and learning
> to set the various scan parameters, I found that it didn't make any
> difference. I don't know if that will be true for the 9950, but I'll bet
> you don't need more than two passes. photo-i certainly got some good
> results without it. But then he was using Canoscan.....
Yes, I've noticed a lot of noise with Vuescan. My theory is that the
Canon software scans at 4800x9600 (the full optical resolution for the
scanner) and then downsamples to 4800dpi to reduce the noise. Or it
could just have a noise reduction option. I agree that with the 2710 I
didn't notice much advantage in the multi-sampling and since it had to
be multi-pass to do it I found that it really slowed down the
scanning.
I should have scanned one of the frames with no multi-sampling and 2
samples so I could compare, I'll do that tonight so I can add that
comparison to the page I'm going to create showing the results from
the scanner.
> I'm not sure separateing the steps that way saves any time overall, but
> it saves a LOT of time during the actual physical scan process. And one
> may then play around with various options on only a few frames to see
> what they do before setting it loose on the whole batch.
Yeah it's probably actually a little slower because it would have to
read each raw file and then process it but it would allow you to
finish the scanning process quicker as you say.
> Yes. done lots of that! At least the 4000 does 4-6 at a time.
Yeah even 4-6 frames at a time is a nice improvement. For the scanning
last night it took about 3.5hrs to scan 24 frames at 64 bit RGBI and 4
samples. That sounds really long at first but when you do the math and
figure out that it's 8.75 minutes per frame it doesn't sound so bad
especially when you consider I was writing 2 files and all the other
settings I was using.
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|