Vuescan and Minolta Scan Dual IV problems: who can help a newbie?

[Alex Stols]

Hello,

I experience some problems with Vuescan (versions 7.6 as well as
8.0b3 and 8.0) scanning color negatives with my new Minolta Scan Dual
IV.

1- there appear faint colored bands on all scans in the direction of
the film holder movement, difficult to see, but present all the same.
At first I overlooked them, but as the edges are perfectly straight,
but they became easier to spot after I did some more scans. No
banding appears when I use the Minolta software.

2- when I set the "Number of samples" to 2 or higher, the results
show very brightly colored bands, all the same width, with sharp
edges. The predominant colors are cyan, magenta, yellow and blue. The
image is present, but only faintly. This phenomenon disappears at
scan resolutions of less than 1600, but not at 1600 and up.

3- the auto focus doesn't seem to work, I have to use manual focus to
get sharp scans. For example auto focus finds a setting of 0.925,
while testing manually I come up with 0.2. When focussing the scanner
is busy a much longer time in v 8.0 than in 7.6. But none of the 2
versions succeeded in focussing the image correctly.

I am very new at this work, so probably I am doing things wrong, but
whatever I try as suggested in the manual or as mentioned here in the
group: things stay as I describe here.

Scanning BW negatives is fine: I have the feeling Vuescan is doing a
better job here than Minolta.

Is anyone here having the same experiences with the same scanner as I
have? As I am a newbie I want to make sure it is not me before I
submit this to Ed. Thank you in advance for any help.

 

[Red Mage]

Hello,

I experience some problems with Vuescan (versions 7.6 as well as
8.0b3 and 8.0) scanning color negatives with my new Minolta Scan Dual
IV.

1- there appear faint colored bands on all scans in the direction of
the film holder movement, difficult to see, but present all the same.
At first I overlooked them, but as the edges are perfectly straight,
but they became easier to spot after I did some more scans. No
banding appears when I use the Minolta software.

2- when I set the "Number of samples" to 2 or higher, the results
show very brightly colored bands, all the same width, with sharp
edges. The predominant colors are cyan, magenta, yellow and blue. The
image is present, but only faintly. This phenomenon disappears at
scan resolutions of less than 1600, but not at 1600 and up.

3- the auto focus doesn't seem to work, I have to use manual focus to
get sharp scans. For example auto focus finds a setting of 0.925,
while testing manually I come up with 0.2. When focussing the scanner
is busy a much longer time in v 8.0 than in 7.6. But none of the 2
versions succeeded in focussing the image correctly.

I am very new at this work, so probably I am doing things wrong, but
whatever I try as suggested in the manual or as mentioned here in the
group: things stay as I describe here.

Scanning BW negatives is fine: I have the feeling Vuescan is doing a
better job here than Minolta.

Is anyone here having the same experiences with the same scanner as I
have? As I am a newbie I want to make sure it is not me before I
submit this to Ed. Thank you in advance for any help.

Yes, this is happening to me also, vuescan 8.0.11. Lots of color
banding, especially green and magentas, but this problem also exhibits
in b/w mode if I set the scan input type to b/w.

 

[Alex Stols]

On Sat, 15 May 2004 06:58:20 -0500, Alex Stols <> wrote:



Yes, this is happening to me also, vuescan 8.0.11. Lots of color
banding, especially green and magentas, but this problem also
exhibits
in b/w mode if I set the scan input type to b/w.

Hello RM,

Glad someone responded at last!
I have been quite busy with B&W negatives and I must say that I love
Vuescan for the way it handles the tasks.
In the mean time all my B&W negatives have been done once - maybe
later on I decide to do them again perhaps when I am wiser in the
trade...

Color I have started to do now: I use the Minolta sw and I get good
results. Only thing is - it is slow and I have to treat all negatives
separately for maximum color and contrast results.

All the time there has been no improvement as to the two main
problems the SD4 shows with Vuescan. No autofocus in any mode. No
multisampling possible and also slight color banding using just the 1
sample scan. I have mailed Ed H. repeatedly sending a problem report,
and I got the impression that he is still working on it. Ed has
stated earlier that the multisampling is now OK - it isn't.
Regarding autofocussing: no response from Ed on that subject as yet.

My question to you is:
Are you having autofocus problems too? If not, what workflow do you
use?

Thanks in advance for any response on these matters!

 

[Ed Hamrick]

I have mailed Ed H. repeatedly sending a problem report,
and I got the impression that he is still working on it. Ed has
stated earlier that the multisampling is now OK - it isn't.
Regarding autofocussing: no response from Ed on that subject as yet.

It's getting closer to the top of my list. I need to first
finish integrating the new Epson 2480/2580 (I bought one today),
fix a problem with the LiDE 20/30 on Mac OS X, and finish integrating
the Plustek 7200 dpi film scanner (Plustek loaned me one).

Regards,
Ed Hamrick

 

[Alex Stols]

It's getting closer to the top of my list. I need to first
finish integrating the new Epson 2480/2580 (I bought one today),
fix a problem with the LiDE 20/30 on Mac OS X, and finish integrating
the Plustek 7200 dpi film scanner (Plustek loaned me one).

Regards,
Ed Hamrick

Glad to hear this! I will praise the day that I can do color
negatives the same fast and accurate way I can do B&W.

 

[Red Mage]

It's getting closer to the top of my list. I need to first
finish integrating the new Epson 2480/2580 (I bought one today),
fix a problem with the LiDE 20/30 on Mac OS X, and finish integrating
the Plustek 7200 dpi film scanner (Plustek loaned me one).

Regards,
Ed Hamrick

Glad to hear it too -- Vuescan is just about useless for me as it is
now... Sometimes the banding is worse than other times, which tells
me its either a calibration problem, or something coming across
uncompensated for.

Something weird too: I tried the silverfast demo, which works OK for
this scanner (but I wont purchase, since I don't like the software
overall), but one thing is shows is 6400 dpi for this scanner! I've
tried it, and I get the resolution I expect, but is it REALLY doing
this, or some kind of software interpolation? Hmmmm.... Guess I
don't really need 100MB files, anyway!

Since I'm archiving right now, my workflow is simple: Optical scan at
highest resolution with minimal adjustments (focus, white balance,
some cropping), then bring into photoshop for first pass color
correcting and some grain reduction. (I use Applied Science Fiction's
grain reducer, which is nice, and Kodak's ROC software, which is
absolutely essential!)

Once it's in photoshop and archived, I can take a copy and work on it
for output.

It looks like vuescan will give me a bit cleaner scans, once the
banding problems are gone...

 

[Alex Stols]

Glad to hear it too -- Vuescan is just about useless for me as it is
now... Sometimes the banding is worse than other times, which tells
me its either a calibration problem, or something coming across
uncompensated for.

Something weird too: I tried the silverfast demo, which works OK for
this scanner (but I wont purchase, since I don't like the software
overall), but one thing is shows is 6400 dpi for this scanner! I've
tried it, and I get the resolution I expect, but is it REALLY doing
this, or some kind of software interpolation? Hmmmm.... Guess I
don't really need 100MB files, anyway!

Since I'm archiving right now, my workflow is simple: Optical scan at
highest resolution with minimal adjustments (focus, white balance,
some cropping), then bring into photoshop for first pass color
correcting and some grain reduction. (I use Applied Science Fiction's
grain reducer, which is nice, and Kodak's ROC software, which is
absolutely essential!)

Once it's in photoshop and archived, I can take a copy and work on it
for output.

It looks like vuescan will give me a bit cleaner scans, once the
banding problems are gone...

I don't know about Silverfast: the resolution is supposedly not above
3200, and in the scanner bake-off by Jim Hutchinson it appeared that
the practical resolution was nearer 20 line pairs per mm. I don't
know how to convert to resolution, but it means that under average
circumstances the number of black lines you can see separated is
about that figure - or about 550 per inch. I reckon that the
'resolution' then is about 1100 dpi or somewhat higher. Still a lot
lower than 3200, and certainly not 6400! You do get 3200 dpi anyway
at the highest setting, sometimes scratches as small as 2 pixels can
be seen clearly.

I'm archiving also, my workflow for B&W is, using Vuescan, the
'advanced' as suggested by Ed H., saving the RAW files and converting
to TIF/JPG to have a quick look. I wish I had Digital ICE )
I've tentatively started to use Minolta sw to scan color, but really
i'm waiting for Ed to get Vuescan right for the SD4

 

[Will G]

Ed, it's good to know that your little program is getting so much
attention. I started reading about it from Steve's D when I was
interested in the Epson scanners in the 2450/3200 class and now I am
about to decide on a 2580-class scanner.

Since you have recent experience with the 2580, can you indicate how
well the 2580 over USB2 compares to the 3200 device over the firewire
connection, using 2400 lpi optical as a reference? I'm very interested
in throughput issues, both with regard to scanning speeds, and in
comparing Firewire 400 vs. USB2 effective transfer rates. Essentially,
does the 11 msec/line of the 2580 overcome the 14.3 msec/line of the
3200 in a real multi-photo production scanning job?

And in general, does the type and speed of the host computer have a
great bearing on the throughput of scanning and color-correcting old
photos? For example, comparing a 466Mhz iBook with Firewire with a
Pentium II-450 (either USB2 or Firewire 400). Both have 384MB memory
and large disks.

Thanks for your comments and advice.
--Will