Minolta 5400 questions

[dseuser]

I came across a problem scanning slides with the Minolta 5400
(original), and would appreciate comments from other users.

The slides are scanned with the Minolta Scan Utility. In the preference,
16-bit linear is selected, and Color Match disabled. No color
corrections are made during scanning. The image in the scan window looks
quite close to the slide, and there are no channel clippings. I believe
that this produces a raw scan.

When the scan is opened in PS CS, it is assigned to the Minolta
Posi-Linear profile and converted to the AdobeRGB98 working space. The
crt monitor is calibrated. The displayed scan looks very different from
the one in the scanner, with a noticeable color shift. PS' histograms
also look different from the scanner's. When rgb info is sampled at the
some location in PS, it is also different from the same sample in the
scanner. The bottom line is that it takes quite a bit of PS correction
to match the slide.

Questions:

- Should the histograms and sample info be different, and why?
- Can this be fixed with a custom scanner profile?
- Can this be fixed any other way?

Thanks.

 

[Robert Feinman]

I came across a problem scanning slides with the Minolta 5400
(original), and would appreciate comments from other users.

The slides are scanned with the Minolta Scan Utility. In the preference,
16-bit linear is selected, and Color Match disabled. No color
corrections are made during scanning. The image in the scan window looks
quite close to the slide, and there are no channel clippings. I believe
that this produces a raw scan.

When the scan is opened in PS CS, it is assigned to the Minolta
Posi-Linear profile and converted to the AdobeRGB98 working space. The
crt monitor is calibrated. The displayed scan looks very different from
the one in the scanner, with a noticeable color shift. PS' histograms
also look different from the scanner's. When rgb info is sampled at the
some location in PS, it is also different from the same sample in the
scanner. The bottom line is that it takes quite a bit of PS correction
to match the slide.

Questions:

- Should the histograms and sample info be different, and why?
- Can this be fixed with a custom scanner profile?
- Can this be fixed any other way?

Thanks.

I have never had any luck with the Minolta scan utility. Why
don't you download a trial copy of Vuescan from hamrick.com and see
if you are happier with the results.

 

[Bart van der Wolf]

The slides are scanned with the Minolta Scan Utility. In the
preference, 16-bit linear is selected, and Color Match disabled.
No color corrections are made during scanning.

As a minimum, you *should* adjust the individual R/G/B channel
exposure levels (inappropriately called gain) to something neutral in
the whites/light grays. If you don't, then the posi-linear profile
will have a wrong base-line.

Bart

 

[Bart van der Wolf]

SNIP
As a minimum, you *should* adjust the individual R/G/B channel
exposure levels (inappropriately called gain) to something neutral in
the whites/light grays. If you don't, then the posi-linear profile
will have a wrong base-line.

Of course, this is another poorly written forgery. I would never write
something so totally wrong.

I love Vuescan and have been using it for years but Vuescan *is* very
buggy. Most of all with the Minolta 5400. I know because I have one.

So due to all the bugs I don't scan with Vuescan anymore but use it as
an image editor. Specifically because of the VueScan D-max bug.

I would urge all people to check the archives. Long before all this
message faking started I've complained about the VueScan D-max bug
repeatedly and have done so for years . The bug is still there and
it's only getting worse.

Bart

 

[Don]

What D-max bug is that?

Don

 

[Bart van der Wolf]

SNIP

What D-max bug is that?

This one:

SNIP

Yes that'll be one of the effects the VueScan D-max bug will cause.

SNIP

And then this one:

SNIP

Which, indeed, seems to have resurfaced. Yesterday I scanned the
'bake-off' image with VueScan 8.1.40 and noticed that the left side of
the histogram was looking funny.

SNIP

And have you forgotten this:

SNIP

SNIP

Unfortunately, to date VueScan is not capable of scanning the Raw data
with a linear gamma. The higher film densities are scanned with
reduced contrast. This results in relatively lifeless shadows in
slides, or dull highlights in negatives. Mind you, it's a matter of
the software not interpreting the scanner response as it should, the
Minolta software does it as can be expected.

SNIP

And so on and on...

Vuescan is just riddled with recurrent bugs.

Bart

 

[Don]

And again, another boring forged e-mail from me.
It's from supernews, and Bart doesn't post from
supernews.

VueScan 8.1 is from more than 2 years ago.
There is no D-max bug in VueScan, and not
a single person has reported one here or anywhere
else in the past 2 years.

VueScan is a wonderful program, and I recommend
it wholeheartedly.

You can download it from:

http://www.hamrick.com/

Don

 

[tomm42]

I came across a problem scanning slides with the Minolta 5400
(original), and would appreciate comments from other users.

The slides are scanned with the Minolta Scan Utility. In the preference,
16-bit linear is selected, and Color Match disabled. No color
corrections are made during scanning. The image in the scan window looks
quite close to the slide, and there are no channel clippings. I believe
that this produces a raw scan.

When the scan is opened in PS CS, it is assigned to the Minolta
Posi-Linear profile and converted to the AdobeRGB98 working space. The
crt monitor is calibrated. The displayed scan looks very different from
the one in the scanner, with a noticeable color shift. PS' histograms
also look different from the scanner's. When rgb info is sampled at the
some location in PS, it is also different from the same sample in the
scanner. The bottom line is that it takes quite a bit of PS correction
to match the slide.

Questions:

- Should the histograms and sample info be different, and why?
- Can this be fixed with a custom scanner profile?
- Can this be fixed any other way?

Thanks.

Try running the scan with out the Minolta profile, always found I have
better scans without their profile with a Scan Multi. Definitely use
the curves control to adjust the color.

Tom

 

[dseuser]

I have never had any luck with the Minolta scan utility. Why
don't you download a trial copy of Vuescan from hamrick.com and see
if you are happier with the results.

The trial version does not support using a profile, or capturing a raw
scan.

Assuming that you scan slides with VS, do you use a profile and scan
raw? Do you see the same problem I mentioned above regarding a different
histogram and rgb samples in Photoshop?

 

[dseuser]

From the archive, the Real Bart suggested neutralizing a film base for
negatives. So I assume that this is a post from him.

I thought that it is for negatives only, but I tried it with an
unexposed slide (all black) and noted a few very interesting points.

If Auto Exposure for slides is enabled in the preference (my normal
scanning setting), the scan preview will show up pixelated and very
bright (rgb in the 2xx range). If the Auto Exposure is disabled, the
scan preview is all black (rgb in the 0 range). Now I wonder whether
Auto Exposure should be left on or off to neutralize the grays, and for
normal scanning.

I also noticed that the mouse pointer maybe sampling at single pixel,
instead of at 3x3 as in Photoshop. With the slightest mouse pointer
movement, the rgb values would change in either of the above scans. With
AE enabled, the rgb values would change drastically, +/- 50 to 100
between rgb. With AE disabled, the change is +/- 5 or so between rgb.
With the mouse pointer sampler so sensitive to movement, where should it
be positioned before adjusting the rgb slides to neutralize a gray?

Is there a yahoo group where I can post these 5400 questions?

 

[dseuser]

Try running the scan with out the Minolta profile, always found I have
better scans without their profile with a Scan Multi. Definitely use
the curves control to adjust the color.

Tom

Not sure what you meant by "scan with out the Minolta profile". Did you
mean scan non-linear, or scan linear and do not assign the profile in
Photoshop? I tried both, and found the scan way too dark with huge
shadow clipping. Correction them this way is darn near impossible.

 

[Bart van der Wolf]

SNIP

VueScan 8.1 is from more than 2 years ago.
and the D-max bug in VueScan is still there today.
Bart himself has reported it here and everywhere
over the past 2 years but the bug persists.

VueScan is just full of bugs.

As much as it pains me but on that I must say I fully agree with Don.
That's why I myself have stopped using VueScan.

Bart

 

[Bart van der Wolf]

I must apologize to all for not writing anything more constructive
recently but faking messages is just taking too much of my time.

I wish I never started but it's too late now... :-(

Maybe next time you all can immediately come down hard on anyone who
starts faking messages to nip it in the bud?

Bart

 

[Don]

And of course, this message again was posted by
me, and not by Bart. Bart doesn't post from
supernews, only I do (and I'm not Bart)

I'm quite capable of scaling this up to several
thousand responses per day, and even automating
these responses if necessary. A bully needs to be
confronted.

I actually love VueScan, and you can download
a trial version from:

http://www.hamrick.com/

You can use www.google.com to search for
VueScan - and you'll find thousands of sites that
recommand VueScan. Use the keyword
"scanner software" and you'll find that VueScan
is the number one recommendation by googe.

Don

 

[Don]

And of course, this message again was posted by
me, and not by Bart. Bart doesn't post from
supernews, only I do (and I'm not Bart)

I'm quite capable of scaling this up to several
thousand responses per day, and even automating
these responses if necessary. A bully needs to be
confronted.

I actually love VueScan, and you can download
a trial version from:

http://www.hamrick.com/

You can use www.google.com to search for
VueScan - and you'll find thousands of sites that
recommand VueScan. Use the keyword
"scanner software" and you'll find that VueScan
is the number one recommendation by google.

Don

 

[Bart van der Wolf]

From the archive, the Real Bart suggested neutralizing a film base
for
negatives. So I assume that this is a post from him.

Yes, I give technical support and supply examples and links for
further reading. The spam messages are obviously not mine since they
are not productive. Besides, I don't post via Supernews.com (just
check the message headers).

I thought that it is for negatives only, but I tried it with an
unexposed slide (all black) and noted a few very interesting points.

With slides you typically want to maximize exposure, but avoid
clipping of highlights. That optimized exposure level will get the
lowest (photon) noise levels, and the highest density range from your
film. The same goes for negative film, but you can remove the mask at
the same time.

If Auto Exposure for slides is enabled in the preference (my
normal scanning setting), the scan preview will show up pixelated
and very bright (rgb in the 2xx range). If the Auto Exposure is
disabled, the scan preview is all black (rgb in the 0 range). Now I
wonder whether Auto Exposure should be left on or off to
neutralize the grays, and for normal scanning.

You can use Autoexposure, combined with a tweak of the R/G/B gain
controls, to obtain know/desired neutrals in the resulting image.

SNIP

Is there a yahoo group where I can post these 5400 questions?

There is, but I'm not sure if it is used much.
<http://groups.yahoo.com/group/minoltadimagescanelite/> .
You can also try Photo.net e.g. at:
<http://www.photo.net/bboard/forum?topic_id=1701> .

Bart

 

[dseuser]

Bart van der Wolf wrote:

Thanks for the answers.

[snip]

You can use Autoexposure, combined with a tweak of the R/G/B gain
controls, to obtain know/desired neutrals in the resulting image.

Can you comment on how to position the extremely sensitive sampler
pointer? For example, by simply touch the mouse, the rgb reading can
change from 0,5,0 to 1,0,5. Adjusting the sliders at either location may
or may not obtain a good neutral.

There is, but I'm not sure if it is used much.
<http://groups.yahoo.com/group/minoltadimagescanelite/> .
You can also try Photo.net e.g. at:
<http://www.photo.net/bboard/forum?topic_id=1701> .

Many lost souls like myself there, but no experts. No flaming either.

 

[Bart van der Wolf]

Can you comment on how to position the extremely sensitive
sampler pointer? For example, by simply touch the mouse, the
rgb reading can change from 0,5,0 to 1,0,5. Adjusting the sliders
at either location may or may not obtain a good neutral.

When I use the Minolta Scan utility, I tend to look at the channel
histograms rather than the sampler pointer. That pointer is then only
used to do a spot check of whether the histogram was accurate enough.

Bart

 

[Bart van der Wolf]

SNIP

I'm quite capable of scaling this up to several
thousand responses per day, and even automating
these responses if necessary. A bully needs to be
confronted.

Yes, it's official! You're definitely tiring now. Just repeating
the same empty promises is a sure sign you just can't take it anymore.

Bart

B-o-o-o-o-o-o-ots, bots, bots, bots, bots... Co-o-o-o-ome, come,
come... He-e-e-e-ere you go:

(e-mail address removed)

Nice bots... Eat!

(e-mail address removed)

 

[Bart van der Wolf]

I forgot again!

For "Los Spambotos"! Vamos amigos!

(e-mail address removed)

And a few for "Le Spambot"! Voila! Chercher le Bart!

(e-mail address removed)

Not to forget my (Dutch) "Het Spambots"! Or should that be "De
Spambots"? Either way: Zeker weten!

(e-mail address removed)

And of course "Die Spamboten"! Aba sicha! (Aber sicher!)

(e-mail address removed)



Bart