Some questions to Minolta Dimage 5400 owners

[WAn]

1) Settings that result in different actual exposure.

 

 

Is it possible to set something like 'favor shadows contrast'

or 'favor highlight contrast' modes, like raising or sinking gamma in

Photoshop except that the setting results in different _physical_

exposure? In other words is there something that can be done at

scanning stage that affects scanning behavior and therefore cannot be

done later in Photoshop?

 

 

2) Ability to scan the same frame twice (or more times) leaving the

scanned area exactly the same?

 

 

Questions 1 & 2 are related: I may want to get 2 scans where the

first has maximum shadow separation, the second has maximum

highlights separation, and then combine them in Photoshop.

 

 

3) Manual focus: how does visual feedback work?

 

When I turn the MF dial -- what happens? Gets a tiny portion of the

image I'm focusing on automatically re-scanned and re-displayed?

 

 

4) Holder quality in relation to film flatness: any softness in grain

across the scanned image?

 

 

5) Black and white point setting, is it possible?

 

 

6) Interpolation ability? -- Just in case I decide (for any reasons)

to increase the resolution beyond 5400 via interpolation...

 

 

 

Thank you.

Andrey

[steven_clark]

1. Yes through Vuescan, probably through the Minolta software too.

2. I've done it and it can work, on the other hand sometimes the slide gets shifted by inertia as the track jerks into motion.

 

5. Black and white levels can be set in the minolta software (if it's anything like the SDIII software) Vuescan will only try to set them automayicly.

6. It will probably have it, use another program anyway.

[stb]

1.Yes. Vuescan does it automatically, but see below.

 

2. Same as 1.

 

3. It works very well. The dial is a bit touchy, though. That said, AF works very well.

 

4. Not good. I have seen areas of grain in grain. This clearly shows that it is an

amateur machine, not an Imacon.

 

5. Yes, not yet experimented.

 

6. No, but not useful IMHO. 5400 dpi is already almost too much. It resolves TechPan

grain...

 

I've had the scanner only 4 days and I am still struggling to find a proper B&W

workflow/settings. Ialready regret I did not get an Imacon 343, but that's off-topic.

Vuescan allows to tune the exposure, but even with max exposure setting, I have

printable highlights it cannot see trhough. I am currently re-trying with the Minolta

software.

 

A clever thing in the Minolta software: I've noticed that when you select B&W negative,

it puts the diffuser on.

[costas_lymbouris]

1) The Minolta software gives full control over the exposure; this includes overall exposure and also R,G and B enabling you to correct any obvious colour cast during exposure. This means you can set it to give a nice 16-bit colour corrected scan without clipping shadows or highlights during EXPOSURE.

 

2) Yes you can do this � I use it to scan negatives with a very wide dynamic range; once for highlight detail then again for the shadows varying the exposure in between. I then blend the images together.

 

3) I have found the auto-focus works well � you can select the point in the image on which to auto-focus. I have not had to resort to manual focus but it is there with software control and also knob

 

4) Tend to use ICE which turns on the grain dissolver so cannot answer this.

 

5) I check the white and black points with the exposure control pointer and set exposure manually to give me what I want. You can set black and white points in the Minolta software post exposure but better to get the exposure right rather than compensate in the software after the event.

 

6) 5400 dpi is the maximum that gives a 230Mbyte 16 bit file

 

You can download the user manual from the Minolta site if you want to get a feel for how the software works. The manual basic and does not tell you how to get the most out of the scanner but worth reading.

[WAn]

Gentlemen,

Thanks a lot for your answers. Very informative.

 

I forgot one more question important for color work:

 

 

7) Is the scanner able to embed the color ICM (ICC) profile into image file?

 

 

Costas, at Minolta site I see only so-called Quick Guide (http://www.dimage.minolta.com/elite5400/index.html), but no downloadable manual. Could you please tell me the address if you mentioned a manual rather than the Quick Guide?

[leif_goodwin8]

7) No. However when you open the scan in Photoshop you can apply a profile. I scan in 16 bit mode, open in PS and apply a profile, perform gamma point tweak, then convert to 8 bit and save. (My PC takes over one minute to open a 200MB 16 bit scan file.)

[costas_lymbouris]

There are 2 "owners manuals" at this link <a href="http://www.minoltaeurope.com/pe/digital/manuals/manuals_en.html">Minolta manuals</a>, one labelled Scansoftware_E.pdf It's 4.6 Mbytes so do not want to download it to see what it is, the other one is the hardware manual.

<p>

 

As Leif says, the profile is not embedded in the output file you have to assign it when you open the file for edit. The Minolta software allows you to select the output colour space and seems to do the conversion to that colour space so can only assume its a bug that the profile is not embedded. If you have time, have a look at the <a href="http://www.photo-i.co.uk/cgi-bin/ikonboard/ikonboard.pl?s=3ef9665d019bffff;act=ST;f=7;t=17">photo-i</a> thread on the 5400 for examples and some issues.

[WAn]

Costas, thank you very much for the links.

 

 

Ad 1).

Having read the photo-i comments I'm somewhat alarmed at the stage when my custom manual exposure and curves are applied: some people guess the transformation is done with 8-bit mode and only then converted to 16 bit. To be sure, let me express my question in other way:

Lets assume I'm interested only in shadow tones, i.e. I want to ignore all the toned that on the normal scan would reside, say, above Brightness 76, but I want maximum information about tones below 76. Before second attempt to scan I set the curves as shown in the attached picture.

--- Will the scanner give me the full tone image (16 bits, no posterisation)?

 

 

Ad 7)

I seem to misunderstand something at very base level about profiles.

My belief is that the profile of a device is a sort of map that translates hardware dependent RGB (or CMYK) values into invariant, hardware independent CIE colors. According to this logic each device must have ITS OWN profile. So scanner must have scanner's profile, monitor -- monitor's profile, printer -- printer's profile etc. But the manual (and people in the photo-i thread) speaks about monitor's profiles supplied with the scanning device! It is very nice if scanner manufacturer is so knowledgeable that it can supply profiles for monitors, but what about profile for the very scanner?

I'm puzzled. Can anyone clarify this point?

 

 

Ad 3)

As I understand there is no instant visual feedback like we have in SLR while focusing. A device determines deviation from 'best' focus and the deviation is displayed as a mismatch between black and white bars. I guess there is no reason to believe that I can manually match the bars better than software can, so there seems to be no point in this sort of manual focus at all. Am I wrong?

[WAn]

Oops, I forgot the attachment<div></div>

[stb]

I am now getting excellent results with VueScan in B&W. The problem was silly: by

default, VueScan select 0.40 as the negative contrast index. Once you give it the

proper one, or reasonnably close, everything is fine.

 

I am now very happy with the Minolta, but even with VueScan, I believe there would

be no contest with an Imacon.

[mendel_leisk]

Just read this old thread, don't know if anyone's still following it.

 

Stephen, D76ci=.40 is indeed at the contrasty end of things. Ed Hamrick is always counseling to leave settings on defaults for openers, but this one instance where you definitely shouldn't. Incidentally, Tmax400 with D76ci=.55 is IDENTICAL to ILFORD XP2 profile, as evidenced by the histograms. I started with XP2, then switched to TMax400 with D76ci=.55 as my new "base", to take advantage of the contrast index. Usually, lowering the number requires lowering brightness as well, and vice versa. I find TMax400 with D76ci=.55 works best for majority of cases. Tmax100 tends to blow out highlights (with neg. film) to readily. Also, there is some sort of bug in Vuescan, if you set Tmaxci's, they aren't remembered when Vuescan restarts, so it's simpler to stick with the D76ci's. They're very similar anyways.

 

Andrey, regarding your question #2, scanning twice and getting exactly the same area scanned, just different contrast or brightness, this is simple using Vuescan raw file.