Scanner Calibration with IT8 Targets

[bob_krueger]

I have searched the archives and failed to find anything that

specifically addresses this question.

 

I have a Canon FS4000US film scanner that I use to scan slides

into Photoshop 6 on a Macintosh running OS 9.1. Using FilmGet,

the included software, the scans are sometimes alright, but

usually not great, as regards color accuracy compared to the

slide. I shoot Velvia a lot, and they're especially bad with that film.

The saturation is good, but color accuracy is not. So I purchased

the professional version of VueScan a week or so ago and have

been playing with it as time permits. The scans with VueScan

are better as regards color accuracy no matter what settings I

use, but have other issues, like inadequate saturation, at

settings that others recommend here at Photo.net. My results

have been worst with the color balance setting of "neutral" that

many others seem to use successfully. "White balance" was

better, but the most accurate scan I've done so far was with the

color balance set to "manual," white and black point settings of

1%, and other color settings left alone. The media type is set to

"image"; I saw no meaningful improvement using "slide film"

and the complementary E6 color setting. I sure wish Ed Hamrick

had more built-in Fuji profiles. To tell the truth, I'm a long way

from knowing what I'm doing with this software at this point, but I

like the fact that it allows me to choose from many variables, that

it works with both of my scanners, and that it will allow me to

calibrate my equipment.

 

The next thing I'm considering doing in order to gain more

control over my system is purchase an IT8 target slide from Wolf

Faust. Since I shoot mostly Velvia (the new 100F much of the

time), if I were buying one slide, the Velvia one would be it,

although I assume it is based on Velvia 50, which I also shoot

sometimes. I also shoot a little Provia, though not much, and

Sensia for motorsports.

 

My main question is - do I want to purchase more than one IT8

target slide, since Wolf Faust also has one for

Provia/Astia/Sensia, and why would I want/need to do that? If the

slide is simply matching the color numbers that are known to be

on the slide with the color numbers that my scanner sees, and

then writing a file to translate one into the other, are not all of the

colors on all of the slides part of the same general color gamut?

And if they are, if I match the colors on one slide, would not the

colors on the other slides be accurate post-calibration as well? If

I need multiple scanner calibration files for multiple film types,

then I guess I would have to reference another file every time I

scan a different film type with VueScan. Is that true? I apologize

for the fact that I'm confused.

 

By the way, yes, my monitor (LaCie 19" Electron Blue) has been

calibrated with a GretagMacbeth Eye One Display colorimeter

and I use the Tim Grey profiles for luster and archival matte

papers for my Epson 2200...and I do use those papers and

Epson inks. So if I could just get accurate scans into the

computer in the first place I could hopefully minimize the amount

of color correction I have to do in Photoshop. I'd very much like to

do that, since I'm not exactly the king of the mountain yet when it

comes to that either. Despite the fact that I have read "Real World

Photoshop" and I'm working on getting better at being subtle, you

should see what I can do to a scan that requires heavy-handed

correction; it ain't pretty!

[qtluong]

The gamuts of two film types are not exactly the same, though largely

overlapping.

On the the page

<a href = "http://www.computer-darkroom.com/it8cal/it8_page_1.htm">

http://www.computer-darkroom.com/it8cal/it8_page_1.htm</a>, which

discusses the variations between different IT targets,

the second to last paragraph says that Wolf Faust thinks

that matching film to target is important. Since you're going to

purchase a target from him, maybe he would be the best to ask this

question to ? I suppose he wouldn't make a variety of targets if

he thought there was no good reason, and the test by Ian Lyons does

show indeed variability.

 

<a href = "http://www.terragalleria.com">Terra Galleria stock photography</a>

[afx]

Get both!

You will see a difference.

I just forgot to switch to velvia and scanned with my provia profile.

The colors are different.

 

Setting vuescan to neutral makes only senese if you have the profiles for the scanner.

 

Ever since I bought the calibration slides, I stopped fiddeling with the color tab in vuescan, the only thing I set is cropping and IR cleaning or not.

 

Any other adjustments are done in PWP later on.

 

cheers

afx

[bob_krueger]

Thank you both very much for your answers. I am going to go ahead and order both the Velvia and the Provia/Astia/Sensia targets from Wolf Faust. Now all I have to do is figure out exactly what to do with the reference files that come in only a choice of MS-DOS or Amiga (does someone actually still have one of those?) formats on floppies. It appears that I can download them from his site once I figure out which of the two Velvia ones to retrieve. Modern Macs don't even HAVE floppy drives. Once I get them, I assume they will go into the VueScan directory somewhere, but I don't claim to know that for sure. So many things to learn; so little time.

 

By the way, Q.T., I visited your site. VERY nice. You surely must do a LOT of traveling.

[leif_goodwin8]

Bob: Basically each type of emulsion has different dyes whose spectral responses peaks at different frequencies, and have different distributions. For example the green dye in Velvia will not peak at the same frequency as the green dye in Provia. The scanner scans the image at three colours - i.e. red, green and blue - to create a colour scan. (I think they simply have three linear CCD arrays. One has a green filter in front, one has a red filter in front, and one has a green filter in front.) The peak transmission of each filter will not exactly match that of the film's dyes. What's more the mis-match varies from film to film. So the result is that each film has its own colour response. The scan can be corrected by application of a profile generated using a program such as Vuescan. Some films such as Fuji Sensia and Fuji Provia are so similar that they can share the same profile. Other films such as Fuji Velvia need their own profile.

 

BTW You can scan with the scanner manufacturers software, and then apply the profile when you open the scan in Photoshop. Or you can apply the profile at the scan stage using a program such as Vuescan. The letter should generate better results, though whether or not it does is another matter.

 

I think you could email Wolf Faust with the names of the slide films you use and he should be happy to tell you which IT8 calibration slides you need.

[royalbox]

I emailed him twice asking a similar question about choosing the right 35mm target and he never got back to me. How did you get on? Did you ever buy them and what results did you get?