LaserSoft Imaging

Hi,

I observe a behavior of HDR which I do not quite understand:

I have a computer generated tif file (based on IT-8 data) which
has a profile assigned. The gamma with which the file was generated
was 2.2 (just as side information).

One specific patch (to which I will refer later) has the following
value:
RGB: 132,130,128
Lab: 53,-4,5

When I convert the file to AdobeRGB profile (in photoshop) the patch becomes:
RGB: 138,142,132 (rendering intend: perceptual)
Lab: 59,-4,5

When I assign the AdobeRGB profile (in photoshop) the patch becomes:
RGB: 132,130,128 (no change because profile is just assigned)
Lab: 55,1,2


Now I open the file with HDR and have the following configuration in
options of HDR:
gamma: 2.2
gamma in HDR: 2.2
CMS Scanner->Intern: None
Intern: AdobeRGB

The upper patch becomes:
RGB: 132,135,126
Lab: 55,1,2

I expected this, it is like asigning a profile in photoshop, therefore
the same RGB but different Lab values.

With gamma: 1.0 and gamma in HDR: 1.0 I get the same.

I expected also this, becaus I assume as long gamma and gamma in HDR
have the same value that both have no influence.

Now I change the options to:
gamma: 2.2
gamma in HDR: 2.2
CMS Scanner->Intern: Embedded profile
Intern: AdobeRGB

I would now expect to get the same behavior as with converting to
the AdobeRGB profile in photoshop.

But I get for the patch:
RGB: 84,87,78 !!!!
Lab: 27,-5,6

This is way off.

If I change gamma and gamma in HDR to 1.0 (both) I get:
RGB: 137,141,131
Lab: 58,-4,5
which is just a little bit off the photoshop conversion (may be the different
Engine).

My question is why is the behavior with embedded profile different
between the two gamma configurations?

Dear Theo,

your problem might be related to the bugfix I just made in SilverFast.
Please send me your image file (the original one with your ICC profile embedded) by email.
I can then check this and eventually send you a bugfixed version back.

Best regards,
Martin Muenier

Here's the email address: martin.muenier@silverfast.com