Digital BW, The Print

From: "jimhayes361" <jimhayes@...>
To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
Sent: Wednesday, May 22, 2002 1:25 PM
Subject: [Digital BW] Tylers new VM curve experiences anyone?


> I'm interested in two main areas for the 1280 pc for the new curves
> (4/22/02). BTW, I'm using Eclipse Satine 190 gms:

Jim,

The curves were specifically designed for Photo Rag so you may need to do
some adjusting for Eclipse Satine. The curves do work well as is on EAM and
most of the Hahnemulhe papers.
>
> 1) About how does each curve print wrt each other. Especially which
> curves posterise least, and how would you rank them from warmest to
> coldest. I asked this question before, and got a response or two. Now
> there are new curves.

The naming convention is cld## = Cool or cool, crv## = neutral, mw## =
medium warm, plat## = platinum and Wm## = warm.

I think that the smoothest are the crv09, mw03 and Wm05 with the others not
too far off. Depends upon the particular image.
>
> 2) I assume the RGB color space to convert to is Adobe RGB. However,
> from reading past posts on Paul's curves, I'm not sure there was ever
> a resolution on whether the original greyscale file should have it's
> profile dropped upon opening, or converted to greyscale gamma 2.2
> (the pshop 5 default).

The final conclusion on that one was that sRGB and Adobe gave the same
result since they have the same gamma, 2.2. The problems were in switching
between color spaces with different gammas. Since you need to wind up with a
gamma of 2.2 in RGB mode, you probably should start there in grayscale.
>
> I recently changed my default monitor greyscale profile from a custom
> dotgain curve to 20%, when I discovered that my new monitor was
> matching the print better at the simpler 20%. However, with Paul's
> curves, when dropping the 20% dot gain profile, then converting to
> sRGB from a default gamma 2.2 greyscale profile as defined by the
> pshop 5 default space, I now seem to get some posterization. Any
> ideas?

I can only suggest that the RGB separation curves are often rather radical.
If you are working in 8-bit and the image is marginal, applying a separation
curve can push it over the edge into posterization. I try to do as much as I
can in 16-bit mode or sometime I have to do the whole thing in 16-bit. Once
again it depends upon the image.

Martin Wesley