Digital BW, The Print

Jim,

I just wanted to point out one difference in our file conversion process.
You converted to sRGB in both PS6 and PS7 while I converted to Adobe RGB in
both. I don't know if this is significant but a variable to keep in mind.

I hope it is just a workflow adjustment issue and not something mechanical.

Martin Wesley
http://www.borderless-photos.de/guests.html



----- Original Message -----
From: "jimhayes361" <jimhayes@...>
To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
Sent: Wednesday, June 26, 2002 9:13 AM
Subject: [Digital BW] Re: pshop 6->7 VM (isolating just 6 to 7 discrepancy)


> Martin, Todd, Paul, everyone,
>
> You all bring up extremely good points. I am going to limit myself to
> two observations and one experiment. I'm doing this because something
> CHANGED on my system and I need to isolate it. This change in workflow
> or settings turned a great print done in Febuary, which I have in a
> light tight box as reference, to a trashy posterized print done two
> days ago. However imperfect or perfect my workflow/ print file/combing
> was in Febuauary, it got worse from the same exact file and same
> settings in June.
>
> I use an 1280 with standard VM inkset, which I (perhaps
> uneconomically) use virgin vacuum filled carts with. I bought my first
> six bottles of ink in January, my second arrived April 9th. So one
> change that happened was I started filling from a new set of bottles.
> I swab tested them to be sure they were not mislabeled and I double
> checked labels as I filled carts.
>
> The second change was that I loaded Photoshop 7 on my system (Win98SE)
> on April 22.
>
> The things that changed are: time, inkset bottles (same batch?), and
> pshop 6---->7.
>
> If the ink gooped up my printer over "time", I may see it get better
> in both ps6 and 7 after running Epson carts in it for a couple of
> days. I suspect this is not the case. Since no CIS is involved we can
> rule out all the little problems it can cause.
>
> If the two sets of inksets differ, MIS would have long heard of it by
> now from many other folks. I think this can be ruled out, escpecially
> since they were bought only 3 months apart.
>
> If when I put a new set of carts in after the OEM ink run, I can, from
> the same file, duplicate the good print using PS 6 and duplicate the
> bad print using PS7, then the variable that messed up the print is the
> conversion algorithim in PS6 vs 7. If the result from this experiment
> is positive, it locks out the other two variables and a lot of other
> things as the culprit. And someone then has to discover what changed
> in PS7.
>
> Oh well. All speculation until I run those two prints- maybe Thursday,
> Friday.
>
> The second note is that I tried Culbertson's technique, suggested by
> Martin, to convert from greyscale to RGB and then apply one of Paul's
> curves in both PS6 and PS7. What's very interesting is that using
> Culbertson's method, PS6 and PS7 give identical mean, SD, and median
> values for my test print using the VM workflow, down to applying
> Paul's curve. BUT, the values in the histogram differ from both PS 6
> and PS7 when the regular greyscale gamma 2.2 convert to sRGB is used
> (Paul's workflow). And of course as already stated way back in a
> previous post, the values differ between PS6 and PS7 using the gamma
> 2.2 greyscale--->sRGB.
>
>  So one gets three slightly different files going to the printer
> depending on whether you use grey gamma 2.2 to sRGB PS 6, grey gamma
> 2.2 to sRGB PS7, or Culbertson's channel conversion to RGB method
> (PS6/PS7 independent).
>
> What is puzzling to me is Martin not getting quite as large a
> difference as I did when he tried the gamma conversion in both ver 6
> and ver 7. It could still be that I'm overlooking some setting, but I
> really can't think of what it is. Maybe it just depends on which file
> one uses(?), not just how dark or light it is?
> Jim H.
>
(snip earlier)