PDN Calibration Confusion (3800)

[Ian Leake]

I'm hoping this isn't a dumb question (if it is then please be gentle with me )...

I'm just starting using PDN and I've hit a strange problem with the calibration palettes. If I print the 'Color Density Range Palette' on my 3800 everything works well, and the best highlight patch is G=255 + B=90 which in Photoshop has an RGB of 0,255,90 as you'd expect. When I take that value and plug it into the 'Tonal Palette' I get a nice greeny/blue palette with RGB ranging from 255,255,255 (step 1) to 0,255,90 (step 101), again as you'd expect.

As I understand the theory, the two patches (step 101 and the selected patch from the CDR Palette) should be identical on the printed negative and on the print. But they're not. Even if I drag the entire Tonal Palette into the CDR Palette they print differently (even though the RGB values in Photoshop are the same)!

I've used an RGB densitometer to measure the two patches when printed from the same file onto the same sheet of film and left to cure for about 24 hours:

CDR Palette Patch = 2.25, 0.71, 0.99
Tonal Palette Step 101 = 2.01, 0.66, 0.96

I can only assume that there's something in the way the file is processed for printing that is causing this. I haven't calibrated the printer yet, but I can't believe that this would cause such a significant variation on a single sheet of film.

Can anyone help me understand what's going on?

My Photoshop CS3 settings are:

• Photoshop Manages Colors
• Printer Profile = sRGB
• Rendering = Relative Colormetric
• Black Point Compensation = checked

My EPSON 3800 is set to do no colour adjustment and I'm using Agfa CopyJet as my film.

Thanks in advance for your ideas.

Ian.

 

[mkochsch]

I would say off the top of my head maybe the two images have different profiles. Or, possibly the method of screening the colour was different. Do it again. Post some scans and samples.
~m

 

[Ian Leake]

That was my first thought too (after the obvious one that maybe I'd forgotten to use the same colour management settings in the print dialogue boxes). But I made the densitometer readings when both palettes were printed at the same time from a single PSD file onto a single sheet of film. That's what's so weird.

I've made several of these two palettes and they've all had the same flaw. So my assumption is that I must be doing something wrong. I'd post the PSD file but my guess is that'd breach licence terms. I suppose I'll have to try again :-(

 

[mkochsch]

Breach "license" terms? Eh?! (American translation: Huh!) Of what? I would think YOU are protected under copyright laws because any and all "original work" (step wedges, images or otherwise) belong to you.
~m

p.s. Have you tried "flattening" the layers of the test image. Have you measured the colour patches using the "Info" palette in photoshop?

 

[Donsta]

Are both source files in Adobe RGB - sounds to me like one may be sRGB and the other is Adobe RGB - for PDN calibration, both should be in Adobe RGB. Go to each and do "edit", "assign profile" to check what each is.

 

[donbga]

Are both source files in Adobe RGB - sounds to me like one may be sRGB and the other is Adobe RGB - for PDN calibration, both should be in Adobe RGB. Go to each and do "edit", "assign profile" to check what each is.

After thinking about it for a while I agree with Don after noticing you mentioned sRGB.

Don Bryant

 

[Kerik]

Maybe ask your question at the PDN forum so you don't have to worry about being sued or arrested.

 

[clay]

You may be falling victim to the achilles heel of any approach that relies on the Epson driver to translate the colors in your file to a combination of ink dots on your printing substrate.

I tried the PDN approach when I got my Epson 7800, and it just didn't work. The epson driver gave me all sorts of weird UV tranmission differences between adjacent steps on the tablets, even though visually they looked continuous. I gave up and used the QTR approach and haven't looked back. I think IJC/OPM will also allow you to have the same kind of control over how the ink is deposited on the substrate.

As mentioned above, using another approach means I don't have any license restrictions, and I can send my curves to whomever I want, not to mention printing negatives for other people for fun and profit.