Prints don't match my monitor. Am I doing something wrong?

I was just given a brand new Epson R300 printer, which I installed today. Monitor is and NEC S-IPS 2070nx panel, calibrated using an i1.

To test the printer, I loaded an image into PS7. I checked the image in a custom proof setting, using the epson icc profile for the premium gloosy paper I have. I tested the image using both perceptual and relative colormetric to determine which one looked best.

I then went into print with preview. I chose the input space as my RGB working space, and the output based on the best soft proof. I then go into the printer properties, choose best photo, the proper printer paper, click ICM and turn off color managment.

The prints I get are close to the monitor, but sometimes have a bit of a color cast, and saturation is degraded. Also I loose a lot of detail in the shadow portions of the image.

Is there something I need to change?

There is a link in my sig.
In it there is quite a bit about CM and softproofing

Sounds like you are doing it right. Maybe the printer/paper profile isn't too accurate?
When softproofing, did you tick the 'simulate paper white' box?
It looks like crap on the monitor if you watch the change (so close your eyes for a sec ), but does give a good preview of what a print will look like.

You monitor is backlit electronically, whereas prints are on paper. There's no way you could ever get the two to match, is there??

Not exactely, but you can get pretty close...

Thanks, everyone.

Rene', I had read your earlier post, and I thank you for that. Also, I was simulating paper in the soft proof. Once I edit the image in the soft proof to make it look like the RGB image and go to print with preview, should I be using the soft proof, as the input space, or the original in my RGB editing space? Is there a specific rendering intent I should use other than the perceptual/relative colormetric I was looking at in the soft proofs?

I was following the procedure outlined here:

http://people.csail.mi​t.edu …on3800/printwor​kflow.html

I would say that the last test prints I did last night were probably about 90% accurate. I just have a very slight green cast in some of the blond hair of my images. Also, very light yellow's come out far too white.

Perhaps it is just time for a printer specific profile. Which is unfortunate as it seems that Walmart prints are a bit more accurate than my inkjet.

BrianAZ wrote in post #3291327
Thanks, everyone.

Rene', I had read your earlier post, and I thank you for that. Also, I was simulating paper in the soft proof. should I be using the soft proof, in cmyk, as the input space, or the original in my RGB editing space? Is there a specific rendering intent I should use other than the perceptual/relative colormetric I was looking at in the soft proofs?

I was following the procedure outlined here:

http://people.csail.mi​t.edu …on3800/printwor​kflow.html

I would say that the last test prints I did last night were probably about 90% accurate. I just have a very slight green cast in some of the blond hair of my images. Also, very light yellow's come out far too white.

Perhaps it is just time for a printer specific profile. Which is unfortunate as it seems that Walmart prints are a bit more accurate than my inkjet.

Are you using geniune Epson ink? Maybe go to Epson's site and download the latest profiles for your Printer/paper/ink combo. Probably a mute poitn but I used to use Epson paper as I like it better than HP, but now I find that I get a color cast (orange in my case) with the Epson and not with the HP. Granted I do not have a profile for the Epson paper with my HP ink so I cant say that the paper is the reason.

Epson paper, and epson ink. All are fresh from the package. Most recent driver loaded from Epson.

BrianAZ wrote in post #3291430
Epson paper, and epson ink. All are fresh from the package. Most recent driver loaded from Epson.

Then I have no idea. Maybe try some Ilford paper???

BrianAZ wrote in post #3291327
Thanks, everyone.

Rene', I had read your earlier post, and I thank you for that. Also, I was simulating paper in the soft proof. Once I edit the image in the soft proof to make it look like the RGB image and go to print with preview, should I be using the soft proof, as the input space, or the original in my RGB editing space? Is there a specific rendering intent I should use other than the perceptual/relative colormetric I was looking at in the soft proofs?

Use the original working space as input space. The soft proof is just that: A *soft* proof (Not "hard" since the working space wasn't actually changed, only simulated...)
Use the same rendering intent as in the soft proof.

BrianAZ wrote in post #3291327
I was following the procedure outlined here:

http://people.csail.mi​t.edu …on3800/printwor​kflow.html

Very nice read, and a good explanation...
Added the link to the sticky

BrianAZ wrote in post #3291327
I would say that the last test prints I did last night were probably about 90% accurate. I just have a very slight green cast in some of the blond hair of my images. Also, very light yellow's come out far too white.

Perhaps it is just time for a printer specific profile. Which is unfortunate as it seems that Walmart prints are a bit more accurate than my inkjet.

Might be... Don't know how much tolerance there is between the different batches of paper, and don't know how accurate the profile is...

You could use a (Q&D) workaround (not the best way, but it might work if the cast is consistent): Print an image, then use an adjustment layer to get the softproof to look like the print on paper.
Then make an 'inverse' correction, to get the softproof like it was. Use that on any image after soft proofing, before printing, as a correction on the profile...

Well, after a bit more testing, I'm thinking it is an issue with the printer icc profile. I went out and bought some Ilford paper and downloaded their icc profiles.

A test print using perceptual shows improvement over the Epson papers. Using relative colormetric is even better still. As a control, I printed the same image at Walmart after converting it to SRGB. The Walmart print is almost an exact rendering of what is on my monitor. The Epson print is just a tiny bit off in color, but unfortunately doesn't have the detail of the Walmart print either. When looking at a portrait on the walmart print, I can easily see the individual strands of hair. On the epson print, I can do the same in the midtone and upper range, but loose that ability in the darker midtones. Probably due to the difference of dye sub, versus inkjet.

BrianAZ wrote in post #3294328
Well, after a bit more testing, I'm thinking it is an issue with the printer icc profile. I went out and bought some Ilford paper and downloaded their icc profiles.

A test print using perceptual shows improvement over the Epson papers. Using relative colormetric is even better still. As a control, I printed the same image at Walmart after converting it to SRGB. The Walmart print is almost an exact rendering of what is on my monitor. The Epson print is just a tiny bit off in color, but unfortunately doesn't have the detail of the Walmart print either. When looking at a portrait on the walmart print, I can easily see the individual strands of hair. On the epson print, I can do the same in the midtone and upper range, but loose that ability in the darker midtones. Probably due to the difference of dye sub, versus inkjet.

I found the same thing (detail) when printing at home versus Costco. So I only print smaller prints, snapshots for the family, and scrapbook stuff for my wife at home. For the nicer, wall hanging type stuff I send it out.