PDA

View Full Version : 2 Epson Printers, inconsistent B&W output question



Paddy McKay
27-Nov-2008, 21:00
Hello folks. I have a frustrating issue with two current gen. Epson printers. At home I have a 2880, and at work, a 4880. From Lightroom, I'm printing B&W on the 2880, using Moab CFG, ICC profile, colour settings, I get perfect and predictable neutral results. Very slight tweaking involved due to high brightness of (calibrated) LCD screen.

The problem comes with trying to get the same results from the 4880. Using all the same settings, gives me a slightly purplish/magenta print. To get even close to the 2880's results, I've had to resort to using Adv. b&w settings (warm/darker), and then fiddling even more with the exposure; the tonal relationships/ exposure are all different though.

Does anyone have clue what's going on here?

Clueless Winddancing
27-Nov-2008, 23:42
That's always the problem wnen one has two wives. Making comparisons is a sickness of the mind's eye.

Ken Lee
28-Nov-2008, 03:51
It's a classic.

For consistent results, you need a separate profile for every printer/paper/ink. I use CHROMIX (http://www2.chromix.com/index.cxsa).

Brian Ellis
28-Nov-2008, 08:32
The Moab (now Legion) web site has profiles for every photo paper they make. I don't know what "CFG" is but I've used Moab/Legion Fine Art Natural and Fine Art Bright White for years with the profiles I downloaded for my Epson printers (first a 2200, now a 3800) from the Moab/Legion site. If you use those profiles for each of your printers and do everything else right you should get good, consistent results from both printers.

Ken Lee
28-Nov-2008, 08:58
Logic suggests that either you did something wrong - or the profiles are inconsistent - or your printer/paper/ink isn't exactly the same as the printer/paper/ink that they profiled.

My guess is that your printer/paper/ink is slightly different - enough for you to see the difference. That's why you may want to get your own profiles made, for your own equipment.

I'm no expert, but if digital computer monitors vary and drift over time, why wouldn't printers ? Being mechanical rather than electronic, wouldn't we consider printers even more prone to variation and drift, than monitors.

I have not worked in the press side of the printing industry, but I have seen enough to know that every time they do a run, they make sure that everything is just right. Compared to the cost of ink and paper, the price of a custom profile is reasonable.

Jim Becia
28-Nov-2008, 10:04
Hello folks. I have a frustrating issue with two current gen. Epson printers. At home I have a 2880, and at work, a 4880. From Lightroom, I'm printing B&W on the 2880, using Moab CFG, ICC profile, colour settings, I get perfect and predictable neutral results. Very slight tweaking involved due to high brightness of (calibrated) LCD screen.

The problem comes with trying to get the same results from the 4880. Using all the same settings, gives me a slightly purplish/magenta print. To get even close to the 2880's results, I've had to resort to using Adv. b&w settings (warm/darker), and then fiddling even more with the exposure; the tonal relationships/ exposure are all different though.

Does anyone have clue what's going on here?

Paddy,

It is my understanding that the 2XXX and the 4xxx lines are not made to the same tolerances. I had a 2200 and a 9600. They both used the same Ultrachrome inks, and I printed the same image using both of them from the same computer, same settings, etc. While the results were close, they were never exact. I just think it's a matter of the 2880 printer not being made to the same tolerances. Jim

Tyler Boley
28-Nov-2008, 11:31
hitting exact neutrals with the RGB driver is very difficult for profiling software, some even have extend gray patches just for that very purpose, since it's an unusual need for most color printers. Even if they do look pretty good, consistency from one printer to the other, even with the same profiling software, may be less than exact. Additionally, the slightest hue deviation in neutrals is very obvious to the eye which excaserbates the problem. It's a tough one for that particular workflow, there may be no really great solution. Learning an ABW workflow may be your next best step.
Tyler

Paddy McKay
28-Nov-2008, 11:31
Just to clarify, the 2880 is producing extremely consistent results, with appropriate ICC profiles. Brian, CFG = Colarado Fiber Gloss. I've been using the unique (for each printer) profiles for each printer, provided by Legion/Moab. I'm going to check with the Tech guy at the college where the 4880 is, since it's not providing the expected results. I suspect that I'm going to have to get a custom profile(s) created for the 4880. I've never used their services, but I 've read about Cathy's profiles http://www.cathysprofiles.com/, and MadManChan's http://people.csail.mit.edu/ericchan/photos/profiles.html

If anyone has info for other highly recommended profiling services, I'd be happy to know.

Thanks again, Paddy

Paddy McKay
28-Nov-2008, 14:54
hitting exact neutrals with the RGB driver is very difficult for profiling software,... there may be no really great solution.

Tyler, it wasn't a problem for the 2880 to hit the (profiled) mark, spot on. The issue with the 4880 is more than just a colour shift; the prints come out with a very different curve/exposure. It's at least 1 1/2 - 2 stops brighter.

I guess I'm not sure if a custom profile will address the exposure difference. If anyone knows, I'd appreciate hearing back from you. I'd rather not waste the money on a custom profile, unless it corrects this problem too.

Tyler Boley
28-Nov-2008, 16:16
recent changes in Adobe, MacOS, and printer drivers has made the current state of profiling less than confident. After recent Photoshop and OSX updates I had to go back to the very original driver for my little old Canon desktop to get it to keep the color managed data path the same as before and prints to look anything like the color before the changes even with the exact same profiles and driver settings. So it could be any number of things, driver versions of the 2 models, how each profile was made, etc etc.
You can avoid blowing your money by finding a custom profiler that gives a money back garauntee, some do.
I occasionally profile for others, and am less than confident in the outcome these days. Way back in OS9, and no color management in the drivers, things were more secure...
Wish I could be more help, it's not a good situation. Hopefully there's something simple in the workflow, a gotcha somewhere to solve. Happen to all of us.
Tyler

Stephen Best
28-Nov-2008, 18:10
Tyler, it wasn't a problem for the 2880 to hit the (profiled) mark, spot on. The issue with the 4880 is more than just a colour shift; the prints come out with a very different curve/exposure. It's at least 1 1/2 - 2 stops brighter.

Are you using CS4 and ABW for the 4880? If so, there's your problem. Adobe has been dicking around with the printing interface in CS4 and it's best avoided until they sort it all out. Maybe in CS5!

Paddy McKay
28-Nov-2008, 22:39
Are you using CS4 and ABW for the 4880?

No, I'm using Lightroom 2.1, and Color settings in the Epson dialog box, no color adjustment ( i.e. designated Moab profile from Lightroom Print module color management)

Stephen Best
29-Nov-2008, 01:01
No, I'm using Lightroom 2.1, and Color settings in the Epson dialog box, no color adjustment ( i.e. designated Moab profile from Lightroom Print module color management)

This problem (see posts by madmanchan)?

http://luminous-landscape.com/forum/index.php?showtopic=29496&st=0

Paddy McKay
29-Nov-2008, 07:57
Stephen, thanks for the link. I'm going to try the workaround of selecting Adobe RGB 98, under LR 2's Print module Color Management Profile setting, then try Epson's ABW. We'll see. I'll try this later today, and report back. Of course, I expect that there'll still be a fair amount of tweaking since this isn't a profile for the paper.

As they say, "welcome to the suck".

Paddy McKay
29-Nov-2008, 16:27
UPDATE: I went to use the 4880 this morning, only to find the printer damaged, and not wanting to accept paper from the bottom tray. It looks like students from the full-time day program have been using the printer all week, and the sliding side paper guide, inside the lower paper tray, has been busted off! Additionally, the software was not in order either, as it kept telling me to load the paper againn, etc. I tried top and bottom loading, all to no avail.

Arghhhh.

Brian Ellis
30-Nov-2008, 11:16
UPDATE: I went to use the 4880 this morning, only to find the printer damaged, and not wanting to accept paper from the bottom tray. It looks like students from the full-time day program have been using the printer all week, and the sliding side paper guide, inside the lower paper tray, has been busted off! Additionally, the software was not in order either, as it kept telling me to load the paper againn, etc. I tried top and bottom loading, all to no avail.

Arghhhh.

If the 4880 is used by students there's no telling what kind of problems they've created even before your latest experience. My students used to develop film in fix, refuse to ever read any instruction manual that came with their cameras, and if they didn't know why something didn't work they'd just beat on it until they completely destroyed it. Students and photography equipment are a combination that was never meant to go together.

Bruce Watson
30-Nov-2008, 14:58
Tyler, it wasn't a problem for the 2880 to hit the (profiled) mark, spot on. The issue with the 4880 is more than just a colour shift; the prints come out with a very different curve/exposure. It's at least 1 1/2 - 2 stops brighter.

Tyler is telling you the truth. Printing B&W with color inks, especially if using the default driver, is fraught with difficulty. And color casts. Because you aren't using the right tools for the job. If the 2880 is working perfectly for you, it's because you got lucky.


I guess I'm not sure if a custom profile will address the exposure difference. If anyone knows, I'd appreciate hearing back from you. I'd rather not waste the money on a custom profile, unless it corrects this problem too.

No one is "sure." Getting the color right for a given machine/ink/paper is what an ICC profile is supposed to do. But that doesn't mean it's supposed to do B&W right. Because that's not what an ICC profile is designed to do. An ICC profile is more about the edges of the gamut than it is about the neutral axis.

To get B&W inkjet printing right, you need either a driver that's designed for the duty (think QTR (http://www.quadtonerip.com/html/QTRoverview.html)), an inkset designed for the duty (think Cone (http://shopping.netsuite.com/s.nl;jsessionid=0a010b441f437894fef00a054fce8a67e379e53cbf22.e3iKaNePch4Re34Pa38Ta38Nbh50?c=362672&it=I&id=119) or MIS (http://www.inksupply.com/qn.cfm)), or both.

Ken Lee
30-Nov-2008, 15:17
No one is "sure."

What Bruce said... as usual !

According to WiseGeek (http://www.wisegeek.com/what-is-metamerism.htm): "What is Metamerism ? When two color samples appear to match under a particular light source, and then do not match under a different light source".

One of the several reasons that I make warm-tone "monochrome (http://www.kenleegallery.com/bronze.html)" images - instead of "black and white" - is that the eye can easily tell when grey is not grey, but will adapt to a monochrome image, and slight variations due to metamerism, with little if any protest.

Another reason, is that I don't want to have dedicated printers, drivers, etc. By the time you get things calibrated, your printer is obsolete. I prefer to use good-old RBG inks, and freely print color and monochrome on the same machine. One printer is already enough to worry about.