Ad
Ad
Ad
Pages: « 1 2 [3] 4 »   Bottom of Page
Print
Author Topic: Epson 4000 color problem  (Read 28147 times)
francois
Sr. Member
****
Offline Offline

Posts: 6870


« Reply #40 on: February 27, 2007, 03:11:26 PM »
ReplyReply

Quote
It either needs to be the default printer to remove the banding OR you need to associate a different profile (not Epson Standard) in the ColorSync utility.

My question is, if you do this from the utility and go back to the Epson Standard, does the banding disappear even if you are not using Default Printer?

The Default bit is going into Print Utility and ensuring the printer you are using is bold (set to default). But I'm trying to determine if you do the switch-a-row with the profiles, you can print without being the default printer (and get no banding).
[a href=\"index.php?act=findpost&pid=103569\"][{POST_SNAPBACK}][/a]
I'll try that tomorrow morning and report.
Logged

Francois
eronald
Sr. Member
****
Offline Offline

Posts: 4117



« Reply #41 on: February 27, 2007, 03:39:01 PM »
ReplyReply

It's interesting, I was having the "greencast" problem while testing an iSis, printing on a Macbook Pro 10.4.9 Intel, drove me crazy, and traced it to the printer drivers, and it suddenly went away after I reset the printing system and set up the (lonely) Epson as default ? Maybe this should be a precautionary measure:

1. Reset printing system
2. Reinstall drivers
3 Set your printer as default, even if you have just one.

What one does with several printers I prefer not to know. Luckily I own several computers.

Edmund
« Last Edit: February 27, 2007, 03:39:46 PM by eronald » Logged
Roy
Full Member
***
Offline Offline

Posts: 196


WWW
« Reply #42 on: February 27, 2007, 03:51:02 PM »
ReplyReply

Quote
On another machine, I tried the trick of associating the Generic RGB profile in the Colorsync utility, made a print (no banding), then reapplied the original Epson Standard profile and kept the 4800 from being the Default printer. No banding. Anyone else?
[a href=\"index.php?act=findpost&pid=103543\"][{POST_SNAPBACK}][/a]

What would be the reason to make the Epson Standard the default again?

I never print using the printer driver to handle colour. Since discovering the Generic RGB as default fix, I have left it that way and never had another problem. No worry that way about what is selected as default.
Logged

Roy
digitaldog
Sr. Member
****
Offline Offline

Posts: 9186



WWW
« Reply #43 on: February 27, 2007, 04:50:25 PM »
ReplyReply

Quote
What would be the reason to make the Epson Standard the default again?

I never print using the printer driver to handle colour. Since discovering the Generic RGB as default fix, I have left it that way and never had another problem. No worry that way about what is selected as default.
[a href=\"index.php?act=findpost&pid=103583\"][{POST_SNAPBACK}][/a]

The reason is, that profile plays some kind of role in the data stream. I substituted it with another profile and the color was different. So I'd prefer to have the factory spec settings but not have to necessarily deal with this Default Printer being set in Print Utility.

And it doesn't matter how you setup the application or driver, the profile found here is affecting the data which is in itself odd and something I wish Apple would clarify.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Ed Foster, Jr.
Full Member
***
Offline Offline

Posts: 219


WWW
« Reply #44 on: February 27, 2007, 09:27:09 PM »
ReplyReply

Quote
On another machine, I tried the trick of associating the Generic RGB profile in the Colorsync utility, made a print (no banding), then reapplied the original Epson Standard profile and kept the 4800 from being the Default printer. No banding. Anyone else?
[a href=\"index.php?act=findpost&pid=103543\"][{POST_SNAPBACK}][/a]
Andrew,
OK, I tried this and got the same results.  I also set another printer as the default, and reset the default profile in the CS Utility for the 7800 to "Factory Profile".  When printing from CS, I selected the 7800 printer in lieu of the default, and noted in the Summary Pane, the colorsync profile was listed not as set in the CS Utility (factory profile), but to Generic.  The resulting print was without problems.

This seems in line with your comments in early February regarding a 3800 - that after you changed to Generic RGB in the CS Utility, there was no banding and that you could not replicate the banding. It is also consistent with your comments above on the 4800.

These experiences seem to be at odds with Epson's comments about NCA and the default printer.

For me, the problem first appeared after the installation of the driver for the 7800.  It seems that way for you too with the 3800 in early February and the recent 4800 - is that correct?  

Once I reset the profile in the CS Utility to "Other" - Generic RGB - the first time, the driver continues to work normally (in NCA mode - default or not) and switches (as noted in the driver summary pane) to a profile that matches the tagging of the print stream (thus causing no CS conversion).  In LR it is the selected custom printer profile, in InDesign and PhotoShop it is Generic.


Ed
Logged

Ed Foster, Jr.
www.edfoster.net
digitaldog
Sr. Member
****
Offline Offline

Posts: 9186



WWW
« Reply #45 on: February 28, 2007, 08:21:57 AM »
ReplyReply

Quote
For me, the problem first appeared after the installation of the driver for the 7800.  It seems that way for you too with the 3800 in early February and the recent 4800 - is that correct? 
[a href=\"index.php?act=findpost&pid=103643\"][{POST_SNAPBACK}][/a]

I think so but there's some hesitation (and a lot of wasted ink and paper) as I'm just not sure what causes this mess and if there's a long term fix (other than doing the default trick which does appear to work).

I called my bud Mac Holbert at Nash Editions. He's having the same problems and noticed them months back. He's be using the Default trick (in Printer Utility) and says the problems are gone.

As to the old switch-a-row with the profiles in ColorSync, I'd like to test this on a few other Macs and also see if the connection (USB versus Network) has any role (I don't think so).

And as you point out, installing a new Epson driver could be an issue. The 4800 is new on this one Mac and since I'm waiting on a network card, its USB. The other Mac's are clean (didn't get install the Epson drivers). Then there's the fact that one machine is an Intel and the other PPC (I don't think that's a factor).

Now here's another test. Substitute the 3800 Standard profile in ColorSync with the 4800 (or vise versa).

I'm almost ready to just use the Default Printer Utility trick because all the variations here add up to a lot of wasted ink and paper! And you never know, Leopard may ship and the entire issue may go away (or get worse).

Bottom line is, if we know users have more than one printer and are using an Epson (some, all?) on a Mac, they should be aware of the Printer Default trick or they may end up with issues on some images with subtle reds of certain hues. And making or using custom profiles? Ouch.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
eronald
Sr. Member
****
Offline Offline

Posts: 4117



« Reply #46 on: February 28, 2007, 08:27:00 AM »
ReplyReply

Although this may be a different issue, my Epson 2400 went into greencast somewhere in the middle of the iSis test. It may have been when I switched to "generic rgb profile" with the colorsync utility or later, I don't know because at that point I was trying to trace an issue which was in fact related to transport damage to the spectro itself.

 The cast is very slight it is only immediately apparent in the neutrals except I guess if you have exceptional vision. Reinitialization of the print system and setting the device as default fixed the cast.

I have repeatedly had "green casts" appear on my own printers over the years. I supect it's some flaw in the print logic. The problem is one doesn't immediately notice it, so a bunch of prints are "not so good" .

Edmund
« Last Edit: February 28, 2007, 08:28:35 AM by eronald » Logged
francois
Sr. Member
****
Offline Offline

Posts: 6870


« Reply #47 on: February 28, 2007, 08:29:31 AM »
ReplyReply

Quote
It's interesting, I was having the "greencast" problem while testing an iSis, printing on a Macbook Pro 10.4.9 Intel, drove me crazy, and traced it to the printer drivers, and it suddenly went away after I reset the printing system and set up the (lonely) Epson as default ? Maybe this should be a precautionary measure:

1. Reset printing system
2. Reinstall drivers
3 Set your printer as default, even if you have just one.

What one does with several printers I prefer not to know. Luckily I own several computers.

Edmund
[a href=\"index.php?act=findpost&pid=103581\"][{POST_SNAPBACK}][/a]
Edmund,
Your suggestions are the first steps when trying to solve weird color issues but we've already done that without any success. I tried with different Macs and all with freshly installed systems. The issue is not as visible as a green cast, it only hapens on some red/salmon gradients. Andrew's test photo (skin tones and the adobe wall) available on his website is excellent to spot this issue. I haven't seen the issue with HP printers - yet -. There's some stange interaction between  the Epson drivers, ColorSync and perhaps Adobe applications with the exception of Lightroom.
Logged

Francois
digitaldog
Sr. Member
****
Offline Offline

Posts: 9186



WWW
« Reply #48 on: February 28, 2007, 08:40:18 AM »
ReplyReply

This isn't a clog issue and no green casts have been seen from day one on any of these printers.

If you try the Printer Test file mentioned or, even easier to see and then measure, the Bill Atkinson 1728 patch target (i0), page one. You can see a number of red patches go off. In fact, I measured the full target using the fix and prior. The one red patch in the upper middle quadrant is off a deltaE 2000 of 9! Other patches of dissimilar colors are less than 1.

There's a very, very small range of red hues that get affected. I suspect there are lots of users who have this issue and don't even know it.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
francois
Sr. Member
****
Offline Offline

Posts: 6870


« Reply #49 on: February 28, 2007, 08:41:14 AM »
ReplyReply

Quote
I think so but there's some hesitation (and a lot of wasted ink and paper) as I'm just not sure what causes this mess and if there's a long term fix (other than doing the default trick which does appear to work).

I called my bud Mac Holbert at Nash Editions. He's having the same problems and noticed them months back. He's be using the Default trick (in Printer Utility) and says the problems are gone.

As to the old switch-a-row with the profiles in ColorSync, I'd like to test this on a few other Macs and also see if the connection (USB versus Network) has any role (I don't think so).

And as you point out, installing a new Epson driver could be an issue. The 4800 is new on this one Mac and since I'm waiting on a network card, its USB. The other Mac's are clean (didn't get install the Epson drivers). Then there's the fact that one machine is an Intel and the other PPC (I don't think that's a factor).

Now here's another test. Substitute the 3800 Standard profile in ColorSync with the 4800 (or vise versa).

I'm almost ready to just use the Default Printer Utility trick because all the variations here add up to a lot of wasted ink and paper! And you never know, Leopard may ship and the entire issue may go away (or get worse).

Bottom line is, if we know users have more than one printer and are using an Epson (some, all?) on a Mac, they should be aware of the Printer Default trick or they may end up with issues on some images with subtle reds of certain hues. And making or using custom profiles? Ouch.
[a href=\"index.php?act=findpost&pid=103747\"][{POST_SNAPBACK}][/a]
Andrew,
I performed the tests and my conclusion is that I need to use either the Default Printer trick or the Generic RGB trick. I've tried all the possible scenarios with the CS Utility application/Printer Default setting switching.
I can get rid of the banding only when my Epson printer is set as default or if I use the CS utility to set my Epson printer to Generic RGB. As soon as I change the system default printer to some other printer, I get banding immediately.

For me, I'll use the Default Printer trick. It seems less "exotic" than changing default printer profiles...

PS: I haven't bothered to try with the developer preview of Leopard... I'll wait until the final version is out and I'll probably waste more ink/paper.
« Last Edit: February 28, 2007, 08:49:17 AM by francois » Logged

Francois
digitaldog
Sr. Member
****
Offline Offline

Posts: 9186



WWW
« Reply #50 on: February 28, 2007, 08:51:49 AM »
ReplyReply

Quote
Andrew,
I performed the tests and my conclusion is that I need to use either the Default Printer trick or the Generic RGB trick. I've tried all the possible scenarios with the CS Utility application/Printer Default setting switching.
[a href=\"index.php?act=findpost&pid=103755\"][{POST_SNAPBACK}][/a]

Sounds like a plan. I think what I'll do next is try the Generic RGB trick, output some targets and compare them to the Epson Standard profile. If the data is essentially the same (or if the gamut of the resulting profile looks bigger/smoother) then maybe the Generic RGB is the way to go.

I'm also trying to get info from Epson about what this Standard profile is all about.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
francois
Sr. Member
****
Offline Offline

Posts: 6870


« Reply #51 on: February 28, 2007, 09:02:11 AM »
ReplyReply

Quote
Sounds like a plan. I think what I'll do next is try the Generic RGB trick, output some targets and compare them to the Epson Standard profile. If the data is essentially the same (or if the gamut of the resulting profile looks bigger/smoother) then maybe the Generic RGB is the way to go.

I'm also trying to get info from Epson about what this Standard profile is all about.
[a href=\"index.php?act=findpost&pid=103758\"][{POST_SNAPBACK}][/a]
From the prints I made earlier this year I noticed that the reds on you test photos (lady sweater) might be more intense. It's not a large difference and only visible when I compare the two prints side-by-side.

Let me know if you see a difference, I'm curious.
Logged

Francois
Roy
Full Member
***
Offline Offline

Posts: 196


WWW
« Reply #52 on: February 28, 2007, 11:32:45 AM »
ReplyReply

Quote
As to the old switch-a-row with the profiles in ColorSync, I'd like to test this on a few other Macs and also see if the connection (USB versus Network) has any role (I don't think so).

And as you point out, installing a new Epson driver could be an issue. The 4800 is new on this one Mac and since I'm waiting on a network card, its USB. The other Mac's are clean (didn't get install the Epson drivers). Then there's the fact that one machine is an Intel and the other PPC (I don't think that's a factor).
[a href=\"index.php?act=findpost&pid=103747\"][{POST_SNAPBACK}][/a]

I have had the problem with a network-connected 4800 using the current Epson drivers on both a PPC Mac and an Intel Mac. The switch to Generic RGB solved the issue on both machines.
Logged

Roy
digitaldog
Sr. Member
****
Offline Offline

Posts: 9186



WWW
« Reply #53 on: February 28, 2007, 11:42:13 AM »
ReplyReply

Good news. It appears that using Generic RGB produces the same output as whatever Epson Standard is. The deltaE 2000 report comparing Generic RGB versus Epson standard shows an average of 0.32 which could account for the instrument or differences in dry down (I didn't let the target just output using Generic RGB dry overnight).

Now I have to hammer on Epson to find out what's the difference here (and why).

Another interesting item. You can go into the Package that contains all the Epson profiles and pull em out so you can have access to them. First of all, the Epson 3800 Standard profile and the Epson 4800 Standard profile appear identical! They plot identically in ColorThink. Then I plotted Generic RGB and its almost a prefect match up too (off a tiny amount).
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
eronald
Sr. Member
****
Offline Offline

Posts: 4117



« Reply #54 on: February 28, 2007, 03:59:56 PM »
ReplyReply

Andrew, where is this Printer Test file you mentioned ?

Edmund
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9186



WWW
« Reply #55 on: February 28, 2007, 04:02:17 PM »
ReplyReply

Quote
Andrew, where is this Printer Test file you mentioned ?

Edmund
[{POST_SNAPBACK}][/a]


[a href=\"http://www.digitaldog.net/tips/]http://www.digitaldog.net/tips/[/url]
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
eronald
Sr. Member
****
Offline Offline

Posts: 4117



« Reply #56 on: February 28, 2007, 05:12:53 PM »
ReplyReply

Quote
http://www.digitaldog.net/tips/
[a href=\"index.php?act=findpost&pid=103853\"][{POST_SNAPBACK}][/a]


Andrew,

 Thank you very much !

Edmund
Logged
francois
Sr. Member
****
Offline Offline

Posts: 6870


« Reply #57 on: March 01, 2007, 02:03:56 AM »
ReplyReply

Quote
Good news. It appears that using Generic RGB produces the same output as whatever Epson Standard is. The deltaE 2000 report comparing Generic RGB versus Epson standard shows an average of 0.32 which could account for the instrument or differences in dry down (I didn't let the target just output using Generic RGB dry overnight).

Now I have to hammer on Epson to find out what's the difference here (and why).

Another interesting item. You can go into the Package that contains all the Epson profiles and pull em out so you can have access to them. First of all, the Epson 3800 Standard profile and the Epson 4800 Standard profile appear identical! They plot identically in ColorThink. Then I plotted Generic RGB and its almost a prefect match up too (off a tiny amount).
[a href=\"index.php?act=findpost&pid=103782\"][{POST_SNAPBACK}][/a]
Yesterday evening I reprinted your test file with both methods and I also got similar results. I cannot tell any difference by the eyes.
Logged

Francois
Roy
Full Member
***
Offline Offline

Posts: 196


WWW
« Reply #58 on: March 01, 2007, 01:53:48 PM »
ReplyReply

Quote
Good news. It appears that using Generic RGB produces the same output as whatever Epson Standard is. The deltaE 2000 report comparing Generic RGB versus Epson standard shows an average of 0.32 which could account for the instrument or differences in dry down (I didn't let the target just output using Generic RGB dry overnight).

Now I have to hammer on Epson to find out what's the difference here (and why).

Another interesting item. You can go into the Package that contains all the Epson profiles and pull em out so you can have access to them. First of all, the Epson 3800 Standard profile and the Epson 4800 Standard profile appear identical! They plot identically in ColorThink. Then I plotted Generic RGB and its almost a prefect match up too (off a tiny amount).
[a href=\"index.php?act=findpost&pid=103782\"][{POST_SNAPBACK}][/a]

If that is true, it is difficult to understand why substituting Generic RGB for the standard profile would make any difference.

However, they don't seem to be the same. The LAB plots of Generic RGB and Epson 4800 Standard displayed in the ColorSync utility show visible differences. Generic RGB is a matrix-based display profile. Epson Standard is a table-based output profile. They use different white points.

The only way to make ColorSync leave color values alone when printing is to make Generic RGB the source and destination profile (see pages 312-314 of Real World Color Management). If Epson Standard is different from Generic RGB (which it is), ColorSync will change the numbers. That agrees with what we see when we try to print with No Color Management and ColorSync puts the Standard profile in the chain.
Logged

Roy
digitaldog
Sr. Member
****
Offline Offline

Posts: 9186



WWW
« Reply #59 on: March 01, 2007, 02:02:50 PM »
ReplyReply

Quote
If that is true, it is difficult to understand why substituting Generic RGB for the standard profile would make any difference.

I agree. Right now, I'm not comfortable substituting Generic RGB when the default setting in the Print Utility, although a drag, seems to fix the issue.

Quote
However, they don't seem to be the same. The LAB plots of Generic RGB and Epson 4800 Standard displayed in the ColorSync utility show visible differences. Generic RGB is a matrix-based display profile. Epson Standard is a table-based output profile. They use different white points.

The 3800 and 4800 Standard appear to be identical. The Generic RGB isn't but gamut wise, it's not far off. I didn't say they were identical and to be honest,I only plotted the gamuts in ColorThink, I didn't look at the tags.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Pages: « 1 2 [3] 4 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad