I’m new to color mgt and an trying to profile an R800 with i1 Photo (bought last week). My objective at this stage is neutral b&w.
I’m using WinXP, Epson inks and Epson Prem Glossy Photo Paper and my monitor is profiled.
I have followed the wizard in iMatch 3.1 and created a profile (strip mode) with the i1 RGB 1.5.txt chart (well dried). When I print grayscale images from photoshop CS (applying the profile and using the no color mgt option in the driver), the prints are not neutral. They are slighly cool. I also tried patch mode with the same chart but that didn’t make any difference.
Does anybody have any suggestions how I could improve the profiling process? Should I be using a different chart? Any help would be appreciated. Cheers.
Refere to the topic “Black and white” I have posted some time ago, I received luminous answers. The printer was the 1290 but it seems to concerne exactly the same problem !
Just a follow up for those who are interested. I built a profile for the R800 using the larger test (actually did it twice just in case). I used Epson Prem Photo Glossy (it was the only paper I could get enough of at the time). For b&w at least, the profile (or at least the output using the profile) was very average. When I used the Eye-One gizmo with Eye-One Share to read various parts of a b&w print, the various tones all read cool (moved around a bit, but all were cool).
Subjectively, the prints were plagued by metamerism. They were all over the place. If anyone has had better luck with b&w on this printer, I’d be very interested to hear about it.
I’ve done limited printing in colour with the profile, but what I have done looks better. Might have to investigate the new Epson 4800/7800 for b&w…
Strange, I’ve bought recently the Espon 1800 that use the inks’cartridges than the R800.
I build the profile with my Eyeone, the results are quasi perfect, looks neutral with very very little metamerism. The profile were done with Profil Maker and the 1100 patches target
I use Epson Archival Matte paper : the results are a little cold (looks like bromide prints) and another pur rag paper called Andalousia , slightly warm (looks like chlrorobromide prints).
Well I use now exclusively this last paper (Andalousia) ; may be hard to find for a non french photographer ???
Just a follow up for those who are interested. I built a profile for the R800 using the larger test (actually did it twice just in case). I used Epson Prem Photo Glossy (it was the only paper I could get enough of at the time). For b&w at least, the profile (or at least the output using the profile) was very average. When I used the Eye-One gizmo with Eye-One Share to read various parts of a b&w print, the various tones all read cool (moved around a bit, but all were cool).
interesting… the cool reading will undoubtedly be influenced by the paper color so for a point of comparison what are the Lab values for paper white?
Also, what are your viewing conditions? This affects neutral inkjet prints so much that any discussion of them should really include viewing conditions.
Subjectively, the prints were plagued by metamerism. They were all over the place. If anyone has had better luck with b&w on this printer, I’d be very interested to hear about it.
metamerism is really tough to deal with if you are only controlling the printer through the RGB drivers. Manufacturer drivers (and Epson is no exception) tend to use less black/gray inks that might be best to keep tones stable. As a result, neutrals are built with a delicate (brittle) balance of CMY and a little black. It’s this balance that fails under certain lighting conditions. If you drive the printer with a RIP you can build CMYK profiles that use more K and reduce this balancing act.
I’ve done limited printing in colour with the profile, but what I have done looks better. Might have to investigate the new Epson 4800/7800 for b&w…
well that is certainly a different platform an it seems to be giving satisfying B&W results.
Regards,
Steve
o Steve Upton CHROMiX www.chromix.com
o (hueman) 866.CHROMiX
Also, if you’re not dead set on an RGB workflow and you’ve only this B&W issue occasionaly, you can build a custom CMYK separation with GCR and then ouput the resulting CMYK file.
So long as your printer driver doesn’t recombine in some way, this would have the same affect as using a RIP to perform the GCR.
It works very well on my new Epson R200 with Photshop 9 in Panther.
…If you drive the printer with a RIP you can build CMYK profiles that use more K and reduce this balancing act…
Also, if you’re not dead set on an RGB workflow and you’ve only this B&W issue occasionaly, you can build a custom CMYK separation with GCR and then ouput the resulting CMYK file.
So long as your printer driver doesn’t recombine in some way, this would have the same affect as using a RIP to perform the GCR.
It works very well on my new Epson R200 with Photshop 9 in Panther.
not sure I understand you here…
if you are not driving the printer w/ a RIP then you don’t have control over the printer using CMYK. You can do work on a CMYK file but as soon as you print it Photoshop has to convert it to RGB in order to pass it off to the OS and print driver…
can you print a K-only gray ramp and get black only ink out?
Regards,
Steve
o Steve Upton CHROMiX www.chromix.com
o (hueman) 866.CHROMiX
Yes, I believe my K only gray ramp is truly K only… it certainly looks different than my CMY ramp (I think that must have something to do with the two extra colors. 1/2 c and 1/2 m.)
Is it the “norm” for Photoshop to do the CMYK to RGB conversion? I thought that happened in the printer driver. (I may be wrong…)
I’ll spit out a target later and see what I can see. I hope I haven’t lead anyone down the wrong path…
Steve: The Epson Premium Glossy reads L* 92.8 a* -1.0 b* -4.1. I am a complete newbie to all this, so I’m not sure what that really means, but on the circle (i1 Share -> Evaluate -> Accuracy), the little crosshair is a mm or two south of the centre.
As far as lighting goes, I have viewed the prints under all kinds (daylight, tungsten, fluro), and that is where the metamerism problems really kick in. Magenta, yellow, green…
Norbert: 1100 patches? With i1 Photo the biggest target I have is 918. Can I use other targets with i1 Match (3.2a)?
I did some profiles for Epson Archival Matte this morning and the b&w prints are much better. Neutral-cool, and comparatively stable. Colour was spot on as far as I could tell.
At the moment I only have Epson papers, I think I’ll experiment with some different ones and see how I go.
if you are not driving the printer w/ a RIP then you don’t have control over the printer using CMYK. You can do work on a CMYK file but as soon as you print it Photoshop has to convert it to RGB in order to pass it off to the OS and print driver…
can you print a K-only gray ramp and get black only ink out?
Yes, I believe my K only gray ramp is truly K only… it certainly looks different than my CMY ramp (I think that must have something to do with the two extra colors. 1/2 c and 1/2 m.)
Is it the “norm” for Photoshop to do the CMYK to RGB conversion? I thought that happened in the printer driver. (I may be wrong…)
due to the OS architecture in Windows, OS 9 and normal OS 10 the OS takes the print job and hands it to the driver. The OS only understands RGB so the application is responsible for conversion of any non-RGB images to RGB prior to hand-off. If a Postscript driver is used then CMYK and other colors can be used.
I’ll spit out a target later and see what I can see. I hope I haven’t lead anyone down the wrong path…
it’s worth taking a look. I suspect you will see a K ramp print using 4 (or more) inks. It may still look different from a CMY ramo though.
Regards,
Steve
o Steve Upton CHROMiX www.chromix.com
o (hueman) 866.CHROMiX
Steve: The Epson Premium Glossy reads L* 92.8 a* -1.0 b* -4.1. I am a complete newbie to all this, so I’m not sure what that really means, but on the circle (i1 Share → Evaluate → Accuracy), the little crosshair is a mm or two south of the centre.
right, well a negative b value certainly means blueish. That could be actual blue paper or (more likely) brighteners in the paper. Either way a 50% gray would measure some of that blue as well so seeing those numbers is not necessarily an error.
As far as lighting goes, I have viewed the prints under all kinds (daylight, tungsten, fluro), and that is where the metamerism problems really kick in. Magenta, yellow, green…
fun isn’t it. One of those physics lessons you could really do without.
Norbert: 1100 patches? With i1 Photo the biggest target I have is 918. Can I use other targets with i1 Match (3.2a)?
sort of. You can hack other target files into Match or use the MeasureTool part of ProfileMaker to measure targets and then bring them into Match for profile building
I did some profiles for Epson Archival Matte this morning and the b&w prints are much better. Neutral-cool, and comparatively stable. Colour was spot on as far as I could tell.
good
At the moment I only have Epson papers, I think I’ll experiment with some different ones and see how I go.
probably wise. I think one of the things you are fighting against is the brighteners in the paper.
Regards,
Steve
o Steve Upton CHROMiX www.chromix.com
o (hueman) 866.CHROMiX
In Jaguar, I used Gimp-Print with ESP Ghostscript to turn my inkjet printers in to PostScript devices (sort of)…
In Jaguar, the system always describes any printer driven by Gimp-Print as a PostScript device and therefore CMYK information is retained. I’m not sure how it would behave in Panther… ESP Ghostscript isn’t used in the OS in Panther so I don’t have a clue how the CUPS would handle the PostScript to raster conversion. (Not to mention Tiger!)
This is not the method I use currently, but it’s something worth investigating. (Note: I don’t know the extent of the color management support with this method)
OK, I’ve downloaded the sample of ProfieMaker and found the 1100 patch target. Getting that into iMatch is a bit more tricky. Can anyone point me in the direction of some help with these tasks? I can’t find them in the quick start guide.
[quote:7e7fa7a5af=“upton at chromix.com”]
sort of. You can hack other target files into Match or use the MeasureTool part of ProfileMaker to measure targets and then bring them into Match for profile building
[/quote:7e7fa7a5af]
OK, I’ve downloaded the sample of ProfieMaker and found the 1100 patch target. Getting that into iMatch is a bit more tricky. Can anyone point me in the direction of some help with these tasks? I can’t find them in the quick start guide.
as far as I know all you need to do is replace one of the existing targets with the new one - while giving the new one the old name…
you may be able to measure this way or you might need to measure in ProfileMaker’s MeasureTool…
I’ve done it before but it was a few versions ago
Regards,
Steve
o Steve Upton CHROMiX www.chromix.com
o (hueman) 866.CHROMiX
…due to the OS architecture in Windows, OS 9 and normal OS 10 the OS takes the print job and hands it to the driver. The OS only understands RGB so the application is responsible for conversion of any non-RGB images to RGB prior to hand-off. If a Postscript driver is used then CMYK and other colors can be used.
Gotcha… have you ever played with Gimp-Print?
I had a feeling you were going to mention that. That’s why I said “normal OS 10” above.
Yes, GIMP-Print does allow CMYK (and more) channel access to printers. It’s pretty cool but also a bit tricky to setup.
In Jaguar, I used Gimp-Print with ESP Ghostscript to turn my inkjet printers in to PostScript devices (sort of)…
yes. There’s a Postscript to PDF interpreter in OS X that converts it to PDF and then it gets printed. We’re using it here in our office to print Postscript to a shared inkjet… amazing
In Jaguar, the system always describes any printer driven by Gimp-Print as a PostScript device and therefore CMYK information is retained. I’m not sure how it would behave in Panther… ESP Ghostscript isn’t used in the OS in Panther so I don’t have a clue how the CUPS would handle the PostScript to raster conversion. (Not to mention Tiger!)
it works fine as far as I can tell…
This is not the method I use currently, but it’s something worth investigating. (Note: I don’t know the extent of the color management support with this method)
it is indeed worth more playing. There does not seem to be any color management support yet so it needs to be done at the application level.
Regards,
Steve
o Steve Upton CHROMiX www.chromix.com
o (hueman) 866.CHROMiX
(I think that must have something to do with the two extra colors. 1/2 c and 1/2 m.)