Print lighting, NEC PA272w Calibrations..Printing...differing results

[Pete JF]

So you're NOT rolling your own profiles for each paper? You're using Canson's supplied profiles? Don't know if it's been mentioned already but printers can drift over time but then that would mean it would affect Epson's ABW profile as well.

To nail down what you describe as brighter looking prints is this in the blacks and shadows only giving an overall washed out contrast where you can still see distinction in the separation in highlight detail going by that small 12 step grayramp in the lower center of your test image. IOW do the blacks look as dense as in the test image?

Hi Tim, thanks for your input..Im pretty satisfied that mine was a soft proofing issue.

As for Printer Manages Color: a crap-shoot. I want to control what I'm producing, and because I'm properly colour-managed, my waste ratio is in the range of 1%.

Hi Mark..I've been making a variety of prints this afternoon-testing out the 'printer managed' scheme. These prints are extremely close to my monitor and more importantly for me..accurate to what I'm happy with in the develop module...variety of subjects with different challenges So far I'm very impressed with the consistency and accuracy that I'm seeing. *seem* like a piece of cake to me, I'm thrilled but feel like I'm cheating a bit..Out of about ten test there is one print that came out lighter than what I have on screen..colors accurate..everything except overall value..noticeably lighter..

[Tim Lookingbill]

As for Printer Manages Color: a crap-shoot.

These prints are extremely close to my monitor and more importantly for me..accurate to what I'm happy with in the develop module...variety of subjects with different challenges So far I'm very impressed with the consistency and accuracy that I'm seeing.

MY print viewing is being done, now, under a Hyperikon 3000k (testing a bit with 4000k Hyperikon too) BW and color prints under same lighting of course..

The 3000K Hyperikon LED's will make a print noticeably warm, far warmer than the 4000K shown in the link below compared to 5000K so I'm having a hard time seeing anyone getting match to a 6500K calibrated display.

http://forum.luminous-landscape.com/index.php?topic=116755.20

But I notice my "All In One" Epson NX330 "Printer Manages Color" out of Photoshop prints with a noticeable royal blue bias where I have to add yellow by reducing the blue channel middle slider in Levels before printing.

What do your prints look like next to an open window during the day? Do they match your display?

[Pete JF]

Hi Tim..I've been checking them under three light sources..a  3000k hyper ikon a 4000k Hyperikon and the halogen floods in my studio..window light looks fine too..

Most of my work these days ends up under gallery lighting or home 'gallery' lighting in the case of private clients or buyers..I always reference that temp range heavily..

Im printing on a 7800 and I'm still running tests in 'printer manages color' as we speak..I don;t see any of that..everything is coming out tight..im sort of waiting for the other shoe to drop..lol

Tim, are you changing your settings in the "Print Matching" menu of the Epson Driver to "Epson Color Control"

Then go to the Print Settings menu and select "Adobe RGB"

In the LR print module then select "Managed By Printer"..

That seems to be the extent of Ctein's scheme for doing this..

[Tim Lookingbill]

Tim, are you changing your settings in the "Print Matching" menu of the Epson Driver to "Epson Color Control"

Then go to the Print Settings menu and select "Adobe RGB"

In the LR print module then select "Managed By Printer"..

That seems to be the extent of Ctein's scheme for doing this..

I don't print out of LR. I print from CS3 Photoshop by first selecting "Managed By Printer" then "Epson Color Control", "AdobeRGB" and "Gamma 2.2". I have the last three saved as a preset. It's an "All In One" so I don't expect perfection especially concerning color tint bias.

For some reason the CS5 Photoshop changed their print dialog box where I can't call up the Epson driver settings like I can in CS3. This flakiness makes me go the easy route and print out of CS3. I can print just fine from Mac OS 10.6.8's "Preview" app and Firefox browser, CS5 PS gives me trouble.

[Mark D Segal]

Hi Mark..I've been making a variety of prints this afternoon-testing out the 'printer managed' scheme. These prints are extremely close to my monitor and more importantly for me..accurate to what I'm happy with in the develop module...variety of subjects with different challenges So far I'm very impressed with the consistency and accuracy that I'm seeing. *seem* like a piece of cake to me, I'm thrilled but feel like I'm cheating a bit..Out of about ten test there is one print that came out lighter than what I have on screen..colors accurate..everything except overall value..noticeably lighter..

Well, can't argue with success; if you are 90% happy and it's working well for you, it's not cheating; just carry on using it. But only for Epson papers listed in the driver, or 3rd-party media that are very close to an Epson Media Type you can select in the driver, otherwise it will not necessarily work well for you.

[Garnick]

I don't print out of LR. I print from CS3 Photoshop by first selecting "Managed By Printer" then "Epson Color Control", "AdobeRGB" and "Gamma 2.2". I have the last three saved as a preset. It's an "All In One" so I don't expect perfection especially concerning color tint bias.

For some reason the CS5 Photoshop changed their print dialog box where I can't call up the Epson driver settings like I can in CS3. This flakiness makes me go the easy route and print out of CS3. I can print just fine from Mac OS 10.6.8's "Preview" app and Firefox browser, CS5 PS gives me trouble.

Tim - This is also your quote from reply # 12 - "This isn't a lighting issue".  Perhaps I'm missing something here, but I simply cannot understand how you come to that conclusion, even though I am definitely not ruling out other very important factors.  However, it would seem that you are using a single bare LED bulb for judging your prints, probably the worst scenario for such activity.  For print/test judging one would normally use a diffused light source that would completely cover the print in a much more even manner.  In no way am I discarding the possibility of using LED bulbs, but I would never depend on one or two bare bulbs to do the job, especially when viewing/judging a print larger than perhaps 11x14.  I was also of the the impression that you were printing in a professional environment and with Pro Graphic Printers, but it would seem that is not the case.  Not to rule out the possibility that an "All In One" can perhaps produce an acceptable print, but in my opinion they are far from what I would ever consider professional printers of any category.  I print with only Pro Graphic Printers because the quality of output is what I and my customers demand, and that's exactly we receive, a professionally produced final print.  Now of course I will admit that some of the files I receive present much more of a challenge than those from photogs who possess at least some knowledge of colour management, and include it in their workflow.  However, it's my job to produce a print that's as good as it can be, regardless of the quality of the image file I have been presented with.  That has always been my job, whether in a traditional lab printing type "C", type "R", Ciba, Dye Transfer etc., or for the past 12 years in the digital lab.  But quite enough of my boring history.

Once more I return to my concern that it would seem at least some of the posters here have no concept of a constant/standard intensity of light under which a print should be viewed/judged.  Perhaps I'm the one who's out of the loop here, since my background always relied on a standard light intensity that most professionals were very comfortable with, as far as producing prints that would be very acceptable under varied lighting situations.  Of course if one is printing for a particular photographer or gallery lighting situation it might be prudent to adapt one's own lighting accordingly if possible.  In my situation that scenario has not been an issue, since I have never received complaints concerning the density of a print I had produced.  And of course I then come back to the LUX value of a lighting setup for viewing/judging prints.  Of course that intensity can also be measured in Foot Candles, EV etc., but it all converts to the same thing eventually, an average lighting intensity for the purpose at hand.  In my business that means printing my own work, but mostly for other professionals and amateurs alike.  Perhaps that "standard" has now become a moving target, one that has no bearing on any sort of standard at all in actuality.  However, in my business that's not a target I can pursue.

It's been very many years since I've had the urge to print from PS CS3 or CS5, and baring some sort of OS catastrophe that scenario will never interest me again.  It is your choice and I respect that, but I must say I do not understand the reasoning behind that choice.  Can you not upgrade your Mac OS to accommodate a newer PS version and print from that?  I am one who has yet to enter the "cloud" and still print from PS CS6 and occasionally LR, but when it reaches the point where the subscription becomes a necessity I will make the jump.  Thus far that has not been the case, and I will soon be upgrading from Mac OS 10.9.5 to 10.11.x.  I have a boot clone with El Capitan installed for testing the printers and other apps before installing the upgrade on my Mac HD, all of which will be done as soon as I am settled after my recent move.   

Gary     

 

 

[Pete JF]

Well, can't argue with success; if you are 90% happy and it's working well for you, it's not cheating; just carry on using it. But only for Epson papers listed in the driver, or 3rd-party media that are very close to an Epson Media Type you can select in the driver, otherwise it will not necessarily work well for you.

Like i said to Tim, sort of waiting for the other shoe to drop..so far so great..we'll see. I have only a few regular fav papers so maybe they'll fall into the happy group.. made a few of prints on Canson Baryta and also Harman Baryta with the same results.

[Mark D Segal]

Those two papers are fairly close to Epson Legacy Baryta in respect to driver handling.

[Pete JF]

It's been very many years since I've had the urge to print from PS CS3 or CS5, and baring some sort of OS catastrophe that scenario will never interest me again.  It is your choice and I respect that, but I must say I do not understand the reasoning behind that choice.  Can you not upgrade your Mac OS to accommodate a newer PS version and print from that?  I am one who has yet to enter the "cloud" and still print from PS CS6 and occasionally LR, but when it reaches the point where the subscription becomes a necessity I will make the jump.  Thus far that has not been the case, and I will soon be upgrading from Mac OS 10.9.5 to 10.11.x.  I have a boot clone with El Capitan installed for testing the printers and other apps before installing the upgrade on my Mac HD, all of which will be done as soon as I am settled after my recent move.   

Gary

I have a had a print station in place for many many years using a G5 and CS2..producing great prints to this day..the G5 and the Sony monitor are failing now and Im moving on, but..the prints off of that deal are stellar, still..i don't know if Tim is just as comfortable as I am with that, or whatever his limitations are but there is a lot to be said for sticking with something that just works..

:  )

[Garnick]

I have a had a print station in place for many many years using a G5 and CS2..producing great prints to this day..the G5 and the Sony monitor are failing now and Im moving on, but..the prints off of that deal are stellar, still..i don't know if Tim is just as comfortable as I am with that, or whatever his limitations are but there is a lot to be said for sticking with something that just works..

:  )

Hi Pete,

I am in total agreement with you on that point.  I still have a G5 that I use very infrequently for the Nikon Scan 3.14, which of course is not compatible with any OS beyond the G5.  I've experimented with Vuescan and the Nikon Super Coolscan 4000, but never with the results I can get with the Nikon driver.  When I have more time I'll get into Vuescan again and take it to the limits of the app.  It doesn't have the fine tuning that the Nikon Scan has, but I might be able to make it work well enough for what I need.  Lasersoft Imaging has a driver for the Nikon scanner but the price is prohibitive, given the infrequent use I have for that scanner now.  I've used Silverfast for the flatbeds for many years, but I'm reluctant to pop for the $$$ required for the Nikon version.  For quite few years I produced all of my work on the G5, but it finally came to the point where needed to step up the Mac Pro.  I now have two of them, one as a backup with cloned drives.  They are both the older style(2010-12), before the "garbage can" version of 2013.  Apple seems to be neglecting the professional imaging market within the past few years, so I'll probably stay with what I have now for quite a long time. 

Gary   

[Pete JF]

Yepp, Garnick..Scanner software is so important..went through the same motions you described with my Heidelberg Lino flatbed /trans/print scanner..SCSI, running off a beige G3..hahahahaha..almost right up till now..One of several beige G3's I found to keep that scanner running and to keep using the Linocolor software which was nothing short of amazing..

After the original G3 started getting very flaky..I tried to upgrade to a G4 with an om SCSI card..Linocolor wouldn't run in the newer OS that the G4 was released on..SO, I tried Vuescan. Seriously, not even worth comparing..no joy whatsoever in every regard..The Linocolor software was top shelf press stuff and was amazing in use and the results were stunning.

Ditched the G4 > went and found another beige g3 and kept it rolling through one more after that..just decommissioned the scanner about 4 months ago or so..felt sort of sad about that.

[Mark D Segal]

Also and very interesting: I was reading Ctein's method for letting Printer Manage etc..

This morning I set up a print according to his settings...Very impressive. I printed an image straight out of the Develop module using his 'Printer Managed' routine and it looks pretty much dead on..This feels promising for me.

Have any of you experimented with this routine? I noticed a thread on this forum about it..

Hi again Pete;

Re your question that I highlighted in bold above, I decided to have a second look at "Printer Manages Color", but I don't really "experiment" - I use scientific methods to produce repeatable and verifiable data that gives insight into the comparative quality of different processes and materials, as you would have perhaps appreciated from the numerous paper and printer reviews I have researched and written for this website.

So, following through with that approach, I prepared a test of "Printer Manages Color using Epson Legacy Baryta paper in my Epson SCP-800 printer. I regret to inform you that the results are absolutely and comparatively sub-standard. You can read-up on my methodology in the already published articles, I shall not repeat it here, except to mention that for printing the test target of course I selected Epson Color Control, Adobe RGB color space(more than enough for the target) and Gamma 2.2 in the printer driver and Printer Manages Color in the Photoshop Print screen. Just last December I had reprofiled my P800 for this paper in the usual way I make custom profiles (i1Profiler, i1Pro2 spectrophotometer, XRite 2033 patch target, dual scan mode for M0, M1 and M2 measurement conditions, etc.)

Key comparative results are as follows for the Printer Manages Color option (with the standard Photoshop Manages Color using my custom profile in brackets beside).

Overall Accuracy, Average dE: 6.05  (0.89)
Neutrals Average dE: 3.14  (1.08)
dE colour bias of the grayscale: 1.68  (0.82)
Three Worst Outcomes dE: Orange 13.6; Purplish Blue 10.3; Blue 13.4; (Cyan 1.6;  50% Gray 1.3; Foliage 1.2)
The "worst outcomes" for my custom profile are indeed not bad at all, the results for those colours under "Printer Manages Color" being considerably worse at 4.8, 3.1 and 5.0 respectively.

The instrument Epson used to make the internal profile that gets selected for Printer Color Management is not the same as my instrument, so a bit of measurement difference could be attributed to variances of the measurement technology itself, however nothing near to the extent of the gaps shown here.

As far as I'm concerned "Printer Manages Color" once again is demonstrably inferior to a high quality custom profiling approach using active colour management. I'll admit, however, this stuff can be charmingly deceptive, in the sense that just looking at the printed test target before any measurement, it doesn't look too bad at all. When the two targets are viewed side by side however, differences of these magnitudes are indeed noticeable.

[Ferp]

Also and very interesting: I was reading Ctein's method for letting Printer Manage etc..   I noticed a thread on this forum about it..

There were several threads in fact, as there were several Ctein articles.  To simplify matters, all you need to read is one detailed and systematic analysis of why Ctein got the results that he did:

http://forum.luminous-landscape.com/index.php?topic=104307.msg857415#msg857415

Mark (MHMG) may not agree with this statement, but I regard it as the definitive debunking of Ctein's position.

[Pete JF]

Hi again Pete;

Re your question that I highlighted in bold above, I decided to have a second look at "Printer Manages Color", but I don't really "experiment" - I use scientific methods to produce repeatable and verifiable data that gives insight into the comparative quality of different processes and materials, as you would have perhaps appreciated from the numerous paper and printer reviews I have researched and written for this website.

So, following through with that approach, I prepared a test of "Printer Manages Color using Epson Legacy Baryta paper in my Epson SCP-800 printer. I regret to inform you that the results are absolutely and comparatively sub-standard. You can read-up on my methodology in the already published articles, I shall not repeat it here, except to mention that for printing the test target of course I selected Epson Color Control, Adobe RGB color space(more than enough for the target) and Gamma 2.2 in the printer driver and Printer Manages Color in the Photoshop Print screen. Just last December I had reprofiled my P800 for this paper in the usual way I make custom profiles (i1Profiler, i1Pro2 spectrophotometer, XRite 2033 patch target, dual scan mode for M0, M1 and M2 measurement conditions, etc.)

Key comparative results are as follows for the Printer Manages Color option (with the standard Photoshop Manages Color using my custom profile in brackets beside).

Overall Accuracy, Average dE: 6.05  (0.89)
Neutrals Average dE: 3.14  (1.08)
dE colour bias of the grayscale: 1.68  (0.82)
Three Worst Outcomes dE: Orange 13.6; Purplish Blue 10.3; Blue 13.4; (Cyan 1.6;  50% Gray 1.3; Foliage 1.2)
The "worst outcomes" for my custom profile are indeed not bad at all, the results for those colours under "Printer Manages Color" being considerably worse at 4.8, 3.1 and 5.0 respectively.

The instrument Epson used to make the internal profile that gets selected for Printer Color Management is not the same as my instrument, so a bit of measurement difference could be attributed to variances of the measurement technology itself, however nothing near to the extent of the gaps shown here.

As far as I'm concerned "Printer Manages Color" once again is demonstrably inferior to a high quality custom profiling approach using active colour management. I'll admit, however, this stuff can be charmingly deceptive, in the sense that just looking at the printed test target before any measurement, it doesn't look too bad at all. When the two targets are viewed side by side however, differences of these magnitudes are indeed noticeable.

hi Mark, I appreciate your attention and time taken to do that. Im using a 7800. I set up my driver according to an article where he runs through all of the settings..no profiles involved. The prints I'm making are dead on matches. I'm no scientist, I'm an artist. I ran a commercial studio for 20 years in a large market, have plenty of experience in getting prints and transparencies readied for and then approved for press..as well as being at countless press checks..some it involving rigorous color matching. I trust my eyes.

I'm not saying that this can't fail, perhaps I jumped in lucky on what I'm printing. I'll take that any day. I'm saying that, so far, I'm very impressed over the span of about 4 different papers. There is something here that can work and i'm seeing it. Like said..it could come crashing down with right (wrong ) image or paper combo. I don't know, we'll see  :  )

I had one bump today with the Canson Baryta..I made some small prints of an image that was always challenging to nail..it printed a little bit flat and lacking the saturation present on screen.. but i was able to nail it with a couple of quick tweaks on the second small test and it looks phenomenal in now in a 16x20..happiest I've ever been with this one..That's all i got  :  )

[Pete JF]

There were several threads in fact, as there were several Ctein articles.  To simplify matters, all you need to read is one detailed and systematic analysis of why Ctein got the results that he did:

http://forum.luminous-landscape.com/index.php?topic=104307.msg857415#msg857415

Mark (MHMG) may not agree with this statement, but I regard it as the definitive debunking of Ctein's position.

Debunk away...This is working for what I've got right now. Clearly, there seem to be multiple situations where this scheme works quite well.

[Ferp]

Debunk away...This is working for what I've got right now. Clearly, there seem to be multiple situations where this scheme works quite well.

If it works for you then far be it from me to tell you that it doesn't.  The problem with printer manages color is that if you don't understand why it works in certain situations then you're not in a position to be able to predict when it will and when it won't.  If you read the extensive comments sections to the three Ctein articles on TOP, especially the last one, then you'll read that Windows users have a particular problem with setting printer manages color in the print dialog in Adobe software. You'll get a silent conversion to sRGB en route to the printer driver and that ain't going to be good for strongly saturated images.

[Mark D Segal]

hi Mark, I appreciate your attention and time taken to do that. Im using a 7800. I set up my driver according to an article where he runs through all of the settings..no profiles involved. The prints I'm making are dead on matches. I'm no scientist, I'm an artist. I ran a commercial studio for 20 years in a large market, have plenty of experience in getting prints and transparencies readied for and then approved for press..as well as being at countless press checks..some it involving rigorous color matching. I trust my eyes.

I'm not saying that this can't fail, perhaps I jumped in lucky on what I'm printing. I'll take that any day. I'm saying that, so far, I'm very impressed over the span of about 4 different papers. There is something here that can work and i'm seeing it. Like said..it could come crashing down with right (wrong ) image or paper combo. I don't know, we'll see  :  )

I had one bump today with the Canson Baryta..I made some small prints of an image that was always challenging to nail..it printed a little bit flat and lacking the saturation present on screen.. but i was able to nail it with a couple of quick tweaks on the second small test and it looks phenomenal in now in a 16x20..happiest I've ever been with this one..That's all i got  :  )

Hi Pete, if I were to repeat my testing on a 7800 (which I cannot) I would expect the results to be similar to what I obtained on the P800. However, if you find printer color management satisfactory for your set-up and purposes, as I said previously, who are we to discourage you - just go for what works for you. This thread just made me curious to go back and double-check my original perceptions of the relative merits of printer versus application colour management and I am now satisfied I know the answer, using very straightforward and reliable methodology.

[Tim Lookingbill]

Tim - This is also your quote from reply # 12 - "This isn't a lighting issue".  Perhaps I'm missing something here, but I simply cannot understand how you come to that conclusion, even though I am definitely not ruling out other very important factors.  However, it would seem that you are using a single bare LED bulb for judging your prints, probably the worst scenario for such activity.  For print/test judging one would normally use a diffused light source that would completely cover the print in a much more even manner.

I have over 5 brands of daylight balanced lights I've tested that put out enough evenly diffused light for my 8x10's. If I was printing poster size I'ld get several of the Hyperikon 5000K 110º floods because they have the best even spread of intense diffused white light I've ever seen in a light bulb. I would not use their 4000K and 3000K to check screen to print matching. Color of light is more important for checking color anomalies due to metamerizm and gamut clipping with the ink/paper combo.

[Tim Lookingbill]

And to point out the importance of using an "accurate to printer characterization" ICC profile for soft proofing I downloaded from Dry Creek Photo one for the Fuji Frontier DL600 at my Walmart. Since Walmart's DL600 prints only in sRGB, I instead assigned the ICC printer profile to the sRGB source file which shows the reddish bias in warm colors and clipping of cyans and jade colors pretty accurately in the source image.

The attached image shows the source file I sent to Walmart and a picture I took of my calibrated 27inch LG sRGB gamut display showing how close the profile compared to the photo I took of the 8x10 print at the bottom viewed under my Philips T8 5000K Natural Light flotubes.

[Garnick]

And to point out the importance of using an "accurate to printer characterization" ICC profile for soft proofing I downloaded from Dry Creek Photo one for the Fuji Frontier DL600 at my Walmart. Since Walmart's DL600 prints only in sRGB, I instead assigned the ICC printer profile to the sRGB source file which shows the reddish bias in warm colors and clipping of cyans and jade colors pretty accurately in the source image.

The attached image shows the source file I sent to Walmart and a picture I took of my calibrated 27inch LG sRGB gamut display showing how close the profile compared to the photo I took of the 8x10 print at the bottom viewed under my Philips T8 5000K Natural Light flotubes.

Hi Tim,

I apologize, but I still don't understand your devotion to Walmart as your go-to photo lab for your work.  I have seen my share of prints from both Walmart and Costco that were terrible examples of anything approaching a good print.  Banding and other issues lead me to believe that QC is severely lacking in both of these retail locations, and therefore, why would you assume that they are in any way interested in producing a quality print, especially given the price range they work in.  I also cannot understand why you would bother to download a profile from Dry Creek for a printer sitting in a Walmart store.  That is anything but a custom profile, and those stores would quite likely benefit from a custom profile even more than someone who actually has any amount of knowledge concerning inkjet printing.  Sorry Tim, but I'm totally confounded by the notion that somehow Walmart has now become thee place to go for quality printing. 

However, to change the subject slightly.  You mention that you have more than five brands of LED bulbs for testing.  Question #1 - do you have the CRI numbers for any of these bulbs and are those numbers substantiated?  Question #2 - At what degree of light intensity are you viewing your prints?  Of course that can be measured in EV or Foot Candles, which can then be stated in LUX or Lumens. 

Gary