Luminous Landscape Forum
Raw & Post Processing, Printing => Printing: Printers, Papers and Inks => Topic started by: Mark Lindquist on December 07, 2017, 09:24:30 pm
-
Wow - after 2 -3 three grueling days of just grunting things out, going back and forth across the country to people like Doug Gray, Andrew Rodney, Brad P., Kers from Amsterdam, Ernst Dinkla from Holland, John Dean, Graham By, etc., Mark McCormick-Goodhart (MHMG) and I have managed to breakthrough to the other side with profiling where angels fear to tread, and I'm here to tell you that the results are staggering!
Mark can fill you in on the specific details much better than I can so I'll hope he chimes in and answers the many questions that will no doubt come up.
First, it took actually 1 hour to make the 6000 patch target profile. I had a partial 44" wide roll of Kodak Semi-Gloss to do testing with - figured if it didn't work, no problem. Having done several profiles prior to this one, Mark MG and I were creating them simultaneously from charts he made with i1Profiler. After long discussions with Doug Gray on the previous thread by Brad P., we came to a stopping point with larger size targets with the Mac towers we both were using. The print would go for about 25 rows on a 44" wide roll and then just stop without finishing the target. This happened to me several times. So when we began a new round of chart and target making today, starting with 1457 patches, then 1877 patches, then 2371 patches, I began doing my printing using the Windows HP Utility rather than the Mac Utility.
Long ago, HP stripped the elegant GUI from the Mac Utility which used to look identical to the Windows Utility in favor of a widget based, streamlined approach. It has always been solid and has worked beautifully until Mark MG found the bug, that it wouldn't print the 2371 target. The windows print sailed right on through. Badda Bing! Huge breakthrough right there. Working on cross-platform (Windows/Mac) has always been important to me and I use VMWare Fusion for Mac since I can run both systems concurrently. So Mark was stopped. He went on to make 2945 patch targets as well as 3605, 4357, and the big Kahuna the 6000 patch target file that no one in their right mind would ever consider doing.
So I continued on, and printed more targets successfully, the 2371, the 2945 and then I just jumped into the deep end and went for the 6000. I thought it might not do it, but had a lot of confidence at this point. Keep in mind that we're just testing at this point. So son-of-a gun, the HP Z3200ps the lone printer out there that comes with an embedded spectrophotometer as part of the standard equipment and price since it was first created as the Z3100 about 15 years ago, crunched through the data and produced a whopper of a 6000 patch target then read it and created a txt file to be converted to an ICC Profile in right at an hour.
We're talking the lowly Z3200ps printer not even on anyone's radar anymore. David just killed Goliath Folks! Bam!
And so I sent my file off to Mark and he did what he did with it, which was to run it breezily through RGB Drop, and I ran my file through ICC Gen, which ran it through Argyll. 4 minutes and I had a magical 6000 patch target ICC profile made in just over an hour, in-house, ready to rock and roll. And man oh man did it ever.
Mark compared the ICC Gen/Argyll to the RGB Drop and the Rel Col is virtually the same. Cha CHING!
So I made a 44" wide print from an excruciatingly difficult print to make - an almost impossible print, and it just came out luminous, buttery smooth, open toned, exceptional transitions, and I have to say EUREKA!!!
So there it is, for those following the Brad P. thread - use the Windows HP Utility to do your profiling even if you're on a mac. Stable and solid, which is why I fought so long and hard with HP to reinstate the Windows Utility after they dropped it. I used Windows 8 which is on my machine, with the matching Utility. Any will work fine however. Windows 7/64 has the Utility, as does XP and Windows 10 (now), and they should all be capable of making any size profile you want and they are available for download on the HP website. The Utility is under "Software". Make your charts however you wish, then print them on the Z3200ps and just grin at others using "more superior" printers, HAHAHAHHAHAHAHAHAH!
Best wishes - I'm a happy camper. Hope someone will enjoy using their Z to do the impossible in an hour. A full write-up and tutorial by MHMG and me, coming eventually on Z3200.com
Thanks for the help from all those mentioned and apologies if I missed you.
Special thanks to Geraldo Garcia for helping me out in the first place. Thanks to Argyll - donate NOW!
Mark McCormick Goodhart - You ROCK, man.
Best -
Mark L
-
That's amazing guys. You have really sparked my interest in revisiting profiling specialized targets my both of my Zs. I've been so busy with Christmas work that Ive only been listening to you guys so far. My plan is to try these expansive targets during the holidays when I'm also going into using Angels workflow for making digital OHP negs for alternative process. That is something I will keep documention on for your site with the various processes I'm going to try out in order to teach a workshop here in April on experimental things.
I am very much also interested in, after hearing all this, in profiling with an extended gray target of big patch size with low gamut and tons of very subtle gray patches to see just exactly how far I can push the Z in the monochrome range.
About 6 years ago a bunch of us got together at my place and shared work, and one guy was making some amazingly rich and clean neutral monochrome prints using an Epson 7800 printer with K3 inks. This was done with cmyk profiling out of Studio Print. I mean you saw absolutely no hit of of any color cast in any of the tonal range. It looked just like a dedicated black and white inkset was used. At the time I never went there because of the limited longevity of Epson's color inks at that time, as well as the difficulty of figuring this out on a 12 ink inkset like the Canon or Hp.
On another topic, is it possible to make extended patch linearization custom targets as well as custom icc profile targets? I doubt that is possible in the hp software but probably not necessary if you are running rgb with extended patch grays.
Has anyone heard of someone having possession of a big grayscale target like that for the icc profile creation to read in one step? One guy from Hong Kong sent me an rgb target that had extended grays along with color patches, about 1775 patches or so. It did give me better shadow detail on the Z through i1 profiler, but for color it wasn't up to my 2033 target for regular color printing but on the z3100 I was using it for black and white .
I still feel like there is more to be squeezed out of the Z in regard to high value dimensionality utilizing the 2 grays, two blacks and light cyan, light magenta, yellow and red in the mix. The beauty of this vivera inkset is the way the hues fade generally at the same rate. How hard would it be to create such a 2000 patch target for that purpose if one doesn't exist? Probably overkill but you wouldn't know if you don't try it. It was certainly possible on the Epson with cmyk profiling.
John
Wow - after 2 -3 three grueling days of just grunting things out, going back and forth across the country to people like Doug Gray, Andrew Rodney, Brad P., Kers from Europe, (or Scandinavia), Ernst Dinkla from Holland, John Dean, Graham By, etc., Mark McCormick-Goodhart (MHMG) and I have managed to breakthrough to the other side with profiling where angels fear to tread, and I'm here to tell you that the results are staggering!
Mark can fill you in on the specific details of the what's, why's, how's and where's much better than I can so I'll hope he chimes in and answers the many questions that will no doubt come up.
First, it took actually 1 hour to make the 6000 patch target profile. I had a partial 44" wide roll of Kodak Semi-Gloss to do testing with - figured if it didn't work, no problem. Having done several profiles prior to this one, Mark MG and I were creating them simultaneously from charts he made with i1Profiler. After long discussions with Doug Gray on the previous thread by Brad P., we came to a stopping point with larger size targets with the Mac towers we both were using. The print would go for about 25 rows on a 44" wide roll and then just stop without finishing the target. This happened to me several times. So when we began a new round of chart and target making today, starting with 1457 patches, then 1877 patches, then 2371 patches, I began doing my printing using the Windows HP Utility rather than the Mac Utility.
Long ago, HP stripped the elegant GUI from the Mac Utility which used to look identical to the Windows Utility in favor of a widget based, streamlined approach. It has always been solid and has worked beautifully until Mark MG found the bug, that it wouldn't print the 2371 target. The windows print sailed right on through. Badda Bing! Huge breakthrough right there. Working on cross-platform (Windows/Mac) has always been important to me and I use VMWare Fusion for Mac since I can run both systems concurrently. So Mark was stopped. He went on to make 2945 patch targets as well as 3605, 4357, and the big Kahuna the 6000 patch target file that no one in their right mind would ever consider doing.
So I continued on, and printed more targets successfully, the 2371, the 2945 and then I just jumped into the deep end and went for the 6000. I thought it might not do it, but had a lot of confidence at this point. Keep in mind that we're just testing at this point. So son-of-a gun, the HP Z3200ps the lone printer out there that comes with an embedded spectrophotometer as part of the standard equipment and price since it was first created as the Z3100 about 15 years ago, crunched through the data and produced a whopper of a 6000 patch target then read it and created a txt file to be converted to an ICC Profile in right at an hour.
We're talking the lowly Z3200ps printer not even on anyone's radar anymore. David just killed Goliath Folks! Bam!
And so I sent my file off to Mark and he did what he did with it, which was to run it breezily through RGB Drop, and I ran my file through ICC Gen, which ran it through Argyll. 4 minutes and I had a magical 6000 patch target ICC profile made in just over an hour, in-house, ready to rock and roll. And man oh man did it ever.
Mark compared the ICC Gen/Argyll to the RGB Drop and the Rel Col is virtually the same. Cha CHING!
So I made a 44" wide print from an excruciatingly difficult print to make - an almost impossible print, and it just came out luminous, buttery smooth, open toned, exceptional transitions, and I have to say EUREKA!!!
So there it is, for those following the Brad P. thread - use the Windows HP Utility to do your profiling even if you're on a mac. Stable and solid, which is why I fought so long and hard with HP to reinstate the Windows Utility after they dropped it. I used Windows 8 which is on my machine, with the matching Utility. Any will work fine however. Windows 7/64 has the Utility, as does XP and Windows 10 (now), and they should all be capable of making any size profile you want and they are available for download on the HP website. The Utility is under "Software". Make your charts however you wish, then print them on the Z3200ps and just grin at others using "more superior" printers, HAHAHAHHAHAHAHAHAH!
Best wishes - I'm a happy camper. Hope someone will enjoy using their Z to do the impossible in an hour. A full write-up and tutorial by MHMG and me, coming eventually on Z3200.com
Thanks for the help from all those mentioned and apologies if I missed you.
Special thanks to Geraldo Garcia for helping me out in the first place. Thanks to Argyll - donate NOW!
Mark McCormick Goodhart - You ROCK, man.
Best -
Mark L
-
Actual file and print, using 6000 patch target ICC Profile.
-
Actual file and print, using 6000 patch target ICC Profile.
You might want to add some extra comment to this post, like "this print is a damned good match to screen, only your iphone shots may vary!"
cheers,
Mark
-
I looked at my X-Rite custom target options today and the end point for patches is the 6000 patch target. So this is the standard max number of patches with their workflow.
HP purchased the license to use that workflow and adapt it to the Z. It's very nice they did.
But have you tried side by side evaluations of the 2000 patch set vs this one with 3 times as many patches to see if you are gaining anything at all? It is hard to imagine you could see any advantage. If you can I wonder why all these years no one has described it until now. Just saying, that's a lot of patches and a lot of reading the unless it is obvious you are really gaining something. If you are gaining something substantial I'm all for it and ready to do it myself on all my printers. With the Z thankfully you don't have to do do it by hand as I have to do with my non ISIS equipment.
John
-
That's amazing guys. You have really sparked my interest in revisiting profiling specialized targets my both of my Zs. I've been so busy with Christmas work that Ive only been listening to you guys so far. My plan is to try these expansive targets during the holidays when I'm also going into using Angels workflow for making digital OHP negs for alternative process. That is something I will keep documention on for your site with the various processes I'm going to try out in order to teach a workshop here in April on experimental things.
I am very much also interested in, after hearing all this, in profiling with an extended gray target of big patch size with low gamut and tons of very subtle gray patches to see just exactly how far I can push the Z in the monochrome range.
About 6 years ago a bunch of us got together at my place and shared work, and one guy was making some amazingly rich and clean neutral monochrome prints using an Epson 7800 printer with K3 inks. This was done with cmyk profiling out of Studio Print. I mean you saw absolutely no hit of of any color cast in any of the tonal range. It looked just like a dedicated black and white inkset was used. At the time I never went there because of the limited longevity of Epson's color inks at that time, as well as the difficulty of figuring this out on a 12 ink inkset like the Canon or Hp.
On another topic, is it possible to make extended patch linearization custom targets as well as custom icc profile targets? I doubt that is possible in the hp software but probably not necessary if you are running rgb with extended patch grays.
Has anyone heard of someone having possession of a big grayscale target like that for the icc profile creation to read in one step? One guy from Hong Kong sent me an rgb target that had extended grays along with color patches, about 1775 patches or so. It did give me better shadow detail on the Z through i1 profiler, but for color it wasn't up to my 2033 target for regular color printing but on the z3100 I was using it for black and white .
I still feel like there is more to be squeezed out of the Z in regard to high value dimensionality utilizing the 2 grays, two blacks and light cyan, light magenta, yellow and red in the mix. The beauty of this vivera inkset is the way the hues fade generally at the same rate. How hard would it be to create such a 2000 patch target for that purpose if one doesn't exist? Probably overkill but you wouldn't know if you don't try it. It was certainly possible on the Epson with cmyk profiling.
John
Hi John -
Thanks for that. I think Mark MG is the one to answer that, however, I must say, that with the Z3200ps, presets for profiles are a very large part of how the prints turn out. HP has included certain Quad-Inks for use in printing true gray neutrality with no visible metamerism. In the properties dialog box there is an opportunity to change ink limits and usually it shows a scale of 100 in the middle, meaning that is where the ink limit is set in relation to their arbitrary "mid-point" of 100 with the number relating to the mid point. So experimenting with that, which is what HP recommends can make a big difference. I think it probably is possible to do what your asking - but MHMG would be best at answering that. But look here at the chart I modified (with red) that shows exactly what areas the presets use Quad-Inks for:
Quad-Inks - HP Z3200ps (http://z3200.com/Working_with_other%20Papers_HP-Z3100_z3200_Printers.htm)
The get together you had years ago sounds very interesting.
Please keep us informed with your experiments about Angels workflow for OHP negs.
Thanks again,
Mark L
-
I looked at my X-Rite custom target options today and the end point for patches is the 6000 patch target. So this is the standard max number of patches with their workflow.
HP purchased the license to use that workflow and adapt it to the Z. It's very nice they did.
But have you tried side by side evaluations of the 2000 patch set vs this one with 3 times as many patches to see if you are gaining anything at all? It is hard to imagine you could see any advantage. If you can I wonder why all these years no one has described it until now. Just saying, that's a lot of patches and a lot of reading the unless it is obvious you are really gaining something. If you are gaining something substantial I'm all for it and ready to do it myself on all my printers. With the Z thankfully you don't have to do do it by hand as I have to do with my non ISIS equipment.
John
I think no one has done it really (probably a few have like Aaron Chan, et al), because the process has not been available mostly, nor the instructions on how to get there without APS. So Mark MHMG is planning on testing to a greater degree once he is sure he'll be able to print the target all the way through, then he can measure it and make comparisons. As for how it looks, I have to say I've been printing with 1728 patch target ICC profiles and I can definitely see a visible difference. Whether it's all that much remains to be seen through accurate testing, but I'm guessing Maybe? Who knows.
ML
-
You might want to add some extra comment to this post, like "this print is a damned good match to screen, only your iphone shots may vary!"
cheers,
Mark
YEAH! I-Phone is as i-Phone does. What can you say. The print studio was also a little dark when I took this picture plus it's at an angle.
Excuses, Excuses, LOL. It does match the screen beautifully though. When it's held up with light on it, it's pretty cool...
ML
-
I have attached a graph showing the tone reproduction curve and the grayscale neutrality plot measured with a 60 patch Aardenburg Imaging QC target of Z3200 output on Moab Entrada Natural 300 gsm printed with the 1877 patch count profile I made with BasicColor RGB drop. I printed with the perceptual rendering tag which is only slightly different from RelCol w/BPC.
The smoothness (lack of bumpiness) in the grayscale neutrality is impressive! This with just the 1877 patches, but that i1Profiler generated patch target has over 170 neutral and near neutral patches, approximately half of them being pure neutral RGB triplets, so this amount is quite a few more than what is included in the standard 12x12x12 grid 1728 patch target.
It takes some experience to interpret the grayscale plot which comes from looking at how quite a number of different printing systems behave, including true monochrome print processes where hue/chroma ripple should be next to nothing. To the novice observer, the grayscale linearity plots may therefore seem a bit opaque to understand, but with experience after looking to see how many modern day printer/ink/media combinations handle this grayscale neutrality test, I can say from experience that I’ve only seen as good a result as this, perhaps a little better, from Cone’s K6 and K7 monochrome ink sets and also Digital Silver Imaging prints mastered on traditional B&W fiber-based wet-processed paper. That the Z is getting into true monochrome printing league while still in full color output mode with just an 1877 patch target is truly impressive.
Higher patch counts might squeak out even more smoothness. That’s the value of feeding the profiling engine more neutral/near neutrals to calculate, and that's what happening in those odd count patch sets (1457, 1877, 2371, etc) generated in i1Profiler. Of course, the printer has to be very well behaved to keep that goodness print to print, but the Z3200 seems to be truly up to the challenge.
All that said, until I can get access to the HP windows version of the driver using software like VMWare Fusion on my Mac, I'm a real slacker on patch count compared to Mark L's "mother of all profiles" made with the 6000 patch target set. However, a 4357 patch counts triggers all those extra neutrals in the i1Profiler patch generation algorithm, whereas the 6000 patch count did not, so it may well be that the 4357 is going to take the final prize, and the 6000 patch set will just be for "bragging rights". When I am able to print and measure some of the higher patch count targets on the Z, I will scale the grayscale Neutrality plot even finer so the we can compare the grayscale smoothness results even more closely. The wider +5, to -10 scaling I use in my standard QC plot is there to accommodate output from both yellowish warm tone papers and bluish OBA-filled papers (which often arc down starkly negative in b* values) and also noisier printers as well that show much more bumpiness in the a* and b* curves.
cheers,
Mark
http://www.aardenburg-imaging.com
-
Thanks for that chart Mark. No one has developed a WIKI so thoroughly as you have for the Z. I learn something every time I go there. Hp had done an excellent job of letting you adapt to the papers that YOU want to use with their internal linearizaton and that alone might be their biggest contribution for doing so without a rip. And as we all know, a rip for 12 channels is a nightmare of complexity.
But the deal is papers like Platine, which my clients prefer for neutral bw these day don't use the quad capability since the MK is not usable on them. And that is no issue because using the PK on matte paper is primarily for greater dmax ( not needed on PK Papers ) and does not help in the lighter than middle gray area that is the real realm of dimensional subtlety in monochrome inkjet output. I think that using a patch set designed for neutralized monochrome but taking advantage of the light color inks could add a lot, if the patch set was made correctly. There is always this "good enough" attitude with color engineers when it comes to that monochrome area. They aren't printmakers, for the most part, they are tech people who didn't spend 20 years in the darkroom printing on chlorobromide or platinum papers from large format negatives. There are two ways you can accomplish this, take the green and blue channels out and replace them with lighter gray ink, which involves redoing the software ( or using studio print ) and the other way is to use the light color inks, which i know is possible, with a specialized target and firmware for reading it ( or use X-Rite's own independent software ). The second way doesn't alter your inks for regular color work from the same printer. Of course for the sake of eliminating metameric failure, gray inks are alway better than color, but that changes the character of a color printer, even with 12 slots.
john
Hi John -
Thanks for that. I think Mark MG is the one to answer that, however, I must say, that with the Z3200ps, presets for profiles are a very large part of how the prints turn out. HP has included certain Quad-Inks for use in printing true gray neutrality with no visible metamerism. In the properties dialog box there is an opportunity to change ink limits and usually it shows a scale of 100 in the middle, meaning that is where the ink limit is set in relation to their arbitrary "mid-point" of 100 with the number relating to the mid point. So experimenting with that, which is what HP recommends can make a big difference. I think it probably is possible to do what your asking - but MHMG would be best at answering that. But look here at the chart I modified (with red) that shows exactly what areas the presets use Quad-Inks for:
Quad-Inks - HP Z3200ps (http://z3200.com/Working_with_other%20Papers_HP-Z3100_z3200_Printers.htm)
The get together you had years ago sounds very interesting.
Please keep us informed with your experiments about Angels workflow for OHP negs.
Thanks again,
Mark L
-
YEAH! I-Phone is as i-Phone does. What can you say. The print studio was also a little dark when I took this picture plus it's at an angle.
Excuses, Excuses, LOL. It does match the screen beautifully though. When it's held up with light on it, it's pretty cool...
ML
That iPhone is close enough, better than I what I get shooting prints with my 2006 DSLR shooting Raw where I have to apply some extensive edits to get a match. I'm assuming the light used is D50 daylight like the Solux or something similar?
What I'm curious about since I had a client way back when I thought I'ld have a career as a consultant who ended up going with the Z3200ps upgrade with on board spectro is a close up shot of a section of the test print that gave you a lot of trouble before the 6000 patch based profile and the after.
It's hard to tell from your posted image what was improved. The print looks great to me.
-
Thank you Mark,
That is what I need to know at this juncture ( before I start to think about putting diluted grays in my Z3100 :) ) which targets give you more dimensionality. Of course "smoothness" and dimensionality are two different, but related things. Cone's inks and Silver and Platinum printing give you a sculptural capability in the high values that I've never come close to seeing from the Z, and boy have I tried with it and the Canon TBW, both of which are very nice but not the same level of fine highlight tonality. Honestly I'm not that far off and it's possible a perfect target could work for me. I'm talking about especially from drum scans of my 4x5 film and very carefully lit still life work that I do. Cone's K7 Neutral inks on the Canson Rag Photo have show on Marks tests to fair better than the Hahnemuhle fade tests I sent in years ago, before they go green, but sadly they are still way behind Vivera ( and Canon ) neutralized grays when shown continuously in daylight. The K7 Carbon Sepia is great but I just don't have many clients who want prints that warm, though I like them quite a bit. The gloss enhancer ease is just a dream on the Z as well with none of this gloss differiental and bronzing garbage.
I am going to make some tests using these patch sets you just mentioned. It is possible that making a "custom" monochrome patch set is not necessary, and that X-Rite has already optimized the values and print color in some of these patch sets. So I'm going to spend some time with the 1877 and 4357 patch sets during the holidays with Platine. I think I'll learn something. I will report back in January what I'm seeing.
John
The smoothness (lack of bumpiness) in the grayscale neutrality is impressive! This with just the 1877 patches, but that i1Profiler generated patch target has over 170 neutral and near neutral patches, approximately half of them being pure neutral RGB triplets, so this amount is quite a few more than what is included in the standard 12x12x12 grid 1728 patch target.
However, a 4357 patch counts triggers all those extra neutrals in the i1Profiler patch generation algorithm, whereas the 6000 patch count did not, so it may well be that the 4357 is going to take the final prize, and the 6000 patch set will just be for "bragging rights".
-
But have you tried side by side evaluations of the 2000 patch set vs this one with 3 times as many patches to see if you are gaining anything at all? It is hard to imagine you could see any advantage. If you can I wonder why all these years no one has described it until now. Just saying, that's a lot of patches and a lot of reading the unless it is obvious you are really gaining something. If you are gaining something substantial I'm all for it and ready to do it myself on all my printers. With the Z thankfully you don't have to do do it by hand as I have to do with my non ISIS equipment.
John
Hi John, Mark L. and I have spent the last few days trying to iron out a few things that perhaps other Z users have done before. First, can one insert a custom patch set and get the Z to recognize it, lay it out according to paper/roll size selected, and print and measure it without choking ;D A subset of the work was, could we find those i1Profiler generated patch counts where neutrals and near neutrals would be maximized and would that help with profile accuracy? To that end, Mark and I have been gradually printing bigger and bigger patch counts, but along the way we both felt like "let's now be buccaneers and try to print the biggest target set the i1Profiler software can make and save to a CGATs file the HP z3200 can read!" That's how we got to the 6000 patch experiment, and Mark L. saved the day because my Mac driver is choking at about 2000 patches. I will probably have to move to the windows version of the driver to resolve the bug that causes the Z to think it has finished printing the target and now moves on to the drying/measuring steps when in fact the target did not get fully printed.
Now to the more overarching question.. Does it matter if one can print so many patches. Is it worth it? When does diminishing returns kick in, e..g at 2000 patches or 3000, or 6000? That's a larger study which will take more time, but I'm particularly keen to do that study because like many other printmakers, I take the attitude that maximum print quality is the holy grail of great printmaking. What Mark L. and I have learned so far is that those oddly configured patch count Xrite targets which put a higher percentage of more neutrals/near neutral RGB triplets into the target set do indeed improve the quality of smaller target sets e.g., 1457 to 1877 total patch count, compared to the traditional 12x12x12 linear 1728 patch set included with the Z3200 driver.
Another common consensus in the industry regarding profile quality is that higher and higher patch counts can actually degrade profile quality when used to profile noisy or not very repeatable printing systems where you are better off just using a smaller set and averaging a number of measurements to gain best average daily output. Hence, Mark L.'s print quality he got using the 6000 patch set demonstrates convincingly that the Z3200 is up to the system output precision and repeatability task such that performance does not go down no matter how many patches (up to 6000) you throw at it to print. Of course, the profiling engine comes into play as well. Newer algorithms do automatic smoothing and outlier detection. Anyway, whether diminishing returns occurs at 1877, or 2371, or 4357, or all the way up to 6000 remains to be determined. It's a great question, and no doubt the answer will be Z3200 specific, not a catch all for every printer on the market today.
cheers,
Mark
http://www.aardenburg-imaging.com
-
Thanks for that chart Mark. No one has developed a WIKI so thoroughly as you have for the Z. I learn something every time I go there. Hp had done an excellent job of letting you adapt to the papers that YOU want to use with their internal linearizaton and that alone might be their biggest contribution for doing so without a rip. And as we all know, a rip for 12 channels is a nightmare of complexity.
But the deal is papers like Platine, which my clients prefer for neutral bw these day don't use the quad capability since the MK is not usable on them. And that is no issue because using the PK on matte paper is primarily for greater dmax ( not needed on PK Papers ) and does not help in the lighter than middle gray area that is the real realm of dimensional subtlety in monochrome inkjet output. I think that using a patch set designed for neutralized monochrome but taking advantage of the light color inks could add a lot...
john
John I’ve been giving this some thought. Looking at the chart that shows Quad ink applications, it occurs to me that possibly there might be one or two avenues open to experimentatition if your open to it with Platine. Looking at the chart, one of the areas I have not experimented with is the “Back-Lit films”.
The fact that these substrrates are essentially smooth and shiny, yet do accept Quad Inks, makes me wonder if maybe some experimentation with Platine might be in order. I’d start with the plain vanilla 464 patch target the Z makes automatically and fiddle with the print characteristics by comparing what the settings are for presets for the pk inks normally used with Platine. A deal breaker could be that you might not be able to turn GE on, but you could check to see if it’s possible in the setup menu of the web browser if not in change print properties. Also, I have for a longtime thought one should be able to run an image back through while fooling the printer thinking it is printing a blank page using a full page Gloss Enhacer setting thereby coating the previously printed image whose profile would not allow GE. So if you are game for some experimenting, this may be one way to use the Quad inkset with Platine. Would however require experimentation. Had to mention this and I may take a shot at it myself one day.
Best,
ML
-
The advantage of using the Z to measure these huge targets, or possibly any target, is that it should be more precise than a human doing the readings a line at a time by hand, in my case without automated reading of the ISIS. This might very well eliminate the need to do multiple readings, or dual scan lines, and average the results. In a way, doing such a huge number of readings might be like doing it's own "averaging" process with a certain amount of built in redundancy, maybe. The proof will be in the prints.
-
The idea of a user created custom media setting is interesting, however, I think you are right in that the gloss enhancer is going to have trouble dealing with the MK ink ( not to mention the paper surface itself.) But anything is possible. The one thing I won't do, which is a bridge too far for me, is doing a dual pass gloss enhancer workflow. I've been there and done that with Piezography dual pass glop and for me it was a mess in keeping the print clean from debris, not to mention the time involved... drying the print twice etc, etc. Even Cone's new dual quad inksets do a single pass gloss overcoat now. I think that is the only way to go practically and HPs current workflow is so beautifully clean and precise, and they were the first to do it.
John I’ve been giving this some thought. Looking at the chart that shows Quad ink applications, it occurs to me that possibly there might be one or two avenues open to experimentatition if your open to it with Platine. Looking at the chart, one of the areas I have not experimented with is the “Back-Lit films”.
The fact that these substrrates are essentially smooth and shiny, yet do accept Quad Inks, makes me wonder if maybe some experimentation with Platine might be in order. I’d start with the plain vanilla 464 patch target the Z makes automatically and fiddle with the print characteristics by comparing what the settings are for presets for the pk inks normally used with Platine. A deal breaker could be that you might not be able to turn GE on, but you could check to see if it’s possible in the setup menu of the web browser if not in change print properties. Also, I have for a longtime thought one should be able to run an image back through while fooling the printer thinking it is printing a blank page using a full page Gloss Enhacer setting thereby coating the previously printed image whose profile would not allow GE. So if you are game for some experimenting, this may be one way to use the Quad inkset with Platine. Would however require experimentation. Had to mention this and I may take a shot at it myself one day.
Best,
ML
-
I don’t know it won’t do it in one go yet John, I’d have to look at it very carefully.
Point taken, however.
Mark
-
Thank you Mark,
That is what I need to know at this juncture ( before I start to think about putting diluted grays in my Z3100 :) ) which targets give you more dimensionality. Of course "smoothness" and dimensionality are two different, but related things. Cone's inks and Silver and Platinum printing give you a sculptural capability in the high values that I've never come close to seeing from the Z, and boy have I tried with it and the Canon TBW, both of which are very nice but not the same level of fine highlight tonality. Honestly I'm not that far off and it's possible a perfect target could work for me. I'm talking about especially from drum scans of my 4x5 film and very carefully lit still life work that I do. ...
Hi John, I've always felt that imaging science lags a little behind with respect to the image quality characterization of those vintage processes that just deliver a little something extra visually which is hard to quantify, e.g., dimensionality as you are referring to it, and I think I know what you mean by smoothness versus dimensionality. It's like when looking at a large format negative beautifully contact printed onto traditional wet process paper. That film/paper combination can record over 30 line pairs per mm of resolution on the print surface all the way into the highlights yet the human eye is only able to differentiate about 6 to 10 lines pairs per mm at any reasonable viewing distance. Nevertheless, this "frequency oversampling" somehow is processed by discerning viewers as extra fine detail and tonal separation, "buttery" smoothness as well to use Mark L.'s words to describe that something extra. It's a fact that the 6 and 7 channel Cone monochrome sets are clear winners in dot structure patterning using inkjet printing technology in those light midtones all the way into the highlights, and wet process silver is, of course, even more finely divided in "dot" structure than even Cone's Piezography as long as the film negative is not enlarged so much as to dominate the grain appearance in the print. This superior "oversampled" fine grain structure in the highlights, even though most viewers can't see obvious grain at normal viewing distances in most inkjet prints made nowadays, I think is part and parcel of what you are referring to as "dimensionality" rather than smoothness in the final print. In this respect, the Z's inherent dot structure may always hold it back from competing at such lofty levels. Then again, I recall that the Iris 3047 had a distinct screening pattern which was incredibly beautiful even though close inspection visually could indeed make out relatively coarse dots. Bottom line for me is that the ultimate dimensionality may always come from analog processes like platinum and silver contact prints, yet superior control over tone reproduction curves and print surface variety can allow inkjet prints to compete in other ways for high praise from discerning viewers.
My goal is just to try to "make my Z be the best it can be" (hey, maybe I should apply for a marketing job :) ). I think it can give other traditional analog monochrome processes a run for their money, but it's probably never going to match them head to head for every conceivable image quality property. It will be better in some respect and not in others. Trick is always to find sympathetic images that can become magic in the viewer's eyes when printed with any specific printing process.
All the best,
Mark
http://www.aardenburg-imaging.com
-
The advantage of using the Z to measure these huge targets, or possibly any target, is that it should be more precise than a human doing the readings a line at a time by hand, in my case without automated reading of the ISIS. This might very well eliminate the need to do multiple readings, or dual scan lines, and average the results. In a way, doing such a huge number of readings might be like doing it's own "averaging" process with a certain amount of built in redundancy, maybe. The proof will be in the prints.
It wouldn't surprise me a bit if the Z doesn't produce more repeatable measurements than the Isis as good as the Isis is. The Isis has wheels every 21mm to grab and feed the paper and it changes slightly the spectro readings and leaves visible tracks depending on the paper type.
I'm actually quite curious about whether the Z has less variance compared to the Isis.
-
It is more than a little ironic that here in the year of our lord 2017, 17 years after I started to work with monochrome digital imaging in a serious way that we are still discussing ways to render the tonal depth of Ansel Adams and Man Rays work ( I had a print of his in my studio a couple of years ago ) from the 1930s.
Having said that, I wouldn't want to revisit the chemical era without my digital tools and versatile pigments. I have a client from New Mexico who has been a master platinum/ palladium printer for over 30 years and when I took his file, added my scanned palladium edges to it and made him 30x40 editions using k7 carbon he was shocked to the point that he doesn't do any more big palladium prints at all. He had a chemical technique for creating outstanding dmax, though for warm palladium prints only, and my k7 prints match his work almost exactly. Why a more stable magenta has not been formulated for k7 neutral I will never know. But I'm not going to grow old waiting for it, I'm going to have to develop something myself I guess if it is at all possible. There are just not enough hours in the day.
Hi John, I've always felt that imaging science lags a little behind with respect to the image quality characterization of those vintage processes that just deliver a little something extra visually which is hard to quantify, e.g., dimensionality as you are referring to it, and I think I know what you mean by smoothness versus dimensionality. It's like when looking at a large format negative beautifully contact printed onto traditional wet process paper. That film/paper combination can record over 30 line pairs per mm of resolution on the print surface all the way into the highlights yet the human eye is only able to differentiate about 6 to 10 lines pairs per mm at any reasonable viewing distance. Nevertheless, this "frequency oversampling" somehow is processed by discerning viewers as extra fine detail and tonal separation, "buttery" smoothness as well to use Mark L.'s words to describe that something extra. It's a fact that the 6 and 7 channel Cone monochrome sets are clear winners in dot structure patterning using inkjet printing technology in those light midtones all the way into the highlights, and wet process silver is, of course, even more finely divided in "dot" structure than even Cone's Piezography as long as the film negative is not enlarged so much as to dominate the grain appearance in the print. This superior "oversampled" fine grain structure in the highlights, even though most viewers can't see obvious grain at normal viewing distances in most inkjet prints made nowadays, I think is part and parcel of what you are referring to as "dimensionality" rather than smoothness in the final print. In this respect, the Z's inherent dot structure may always hold it back from competing at such lofty levels. Then again, I recall that the Iris 3047 had a distinct screening pattern which was incredibly beautiful even though close inspection visually could indeed make out relatively coarse dots. Bottom line for me is that the ultimate dimensionality may always come from analog processes like platinum and silver contact prints, yet superior control over tone reproduction curves and print surface variety can allow inkjet prints to compete in other ways for high praise from discerning viewers.
My goal is just to try to "make my Z be the best it can be" (hey, maybe I should apply for a marketing job :) ). I think it can give other traditional analog monochrome processes a run for their money, but it's probably never going to match them head to head for every conceivable image quality property. It will be better in some respect and not in others. Trick is always to find sympathetic images that can become magic in the viewer's eyes when printed with any specific printing process.
All the best,
Mark
http://www.aardenburg-imaging.com
-
That is a very good point. The profiling capability of the Z may just be getting started thanks to these guys. I'm very interested in what can be done, with these auto measured big targets. There are so many things this printer can do and HP never marketed it. Their sales-tech guys are MIA.
It wouldn't surprise me a bit if the Z doesn't produce more repeatable measurements than the Isis as good as the Isis is. The Isis has wheels every 21mm to grab and feed the paper and it changes slightly the spectro readings and leaves visible tracks depending on the paper type.
I'm actually quite curious about whether the Z has less variance compared to the Isis.
-
...Why a more stable magenta has not been formulated for k7 neutral I will never know...
Probably because it's not common knowledge that there's a color shift fading issue due to the addition of a less light fast magenta pigment blended in with the carbon pigment to achieve the colder toned Piezography ink sets.
That said, the Piezography neutral shade lightfastness issues pale in comparison to all the "archival pigment print" nonsense being played out in art galleries all around the world nowadays. Take any pigment inkjet printer, use any media you want, loaded with OBAs or not, and call it an "archival pigment print". Market resistance now apparently resolved. "Archival pigment" appears to be an incredibly reassuring combination of words to the buyer even though there is no formal definition for what it means!
-
It wouldn't surprise me a bit if the Z doesn't produce more repeatable measurements than the Isis as good as the Isis is. The Isis has wheels every 21mm to grab and feed the paper and it changes slightly the spectro readings and leaves visible tracks depending on the paper type.
I'm actually quite curious about whether the Z has less variance compared to the Isis.
I wouldn't count on it Doug. Referring to Geraldos thread, Samuel Johnchia had this to say:
Hi Geraldo,
Thank you for the information. I assume that the first reading pass was after sufficient dry down time, not right after the chart was printed? Then the several hours between readings should not matter enough. It would be good to know that the dE differences is almost entirely due to the instrument's limitation, and we are not looking at dry down changing the dE also.
Otherwise the repeatability is about the same as what one can get with careful hand scanning with the i1 Pro II, in my tests. The repeatability is limited by the measuring device in this case - the same i1 Pro base technology. However it is less likely for the automated Z3200 system to make errors due to hand slippage or inconsistent or too fast/slow movement, so human error is minimized. Occasionally I catch an error of about 1 dE2000 when hand scanning in the past. Sometimes a freak accident is much worse. It was good to see confirmation from Damir that it is possible to hack the software to scan any number of patches. If I had a Z3200 and no iSis I would use it for scanning too.
The iSis repeatability is on a whole different level. Here is a comparison of scanning a (more challenging) 2808 patch target. The max dE difference is more than an order of magnitude better and the same could almost be said of the average dE:
Number of Samples: 2808
Delta-E Formula dE2000
Overall - (2808 colors)
--------------------------------------------------
Average dE: 0.01
Max dE: 0.03
Min dE: 0.00
StdDev dE: 0.00
Best 90% - (2526 colors)
--------------------------------------------------
Average dE: 0.01
Max dE: 0.01
Min dE: 0.00
StdDev dE: 0.00
Worst 10% - (282 colors)
--------------------------------------------------
Average dE: 0.02
Max dE: 0.03
Min dE: 0.01
StdDev dE: 0.00
If you go back to that thread, there's a lot of info re: what the Z3200ps can do, as Geraldo made several tests:
http://forum.luminous-landscape.com/index.php?topic=114961.msg948233#msg948233
The HP Z3200ps is really good at what it does, but it's not an i1Sis the larger of the two of which costs more than the entire Z3200ps INCLUDING the spectrophotometer. It's important to acknowledge that what it does is indeed awesome, but not overstate what it is, I think.
Pushing it to it's very edge and it's limits will tell, but it's simply an embedded spectrophotometer that produces at this point in my mind world class results. That's plenty good for me, and for the money, it is absolutely incredible. Geraldo went on to clarify what the Z could do under careful scrutiny and that's worth looking at. Have to say thanks again to Geraldo for taking us down this path.
Best,
Mark
-
You got that right. I may be setting up the neutrals in a 24 inch printer just because I miss them so much. If they are all n limited edition portfolios or exhibited frequently they could last a generation or more. Just wish they had the Epson magenta mixed in instead of what the have.
I realized a way to convey the "dimensionality" aspect I was describing, it is like Leonardo's glazes in oil painting, a gradual process of building up of tonality lighter lighter lighter that wraps around lit objects. Jon deserves a lot of credit for being able to move it that far. As for modern Ilford silver prints , they look dead to me compared to what we used to do with more silver in those old papers. If you make your own silver chloride paper you could it though I assume and print from digital negs....
Probably because it's not common knowledge that there's a color shift fading issue due to the addition of a less light fast magenta pigment blended in with the carbon pigment to achieve the colder toned Piezography ink sets.
That said, the Piezography neutral shade lightfastness issues pale in comparison to all the "archival pigment print" nonsense being played out in art galleries all around the world nowadays. Take any pigment inkjet printer, use any media you want, loaded with OBAs or not, and call it an "archival pigment print". Market resistance now apparently resolved. "Archival pigment" appears to be an incredibly reassuring combination of words to the buyer even though there is no formal definition for what it means!
-
I'd like to thank all of you involved in working out this new profiling solution. I've been following along (the best I could) from the beginning of Brad's post. I'm new to the Z and so far have been very happy with it's output with the exception of the B&W linearity.
It wasn't a problem for the majority of my work but I have a few "tough" images that were showing stepping. Thanks to you all, problem solved! I printed a 2033 patch target and ran it through Argyll CMS, worked like a dream. I'll be donating soon!!!
Just a few things I found that may be specific to my set up but maybe not. When printing or scanning a target patch, don't let your computer go to sleep. It breaks communication between Utility and the printer and in my case could not get it reestablished. In fact it caused
a Runtime error and I had to reboot the printer to start the process over. Because of my all too short attention span I did this twice (I then remembered to turn off sleep). :o
One more thing I discovered. I've been running a Canon for years and it would always stop printing when I was stupid enough to try to print a job without enough left on the roll. The Z did not :( and made a mess of itself. Is this normal? Do I have a problem with my printer? I know I should pay closer attention but really thought I had enough paper to complete the job. I may have been a bit too exited make the profile as well. ::)
Once again, thank you all so much. I would not have figured this out on my own. I am now making the best prints I've ever made, at least to my eye. This was like an early Christmas present!
Mike
-
Mike - in many cases, if the paper is at the end of the roll, it is glued on and can't let go. In that case, the printer may continue printing on the same line until the print is done. So if it has run out of paper and just kept printing in thin air, I wouldn't know what to say. Best thing to do is clean it up the best you can and keep printing. You should be fine.
Glad to hear you're liking your Z and making great prints. This should be the beginning of a great partnership with that printer.
Best,
Mark
*EDIT - By the way, you should always create the paper, profile it and do advance paper calibration. It's a good idea to do this because once in the printer's list of custom papers, it will remind you when to do re-calibrations. Also, it keeps everything organized and tidy. But as you saw, it can be done otherwise as well - it's just that the Z does such a good job of organizing, why would you not want to let it do it?
-
Mike - in many cases, if the paper is at the end of the roll, it is glued on and can't let go. In that case, the printer may continue printing on the same line until the print is done.
Yeah, this is exactly what happened. I made the mess when I removed the paper - some of the ink on that way over printed line got on the rollers. The 6300 would cancel the job when it felt the resistance of the taped roll.
As far as this being the beginning of a great partnership, I agree. I only wish I went this route years ago. Thanks again Mark for all your effort you put into the Z website, it's such a great resource.
*EDIT - By the way, you should always create the paper, profile it and do advance paper calibration. It's a good idea to do this because once in the printer's list of custom papers, it will remind you when to do re-calibrations. Also, it keeps everything organized and tidy. But as you saw, it can be done otherwise as well - it's just that the Z does such a good job of organizing, why would you not want to let it do it?
This is exactly what I'm doing Mark and I agree it's a great system. What did I say to make you think otherwise?
I'm looking forward to hear Mark's (MHMG) analysis of the different patch counts. I went with 2033 because it had the extended grays and I like the number. I'm not all that scientific :) I would like to know at what number the patch counts start to yield little improvement, if that's even the case. I do know that I'm really happy with what I got out of 2033.
-
As far as this being the beginning of a great partnership, I agree. I only wish I went this route years ago. Thanks again Mark for all your effort you put into the Z website, it's such a great resource.
This is exactly what I'm doing Mark and I agree it's a great system. What did I say to make you think otherwise?
. I may have been a bit too exited make the profile as well.
See above Mike - I may have misunderestood you. Anyway, all is good - thanks for the kind words.
We’ll be sure to post results.
Mark
-
Hey Mark, here's something that could be helpful. From the old post. Didn't want to clog up the new one.
http://forum.luminous-landscape.com/index.php?topic=121939.msg1016004#new
-
Hey Mark, here's something that could be helpful. From the old post. Didn't want to clog up the new one.
http://forum.luminous-landscape.com/index.php?topic=121939.msg1016004#new
Hi Brad, just to confirm, yes this the exact work flow MHMG and I have been discussing and Mark MHMG has been working on. So yes, you have the correct work around.
As far as adding the patch target size file in under the list, clicking the “+” sign is the correct method, and then using the drop-down menu, the new target will show up. Install profile is the way to go, as Geraldo pointed out in the original thread in order to get the newly generated /ICC file included in the custom paper list. Thanks for listing the steps. We’ve been doing this, but haven’t posted it yet, wanting to finalize our test results. I ran into a glitch with the Windows HP Utility, where although it would print and measure the 44” wide 6000 patch target, it failed to print and measure the 4357 target on 24” Wide Moab Entrada.
After much screwing around, I put a 6 ft. Long table in behing the printer, cut the roll into a sheet that was about 5+ feet, and using load paper with sckew check, was able to get the spectro to read the chart using the very same workflow of “measure a previously printed chart”. MHMG is testing the buginess of printing tiff files with Apple Color Sync, etc., and we are continuing to carefully plod through various iterations until we know for sure what bugs are and aren’t in both the Mac and Windows HP Utilities before publishing the results. Your list of steps will come in handy when we do come to that point. Very good. It also confirms our results.
Mark