August 2013 - colorsync-users

Re: i1Profiler linearization
by Mike Strickler 07 Aug '13

07 Aug '13

It's not so clear at all. It's really hard to analyze your problem from afar; there are too many variables: Different RIP version, old profile, possible change in media, and of course the likelihood of printer drift. Go back to media manager and print the charts. If you still have smooth ramps, no lowering of density or shouldering of chroma, just make a new profile. You need one anyway. Version 10 is very much improved software, by the way. You'll find much of the business of ink limiting and linearzing is handled for you. Highly recommended. Mike Strickler MSP Graphic Services > > Message: 1 > Date: Wed, 07 Aug 2013 07:35:55 -0500 > From: Scott Martin <scott(a)on-sight.com> > To: David Wollmann <david(a)lenzart.us>, > "colorsync-users(a)lists.apple.com List" > <colorsync-users(a)lists.apple.com> > Subject: Re: i1Profiler linearization > Message-ID: <234FBB78-E916-41BB-9722-070C189BB6B9(a)on-sight.com> > Content-Type: text/plain; charset=us-ascii > > >> The original testing and ink limiting for the media was done using Onyx 6.5, that media library was imported into Onyx 7.3.2. It should have brought the linearization over in the import. > > > That explains it! You just can't get a good profile on a printer like that without starting with a good lin. You might also double check that the other settings carried over as well. > > If you feel up to the task, it might be smart and start the whole ONYX Media Manager "Create new profile" over from scratch and go through the whole process to make sure you're optimizing everything from the start. On a solvent printer the results are so great when everything is perfectly dialed in. Unfortunately with a lot of the supplied profiles that's not the case. > > And BTW you can still use i1Proifler for the final profile. > > Scott Martin > www.on-sight.com > >

1 0

Re: RGB Drivers in RIPs
by tlabarbera 07 Aug '13

07 Aug '13

>>From: Ernst Dinkla <E.Dinkla(a)onsneteindhoven.nl>: The technology of the 1270, the dye ink, the media, the color management and the applications to print from, even the camera then; one would not be satisfied with any part of that now. I started with an Epson 700 later moving to an 1160 (refurbished) and was pleased with the output on both--other than of course dye inks which were pretty much the only option--using rudimentary (by current standard) CMS. Others, like my friend, were less than pleased with their output. I currently run an Epson 4000 and again, have been pleased with the output. >>True, there can be a rotten apple in the pie today too. Oh definitely, but I think in many cases that might be due more to a defective/poorly built individual printer because I think the printer manufacturers have come a long way since the 1270 days--how could they not...'-}} Terrie http://tlbtlb.com/ tlbtlb(a)mail.com

1 0

Re: Colorsync-users Digest, Vol 10, Issue 183
by Millers' Photography L.L.C. 07 Aug '13

07 Aug '13

Things are a bit better now, but there are still plenty of driver niggles such as roll paper settings not sticking in 10.8 and various utility screw ups. In my case with the Pro 9900, Epson driver, roll paper setting sticks when I set for sheet. Hmmm. Do wish I knew how to prevent this. Cheers David David B Miller, Pharm. D. member Millers' Photography L.L.C. dba Spinnaker Photo Imaging Center Bellingham, WA www.spinnakerphotoimagingcenter.com 360 739 2826 On Aug 7, 2013, at 4:24 AM, colorsync-users-request(a)lists.apple.com wrote: > > From: Martin Orpen <martin(a)idea-digital.com> > Subject: Re: RGB Drivers in RIPs > Date: August 7, 2013 12:35:44 AM PDT > To: "'colorsync-users?lists.apple.com' List" <colorsync-users(a)lists.apple.com> > > > On 7 Aug 2013, at 00:25, Spinnaker Photo Imaging Center <spinnakerphotoimagingcenter(a)dnmillerphoto.com> wrote: > >> What does " place a rectangle in the middle of a page" mean? > > > I mean that the best selling point of a RIP for me personally is being freed from the nightmare of OS and driver screw ups -- of which there have been so many. A lot of fine art printers have to maintain a Mac museum setup - CS4 Photoshop, 10.6 and an old Epson driver because changing any of this production chain is too painful. > > Things are a bit better now, but there are still plenty of driver niggles such as roll paper settings not sticking in 10.8 and various utility screw ups. > > For example, why can't X-Rite's SpectraProofer Utility linearise Velvet Fine Art media on a 9900? It can do it on a 4900 but not a 9900. > > Look under the hood on these products and you'll find some ugly bits of programming lashed into Frankensoftware that loads badly, behaves badly and is guaranteed to break with every OS or dependent app upgrade. > > Using a RIP means that you get to send all your bug reports to one place: the GMG Help System, Have just one tech support person tell you that "nobody else has reported this problem". And then smile when the sales team sell the bug fix as a feature when they next visit. > > It was just over a year ago that I posted here about 100c & 80y issues with GMG and Epson's x900. Have you seen GMG's brand new x900 Quality Check test page? > > There are a lot of 80y patches on it…. :-) > > -- > Martin Orpen > Idea Digital Imaging Ltd >

1 0

Re: i1Profiler linearization
by David Wollmann 07 Aug '13

07 Aug '13

Mike, thank you for your comments, they have helped chase out the cobwebs of my own memory. I was living in a pipe dream where I thought for all to brief a time that could do everything on my Mac. The original testing and ink limiting for the media was done using Onyx 6.5, that media library was imported into Onyx 7.3.2. It should have brought the linearization over in the import. This had all been imported a few months ago. Everything had been working and color was good on this production machine. I do need to work in between print runs. A head was changed on the printer. It was time to pull things back in line and tighten the profile. This was also a new install on a "newer" machine lacking some of the old ports such as serial for a DTP41. Using the i1Pro for the first time on the computer. Onyx would go through the setup but would fail at calibration. I finally found a download from Xrite for the i1Pro with the drivers for XP, even though Onyx shows the i1Pro in its list. I can now calibrate but I am limited to doing spot readings of the new Lin. target. No matter what setting I choose it will only let me spot read the 217 patches of the Lin. target. Not precise at all. The output would be considered low res, thus I was originally looking for a way to average the data. Now that I'm back in Onyx I can set the number of readings there very easily, but thats where easy ends. My first attempt at spot reading gave me a set of numbers with one bad patch, reading spiking too low. Onyx will not let me back in to make one new reading to correct the patch, not after you say OK. I had to start all over again. Now I have better numbers but the magenta tries to be a bit of a saw tooth in the 3/4 tones. This is effecting smooth gradients which were fine with the old setup - Linearize in Onyx but build the profile with Monaco - the old version of the RIP did not support building ICC profiles, just calibration Now updating to i1Profiler. I will need to spot read the chart again. Does anyone know how to get the i1Pro to read all patches in row with Onyx 7.3 and XP? Thank you, David > Perhaps my attention span is deficient and I've just missed something, but what is the actual objective here? To produce a better profile in the first place by improving the linearity of the printer output or returning output to a baseline via recalibration? The two may be treated together though in most instances with inkjet printers they are distinct concerns. As Terry points out, the linearization in i1P has nothing to do with calibrating devices but is designed to assist in the profile creation process. I don't think this addresses the original question. > > If the only concern is making a good profile, linearization does not need to be perfect by any means: Pretty good is good enough, and the profile should straighten out any remaining kinks. It should go without saying that great attention should be paid in RIPs to ink limiting, as this directly impacts profile quality. Scott may overstate the credentials needed for this, but it's best left to those who do it regularly and have had a bit of training especially if the materials are difficult to ptint on. > > Recalibrating color that has drifted is a far more interesting subject, and there are various solutions. They fall roughly into three categories: pre-profile, reprofile, and post-profile.Typical "relinearize" routines are in the first category. They treat the unprofiled output of each color channel. The idea is that inkjet printers drift in only one way: printheads deliver more or less color in each channel; correct the individual color channels back to a baseline and all colors will fall into place--the profile will be revalidated. IN some cases, the ink limits are readjusted as well. This is sometimes called a 2-D correction, as it does not directly treat the aggregate result. The shortcoming is that it isn't accurate enough to correct slight CMY gray balance variations that can still be objectionable. It works well enough for larger corrections. A variation on this is a 3D approach that also predicts the resulting CMY gray balance of the suggested channel adjustments. Onyx appears to have implemented this approach, which is well established in the offset press world (Curve3, Bodoni PressSign, Alwan Print standardizer, et al). > > The reprofiling approach is self-explanatory. It should be noted that if solid ink values have dropped below an acceptable level because of clogged heads or other issues this must first be addressed as the new profile cannot correct this. Reprofiling is time-consuming and not generally used in the heat of production. > > Finally, there are the recalibrations based on measurement and correction of profiled output, another "3D" approach. These can be in the form of a Lab correction LUT inserted before the output profile or a device link. This is a lot like reprofiling but can preserve the precision of the original profile and gently nudge all the data points with smooth curves. Very small charts may be used, so this approach can be used in the midst of a job. I personally favor this approach for inkjet printing. It is highly accurate and fast. > > Mike Strickler > MSP Graphic Services

2 1

Re: RGB Drivers in RIPs
by Robert Crow 07 Aug '13

07 Aug '13

... ">A corollary of what we've been saying here is that "Printer Manages Colors" aka Vendor Color works amazingly well for photo printing if you're using vendor media and vendor drivers. " ...personally I have not been bold enough to hit the 'printer manages' button yet! We are still talking about a colour managed workflow using custom profiles (ok - occasionally the canned profile) via Application managed colour and the printer vendors OS/RGB driver. In other words what we get when we print carefully through photoshop and the Epson OS driver with our profiles. Rips like Mirage and Overdrive use the same route - but with a far better front end and I've been told, off list, that some other RIPs using RGB drivers use the printer manufacturer's more adjustable SDK version of the driver rather than the OS - I need to test those becuase in the past I found some which claimed RGB drivers still gave less pleasing results to the OS driver The issues I've had with some rips was not the colour per-se, but the screening and break up in the gradients. It is an issue driven partly by colour though in my opinion... ...for example, if you print a pure blue gradient via the Epson canned profile for their Premium Semigloss paper, it is evident (to me at least) that some 'purity' of colour has been sacrificed in order to achieve a smoother gradient (versus my custom 'colour accurate' profile). Perhaps this is where photographers are at odds with the graphics world? i.e smooth transitions of colour and tone vs absolute colour accuracy? After all, a toned (or duotone) image is often favoured over a purely neutral one in for this reason. RIP manufacturers naturally concentrate on colour (often in blocks) hitting precise Lab values whereas the inkjet manufacturers know that photographers are interested in balancing colour against a more subjectively pleasing smoothness. We now print lots of fine art photography and lots of short order graphics work, so I need the best of both worlds! I might have to run two types of workflow but it is the more complex nature of PDF's, Fonts, imposition etc that is driving the need for automation and a production rip. Thanks for all the suggestions so far - I have some good leads to follow up and always interested in hearing more. Rob ------------------------------ >Message: 10 >Date: Wed, 07 Aug 2013 03:40:07 +0200 >From: edmund ronald <edmundronald(a)gmail.com> >To: Spinnaker Photo Imaging Center > <spinnakerphotoimagingcenter(a)dnmillerphoto.com> >Cc: "'colorsync-users?lists.apple.com' List" > <colorsync-users(a)lists.apple.com> >Subject: Re: RGB Drivers in RIPs >Message-ID: > <CAJe_Magm21h_JL0NHRsyo2VDL5o5+S-k7ERUmPrX2tSYay4rmw(a)mail.gmail.com> >Content-Type: text/plain; charset=windows-1252. > >A corollary of what we've been saying here is that "Printer Manages >Colors" aka Vendor Color works amazingly well for photo printing if >you're using vendor media and vendor drivers. Of course if you use a >RIP you have much better layout facilities; but then I've never looked >at the software that came with the printer, maybe it can help there. > >I think there was a huge change about 2 generations ago, and now the >big Epsons are very stable and have good drivers and good profiles; >this is a surprise to those of us who owned the previous generations >and smaller inkjets. >I used to run a 9600, and now have an Epson 3880 sitting on my shelf, >and there is simply nothing bad I can say about it apart from the ink >costs. > >Edmund ________________________________________ --------------------------------------------------------------------------- Please note that Middlesex University's preferred way of receiving all correspondence is via email in line with our Environmental Policy. All incoming post to Middlesex University is opened and scanned by our digital document handler, CDS, and then emailed to the recipient. If you do not want your correspondence to Middlesex University processed in this way please email the recipient directly. Parcels, couriered items and recorded delivery items will not be opened or scanned by CDS. There are items which are "exceptions" which will be opened by CDS but will not be scanned a full list of these can be obtained by contacting the University.

2 1

Re: Colorsync-users Digest, Vol 10, Issue 182
by Mike Strickler 07 Aug '13

07 Aug '13

Well, it's not quite that simple. While it's true (and I've already stated this at least twice) that manufacturer's drivers produce excellent color and screening, RIPs that are designed for press proofing have features specifically designed for producing and maintaining the tightest possible match to the target reference press, and verifying this compliance with integrated measurement tools. Only very recently have some proofing RIPs attempted to do the same with RGB drivers, and it is only in RIPs that you will find such capability. This is of little concern to the photographer making fine-art prints; likewise he may have no use for nesting, tiling, step and repeat, contour cutting, and a hundred other features that reside in RIPs. What they do care about is reasonably accurate color, the largest possible gamut, smooth screening, and low cost. I can't evaluate the statement Epson has actually "done their best" to keep useful features out of their drivers, but I can say that "avoiding investing in a RIP" would not be something to "smile about" if you had to make a living either proofing or production printing. Large- and grand-format display printers are not sold with RGB drivers as they would be largely irrelevant in that sector, whie pinching pennies on productivity software for a $20,000-4000,000 printer would be regarded as doubtful economy. Likewise in a prepress shop where the proofer must be able rto simulate multiple reference press conditions, including spot colors, on the fly, in multiple print queues. I don't know about you, but that's hardly boring, and it can't be done without dedicated RIP software. Mike Strickler MSP Graphic Services > > Message: 5 > Date: Tue, 06 Aug 2013 21:56:33 +0100 > From: Martin Orpen <martin(a)idea-digital.com> > To: "'colorsync-users?lists.apple.com' List" > <colorsync-users(a)lists.apple.com> > Subject: Re: RGB Drivers in RIPs > Message-ID: <048A41D9-AF6F-4F9F-8174-461EA8FCAD43(a)idea-digital.com> > Content-Type: text/plain; charset=windows-1252 > > On 6 Aug 2013, at 19:05, Mike Strickler <info(a)mspgraphics.com> wrote: > >> RIPs are enormously helpful in automating and foolproofing repetitive tasks and there is no longer any reason for fine-art printers to avoid them > > > Sure they're helpful. > > But the original, big selling point from RIP vendors (and colour management experts) was that RIPs would *always* produce *better* prints than were possible with the manufacturer's drivers. > > Now that users are realising that the drivers on printers like Epson x900 series are too difficult (or risky) for the majority of RIP vendors to bypass, the extravagant claims of superior printing have gone… replaced with boring stuff like "automating and fool-proofing repetitive tasks". > > Admittedly, Epson and Apple have done their best to make it as difficult as possible to repeatedly place a rectangle in the middle of a page, but there will be plenty of fine art printers (and loads of photographers) who have saved their money and avoided investing in a RIP who are smiling to themselves right now :-) > > -- > Martin Orpen > Idea Digital Imaging Ltd >

1 0

Re: Colorsync-users Digest, Vol 10, Issue 179 (Mike Strickler)
by Robert Crow 07 Aug '13

07 Aug '13

I too have tried many RIPs on ink-jets over the years (mainly for photographic reproduction up till now) and found that none of them came close to the manufacturer's driver in terms of what satisfies us as a good photographic fine art print. (I am sure they all do a brilliant job matching press conditions etc! and I do remember the old days when the RIPs used to beat the vendor's driver for BW neutrality etc). We came to the conclusion that the manufacturers 'special sauce' for driving these printers was not being made available to the RIP makers, so it is a relief to see them start to embed it... A relief because many RIP productivity tools such as automated layout etc are highly desirable to us but require compromising on output (in our opinion). Until now I only am aware of DInax Mirage, Colorburst Overdrive and EFI XF giving this option but I would like to know the extent to which EFI XF is using Epson's RGB driver (i.e is it basically the Epson Driver?) The other two rips are great but are not really designed for bigger production environments and automation. I'd be interested to know of any fully automate-able, integrate-able, server side runnable RIPs which use the printer manufacturers 'RGB' drivers and of anyone's experience with them (on or off list) thanks, Rob. >Message: 1 >Date: Mon, 05 Aug 2013 12:21:18 -0700 >From: Mike Strickler <info(a)mspgraphics.com> >To: colorsync-users(a)lists.apple.com >Subject: Re: Colorsync-users Digest, Vol 10, Issue 179 >Message-ID: <FD9B5268-FE52-4034-B199-B70ED3031B72(a)mspgraphics.com> >Content-Type: text/plain; charset=us-ascii >Though I don't sell it, Colorburst Overdrive makes exclusive use of printer mfr RGB media sets, no CMYK or CMYKn. EFI eXpress or Fiery XF (which I do sell) have embedded RGB drivers for a number of models. I have not made a survey of all the >RIPs that have RGB drivers. One requirement is that the printer model actually has RGB input. This is the case for all the "photo" printers from Epson, Canon, HP. Production printers (Mutoh, Mimaki, Roland, HP Scitex, VUTEk, Durst, et al) typically >have no RGB input .and one must drive them with a RIP in CMYK or CMYKn. My shopping advice: All RIPs that use RGB output are using the same maufacturer-written drivers, so base your decision on other features and price. RIPS, especially for >fine-art and signage applications, are mainly about productivity and not color, and this extends to RGB printing as well. (Please no deluge of claims from RIP mfrs on superior color from their CM engines, or about proofing--or begin a new thread >for it!) >Mike Strickler >MSP Graphic Services --------------------------------------------------------------------------- Please note that Middlesex University's preferred way of receiving all correspondence is via email in line with our Environmental Policy. All incoming post to Middlesex University is opened and scanned by our digital document handler, CDS, and then emailed to the recipient. If you do not want your correspondence to Middlesex University processed in this way please email the recipient directly. Parcels, couriered items and recorded delivery items will not be opened or scanned by CDS. There are items which are "exceptions" which will be opened by CDS but will not be scanned a full list of these can be obtained by contacting the University.

6 9

Re: i1Profiler linearization
by Mike Strickler 07 Aug '13

07 Aug '13

Perhaps my attention span is deficient and I've just missed something, but what is the actual objective here? To produce a better profile in the first place by improving the linearity of the printer output or returning output to a baseline via recalibration? The two may be treated together though in most instances with inkjet printers they are distinct concerns. As Terry points out, the linearization in i1P has nothing to do with calibrating devices but is designed to assist in the profile creation process. I don't think this addresses the original question. If the only concern is making a good profile, linearization does not need to be perfect by any means: Pretty good is good enough, and the profile should straighten out any remaining kinks. It should go without saying that great attention should be paid in RIPs to ink limiting, as this directly impacts profile quality. Scott may overstate the credentials needed for this, but it's best left to those who do it regularly and have had a bit of training especially if the materials are difficult to ptint on. Recalibrating color that has drifted is a far more interesting subject, and there are various solutions. They fall roughly into three categories: pre-profile, reprofile, and post-profile.Typical "relinearize" routines are in the first category. They treat the unprofiled output of each color channel. The idea is that inkjet printers drift in only one way: printheads deliver more or less color in each channel; correct the individual color channels back to a baseline and all colors will fall into place--the profile will be revalidated. IN some cases, the ink limits are readjusted as well. This is sometimes called a 2-D correction, as it does not directly treat the aggregate result. The shortcoming is that it isn't accurate enough to correct slight CMY gray balance variations that can still be objectionable. It works well enough for larger corrections. A variation on this is a 3D approach that also predicts the resulting CMY gray balance of the suggested channel adjustments. Onyx appears to have implemented this approach, which is well established in the offset press world (Curve3, Bodoni PressSign, Alwan Print standardizer, et al). The reprofiling approach is self-explanatory. It should be noted that if solid ink values have dropped below an acceptable level because of clogged heads or other issues this must first be addressed as the new profile cannot correct this. Reprofiling is time-consuming and not generally used in the heat of production. Finally, there are the recalibrations based on measurement and correction of profiled output, another "3D" approach. These can be in the form of a Lab correction LUT inserted before the output profile or a device link. This is a lot like reprofiling but can preserve the precision of the original profile and gently nudge all the data points with smooth curves. Very small charts may be used, so this approach can be used in the midst of a job. I personally favor this approach for inkjet printing. It is highly accurate and fast. Mike Strickler MSP Graphic Services

2 2

Re: i1Profiler linearization
by Tyler Boley 07 Aug '13

07 Aug '13

Ergosoft... manufacture's provided RGB, or N-channel, CMYK, etc. I've been using it for fine art for years, but on a stand alone PC I send files too, so not sure of your other stated requirements. Tyler > From: Robert Crow <R.Crow(a)mdx.ac.uk> > ... > ... > > ... I'd be interested to know of any fully automate-able, integrate-able, server side runnable RIPs which use the printer manufacturers 'RGB' drivers and of anyone's experience with them (on or off list) >

2 1

Re: i1Profiler linearization
by Bruce J. Lindbloom 06 Aug '13

06 Aug '13

On Tue, Aug 6, 2013 at 12:40 AM, Scott Martin <scott(a)on-sight.com> wrote: > > Two questions: > - How do I convert spectro readings to something useful eg. density > measurements? > On my website (Calc section) I have Excel spreadsheets that perform many spectral conversions, including density. If instead you want to "roll your own," the tables needed to perform density calculations may be found in ANSI CGATS.5-1993 (or probably a newer revision). Bruce -- Bruce J. Lindbloom bruce(a)brucelindbloom.com http://www.brucelindbloom.com

1 0