Re: i1Profiler linearization
Re: i1Profiler linearization
- Subject: Re: i1Profiler linearization
- From: David Wollmann <email@hidden>
- Date: Tue, 06 Aug 2013 23:04:37 -0600
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
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden