Re: Colorsync-users Digest, Vol 8, Issue 70
Re: Colorsync-users Digest, Vol 8, Issue 70
- Subject: Re: Colorsync-users Digest, Vol 8, Issue 70
- From: Mike Strickler <email@hidden>
- Date: Wed, 06 Apr 2011 13:08:12 -0700
No, Terry, reprofiling is not needed. But let's back up a bit. First, yes, the linearization is colorimetirc, which means it's not densitomtetric, which is what the original question was about. So that's settled. Next, printer linearization is always by definition all about the primary colors alone. It has no other common-sense meaning. Relinearization should, if designed right, also include some means of adjusting the ink density of primaries, including the 100%, as printers tend to drift in this way, not in the way they overprint, for example. I'll expand on XF's options farther on.
Back to the baselin procedure. I was asked about the entire process, which sets ink limits (individual, total) and light-dark ink levels, as well as linearizes the primary ramps. A QC target, with overprints, is printed and measured at the end as a check that the net result of the ink limits (primarily) does not produce any anomalies. Notice I said "measured" and not merely looked at--so it is not merely a "reference," as you surmised. I assume (but can't confirm at this point) that some tweaking of the individual, 2- and 3-C, and total ink limits then takes place. It is sensible and should not be so controversial.
Now for recalibration. The whole idea of a relin is to adjust for a change in printer output, and that change takes place in the individual color channels. With XF this also includes a small profiling-like chart, 180 patches, like GMG's chart but smaller. It has a lot of 2- and 3-C overprints, including some with black (which is sensible considering that many inkjet blacks aren't "black" at all). This relin process is available on an automatic/scheduled basis with the Epsons that have a Spectroproofer. This process also contains an iterative phase.
The relin works fairly well, but there is more available. In most cases I prefer to reoptimize the profile. Here one can choose between a small chart or larger ones, including up to a full IT8.7/4. This is iterative, as with GMG, and can bring dEs down to very low levels. I do like the tiny 46-patch "Offset profiling Chart," which does almost as good a job as a full IT8.7/4, but with which is it possible to do 3 or 4 iterations very quickly. ( It is worth considering how much accuracy to expect (and how big a chart to use) when an inline spectro is used, but that is another topic.) It must be initiated manually but can proceed without any intervention and can be started at the end of a shift. Please note: In no cases do we make a new profile.
Not to be too hard on you, I think it's worth saying that one should have some experience with a product before making too many statements about how they work. It's important that readers are not misinformed. In reality, there are several proofing RIPs that operate at a high level of elegance and sophistication, not just one, and often the technical differences we dwell on are less important than the overall design philosophy and how well they fit with the practical needs of users. EFI XF does most of its geeky stuff "under the hood" and presents a wizard-driven interface with judiciously chosen opportunities for manual override. Like any RIP, it takes some training and getting used to, but it is highly adaptive and extremely friendly and productive in everyday use. That's not an easy goal to achieve, and it takes a lot of technology to get there.
Mike
>
>
> ------------------------------
>
> Message: 5
> Date: Wed, 06 Apr 2011 14:18:52 -0400
> From: Terence Wyse <email@hidden>
> Subject: Re: Epson 4900 Calibration
> To: Mike Strickler <email@hidden>
> Cc: email@hidden
> Message-ID: <email@hidden>
> Content-Type: text/plain; CHARSET=US-ASCII
>
> EFI's "calibration" is strictly a colorimetric linearization....no overprints are used AFAIK for the re-calibration (the quality control target is used as a reference but that's all).
>
> The confusion is that during the initial base linearization, overprints are used to set total ink limit and in some of the other steps during the *initial* base linearization. The per-channel ink limits use L*a*b* (limits are set based on chroma primarily) and the L*a*b* values set during the initial base lin are used later for a re-lin. But an EFI re-linearization/calibration is fundamentally different than GMG...GMG uses about a 500 patch target for re-cal that includes overprints and will adjust the overprint combinations as well as the pure ink ramps....GMG essentially uses a "mini profile" for calibration.....and to be precise, GMG only considers the C+M+Y overprints....K is treated using L* only.
>
> Because EFI uses more-or-less a re-linearization process, re-profiling is almost always required to truly bring the system back into complete calibration....with GMG that is not the case.
>
> Terry
>
_______________________________________________
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