Re-profiling

[apetitphoto]

how often, if ever, does a printer need to have profiles done?
Assuming ink formulations, paper formulations, don't change.

 

[palombian]

The profiles are used in the printing software, not in the printer.
I can imagine the output of the nozzles can change with age, but for an ordinary consumer or semi-pro printer this means the printhead needs a change.

 

[Ink stained Fingers]

Assuming ink formulations, paper formulations, don't change.

What is the confidence level for your assumption ? What is your threshold for a re-calibration ? You can apply all quality control methods if you want to ... if you have the time -- if you think it's worth the effort . e.g. you could print a proof type verification patch pattern on all your prints so you have a kind of ongoing check in place .. and you can get your spectro recalibrated every year etc. And how are you doing your profiles - printing patch sheets multiple times , scanning each several times and averaging the readings - and more

 

[mikling]

Depends on the accuracy you want. The Canon Pro-1000 and larger brethren have a densitometer to care of that. Periodically you can recalibrate ( not reprofile) the printer to meet the original specs. This can cover small shifts in papers, ink and ennvironmental conditions. Even if using OEM inks all the time. Even if you do not put in new inks, for cost of one print, you can automatically recalibrate and thus the ICC you had been using for a long time is retained or is more closely retained without recreating a new one.
Not really important on a refill board but they want or the intended markets needs this level of accuracy or more correctly, repeatability.

 

[apetitphoto]

"Ideally" there are any number of reasons to reprofile. But practically is there a need beyond known paper no ink formulation changes? That is, do printers drift enough in their application of ink to warrant re profiling on a regular basis? Monitor profiling tools, for instance, say to do it monthly, weekly, daily even, depending on various factors.

 

[mikling]

Calibration is to take care of physical shifts which will occur. The profiling is to take care of the behavior of the colors on a known basis.
Profiling is typically thought of being one thing where there are actually two things that go on. The calibration is assumed to be in order or always correct. Even if it isn't will not drift. So essentially the profiling we think of is actually characterization and calibration that will remain in place forever. If there is a drift, then people would reprofile. This reprofiling might have been the result of something that might have been correctable with a recalibration.
So for most people..characterization is assumed to be non measurable and one is not able to do anything about it.
The reality is that calibration DOES have a place on an Epson as well. The larger machines all have it. On smaller machines it is needed if one were to remove or replace the printhead or logic board. Part of this process requires that one get an Spectro to set the printer back up to specs. With the Pro-1000, this is not necessary because the densitometer is user accessible by software.
This is why the 1000 is so interesting. It gives you all the toys and bells and whistles of the big machines that can fit on a sturdy desk but inexpensive it is not.

 

[Roy Sletcher]

how often, if ever, does a printer need to have profiles done?
Assuming ink formulations, paper formulations, don't change.

Here is a cheap lazyman's approach. Simple and quick if you feel you are happy with the current output.

1: Print a standard reference image. (Suggest the Outback image) on your paper of choice using your current preferred profile.

2: File it in a dark place.

If at any future time you suspect colour change, variation, drift or whatever. Then reprint the same image under the same condition and compare. If prints are for personal use and you cannot see any difference all is well. Or is it- we know our brain fools our eyes more than we like to admit?

If you are printing commercially, or for the approval of others, then a spectrometer will allow you to objectively compare the readings of specified patches and measure the Delta E variation.

If you get significant variation - that is where the fun starts.

 

[stratman]

Given your parameters, if all you do is reprint things you already ran through your favorite photo editor and have saved, then follow what others have recommended about printing an archival copy, store properly and then occasionally print the same image and look for changes and then consider recalibration of the printer after all other variables have been examined and corrected as necessary.

If, however, you are working with images that require editing, ie using your monitor to make changes, and you want to keep things as close as possible to WYSIWYG between the monitor and printer, then the best bang for your buck will be to calibrate your monitor at regular intervals and if you are working under different lighting conditions (or repaint the room, for instance) from the previous calibration. Monitors and calibration equipment drift over time.

It can become quite personalized in how much you want to do and how often.

 

[Roy Sletcher]

Given your parameters, if all you do is reprint things you already ran through your favorite photo editor and have saved, then follow what others have recommended about printing an archival copy, store properly and then occasionally print the same image and look for changes and then consider recalibration of the printer after all other variables have been examined and corrected as necessary.

If, however, you are working with images that require editing, ie using your monitor to make changes, and you want to keep things as close as possible to WYSIWYG between the monitor and printer, then the best bang for your buck will be to calibrate your monitor at regular intervals and if you are working under different lighting conditions (or repaint the room, for instance) from the previous calibration. Monitors and calibration equipment drift over time.

It can become quite personalized in how much you want to do and how often.

You are misrepresenting what I wrote, and changing the OP's original question. I will accept this is "ON ME" for not being clear. So here we go again:

The original question was, "how often, if ever, does a printer need to have profiles done?
Assuming ink formulations, paper formulations, don't change."

There is some ambiguity in terminology, but it is reasonable to assume he means a printer's ICC profile for a specific paper and ink.

No Mention of monitor profile - a factor you introduced.
No mention of WYSIWYG - another factor you introduced.
It also has nothing to do with editing on a monitor, or even WYSIWYG, which you introduced into the discussion.

What I was trying to say, and I will use different words, but try and retain the KISS approach:

When he has the first satisfactory profile that meets his requirements. All that is needed is to NOW PRINT A REFERENCE IMAGE UNDER CONTROLLED CONDITIONS USING THAT PROFILE.

The reference image I suggested was from:
http://www.outbackphoto.com/printinginsights/pi048/essay.html

But there are many others of equal value.

At some point in the future the OP will have problems with the reproduction of a printed image.

At that time he reprints the test image under the same controlled conditions. He should have an exact match with his original test image. If not the trouble shooting procedures should be obvious.

THe answer to the OP's question should be apparent - whilst the printer, paper, ink parameters remain constant the ICC profile is valid. My long winded explanation was how to verify this fact.

This is not something I dreamed up. It is SOP (Standard Operating Procedure) for most serious printmakers.


Everything should be made as simple as possible, but not simpler.


Get it!
Got it!
Good!



rs

 

[RogerB]

When he has the first satisfactory profile that meets his requirements. All that is needed is to NOW PRINT A REFERENCE IMAGE UNDER CONTROLLED CONDITIONS USING THAT PROFILE.

The reference image I suggested was from:
http://www.outbackphoto.com/printinginsights/pi048/essay.html

But there are many others of equal value.

At some point in the future the OP will have problems with the reproduction of a printed image.

At that time he reprints the test image under the same controlled conditions. He should have an exact match with his original test image. If not the trouble shooting procedures should be obvious.

THe answer to the OP's question should be apparent - whilst the printer, paper, ink parameters remain constant the ICC profile is valid. My long winded explanation was how to verify this fact.

This is not something I dreamed up. It is SOP (Standard Operating Procedure) for most serious printmakers.


Everything should be made as simple as possible, but not simpler.


Get it!
Got it!
Good!



rs

I agree entirely with Roy's recommendation but for one detail. In my experience the Outback image is great for evaluating the "real world" capability of the printer. However, in my experience it's not easy to spot small colour differences when you have two prints and have to switch your gaze from one to the other.

I printed commercially for 7 years and as part of my QC I printed a reference image for each paper that I used at least once a month. The reference that I used was, and still is, an electronic version of the Xrite/GM CC24 but in a linear format. My reference is trimmed after printing to make a full-bleed (borderless) strip that I can lay directly on top of the print that I want to check. Just eyeballing the two I can easily see changes of less than 2deltaE in the less saturated colours and even smaller in the greys. That's very hard to do with two copies of the Outback image.

Having said all that I can honestly say that my 3880 has never needed re-profiling apart from one occasion when I made a mistake in re-filling. Of course more consumer oriented printers may not be quite so stable and some third-party inks may vary from batch to batch.

Here's the CC24 that I use. A big advantage is that you can print it in the margin of a scrap print or do multiple test prints on a dedicated A4 (or 8.5 x 11) sheet over a long period of time. I've added a ZIP file with a TIFF of the image which is in the AdobeRGB(1998) colour space for anyone that would like to try it.