Do I need to calibrate my printer every two weeks also?

[rustyair]

I just got the Colormunki photo and calibrated both monitor and Epson 3880.
I will probably re-calibrate the monitor every 2 weeks but do I need to re-calibrate 3880 printer every 2 weeks as well? Or is it just one time thing? Maybe annually? I only use Premium luster.

Thanks,

[paulr]

In my experience inkjet printers are stable. Depending on what kind of papers you use and how critical your expectations are (like, are you trying to print identical editions across paper batches) it might make sense to create a profile for each new batch of paper. I've never felt the need for this. I find the generic profiles work really well on the the 3880. The papers I've used have been consistent batch to batch. In fact, prints from my new 3880 match prints from my old one almost perfectly. I don't know if this remains true when the printer gets very old.

[Greg Miller]

The 17" and larger Epsons have a closed loop calibration system, designed so that every printer leaving the factory will print the same (given the same paper and profile)(which also allows the reliable use of generic profiles). The printer should not vary as it ages. At least until the point were it cannot self adjust itself enough to stay within the predefined standards.

[Leigh]

I don't know where the two-week recommendation came from.

It sounds like marketing hype from somebody trying to justify their existence.

Modern digital hardware is very stable over months and years, not days.

I would suggest rechecking the printer whenever you change the ink cartridge(s), since there can be some variation.

- Leigh

[pherold]

Monitors will tend to vary over the course of a month or so. We generally recommend recalibrating a monitor between 2 - 4 weeks.

Printers will not vary anything close to the way monitors do. So as long as everything remains constant: the printer, the ink, the paper type - you should be able to get the same results using the same printer profile. This reminds me of a color management "myth" that Steve Upton wrote about years go: Color Management Myths A good profile gone bad (http://www.colorwiki.com/wiki/Color_Management_Myths_6-10#Myth_.2310:_Profile_Rot_or_.22A_good_profile_gone_bad.22)

[algarzai]

I own a gallery and print shop. We calibrate monitors every 2 weeks and printer/paper combo every time we received a new batch of paper. We promise high the highest possible quality.
Having said that, i have compared a print from 2 different batches of the same paper using the profile created on the older batch. it is different. it doesn't look bad. it looks excellent. i would definitely use it. if i sold it to a customer he wouldn't tell the difference. We just have to do it to live up to our promise because some people do really require high quality.
in a nutshell. keep using your profile until you start noticing something wrong. then go ahead and re do it. be sure to optimize your profile for black and white and for color prints. You might want to save the profiles something like "semimatte_3_11_2013_FallColors.icc" i think this will be more beneficial when using colormunki as opposed to doing a new profile for every new batch.

[bob carnie]

plus 1

You need to check your monitor very rarely.


I don't know where the two-week recommendation came from.

It sounds like marketing hype from somebody trying to justify their existence.

Modern digital hardware is very stable over months and years, not days.

I would suggest rechecking the printer whenever you change the ink cartridge(s), since there can be some variation.

- Leigh

[bob carnie]

I have never found this to be the case.

but then again I colour correct by the numbers, they never lie.

QUOTE=pherold;1075005]Monitors will tend to vary over the course of a month or so. We generally recommend recalibrating a monitor between 2 - 4 weeks.

Printers will not vary anything close to the way monitors do. So as long as everything remains constant: the printer, the ink, the paper type - you should be able to get the same results using the same printer profile. This reminds me of a color management "myth" that Steve Upton wrote about years go: Color Management Myths A good profile gone bad (http://www.colorwiki.com/wiki/Color_Management_Myths_6-10#Myth_.2310:_Profile_Rot_or_.22A_good_profile_gone_bad.22)[/QUOTE]

[sanking]

"Printers will not vary anything close to the way monitors do. So as long as everything remains constant: the printer, the ink, the paper type - you should be able to get the same results using the same printer profile."

+2

In my experience there is almost no variation in results by printer so long as you use the same paper type, ink set, setting and profile. Over the years I have printed many ink calibration pages with QTR and there has been almost no difference in results with the Epson ink sets. I have seen some difference in results with all gray ink sets, not sure whether this is due to my refill procedures or to the inks themselves.

These remarks assume that you maintain the printer so that all the nozzles are firing.

Sandy

[Greg Miller]

but then again I colour correct by the numbers, they never lie.

Some day perhaps you will elaborate on this. What does colour correct by the numbers mean? How would one colour correct, say, a landscape image by the numbers?

[bob carnie]

Ahhh the million dollar question


Its tough but when you look at the colour numbers daily, and start having an imprint , one can start understanding when certain values do not make sense and lead you to the right balance.

Margulis wrote a couple of books on this, a third one is coming out. It was the hardest part of the PS package for me to learn.



Some day perhaps you will elaborate on this. What does colour correct by the numbers mean? How would one colour correct, say, a landscape image by the numbers?

[Greg Miller]

Ahhh the million dollar question


Its tough but when you look at the colour numbers daily, and start having an imprint , one can start understanding when certain values do not make sense and lead you to the right balance.

Margulis wrote a couple of books on this, a third one is coming out. It was the hardest part of the PS package for me to learn.

So how would you know something is out of balance when for a scene shot twice; once in open shade or twilight)(cool light) and again during when lit with golden hour light (warm light)? Same scene but very different color balance. Neither is right or wrong. If it is just an "understanding" gained from experience, then I am niot buying that it is a valid repalacement for proper calibration ad profiling.

[polyglot]

I presume that by "from the numbers", bob means the numbers measured by the profiling sensor and therefore the numbers in the ICC profiles. As opposed to visually where as you say, different lighting gives very different balance.

[paulr]

We just have to do it to live up to our promise because some people do really require high quality.

Calibration really doesn't have anything to do with high quality. It has to do with consistency, if you need to make identical prints across batches of paper. And it has to do with the ease and economy of achieving high quality.

You can make absolutely top-quality prints without any calibration whatsoever. It just requires a lot more time and materials and trial and error. This is how most of us did all our printing in the darkroom.

[bob carnie]

Yes of course your example has different colour balance, but you will be able to pick up clues by constantly reading the info palette and seeing where the numbers land. Over time you will be able to know when the numbers do not make sense. I mention Margulis as this is territory that you are taken in with him , and its very heady to grasp when you are working
on his tests.
Also in your example I would suggest you are pointing to a very subjective colour subject and any variation of balance would look fine to you or I . But if you include a grey road, black tarmac and some nice white puffy clouds then my task is much easier. With landscape the proof is always in the print , does it look right to you.


For example... One test is where Margulis takes a colour scene and changes the subjects colour balance **bad balance** but you are not allowed to see the original scene. The student is forced to turn the monitor to BW monitor reading but still able to see the colour numbers via the info palette.
Then you are on your own to colour correct the scene, All you have for reference is a black white rendition of the scene and your LAB numbers. After you are finished the class turns back on the Colour to see how each person has done.

One needs to rely on the info palette and start thinking, **CAN THIS COLOUR BE RIGHT. ** I found this to be a very difficult challenge and since we were competing with others , where I failed others succeeded and by listening to them describe how they figured out when a colour was off , well for me at least an eye opening experience.
I have been colour correcting with filters my whole career by looking at a print, comparing it against a colour ring around for visual colour clues.
This method was very interesting and took quite a few attempts before I got the hang of it and though I still correct on the table with filters, I use the numbers in PS to help guide me.

Dans point is that colour blind people, or people with poor colour judgment can indeed colour correct by reading clues in the image.
I have seen this myself in three of his classes and know it is a very valuable tool to use.


None of this has anything to do with balancing your monitor every two weeks, I am not arguing but curious, as I have not seen this need to play with something that seems to be working now over four years.
We had our colour management export come in and profile 5 papers at the beginning of the year, using the new Xrite system, I had him check my monitor and he said it passed.
I know my monitor is reliable as my prints coming out are what I think they should look like and I have a viewing box right beside my monitor that I can check.
I am curious to understand the reasoning to calibrate ones monitor every two weeks, or even more than twice a year, is this something you do Greg? and if so would you elaborate why you do this and as well the benefits you see.

I do use numbers , maybe more than others as I am always moving from paper, silver, inkjet, ect. that all require different end points, which is IMO extremenly important, from the initial
scan right through to the final print on any media.

Also I am very stubborn and do not accept a white balance keystroke in Lightroom and call it a day.

[Greg Miller]

I color correct my color landscape images by eye. I have printed for a photographer with extremely critical color sensitivity. He can consistently detect color shifts of a Photoshop value of 1 in a print. This forced me to see color more critically. With color landscapes, and the time of day that color landscapes are typically made, I never expect a rock to be gray or puffy white cloud to be white. I have never found the Photoshop white balance tools to be of any value either.

[bob carnie]

I can agree that landscapes are more subjective ,and much easier to determine by looking at the image on screen and then on print,no doubt, I like the fact that I can read a rock or rocks and the puffy white cload and analyse what the numbers tell me and helps me make very small adjustments.

Another example ... I started this whole LAB thing by a thread Chris Jordan started years ago on this forum, basically he was talking about bringing casts out of neutral areas using LAB and setting anchor points around the offending colour and bringing the scene back to pleasing balance.
Doing this could be done by visual inspection, but really understanding the numbers and how to fudge the curve with very precise moves is much easier.

[paulr]

Bob, I think those skills of correcting by the numbers are valuable. I don't think they're completely a substitute for subjective evaluations, just as I don't think a calibrated system is a substite for evaluating the print afterwards. Both just streamline the process.

Part of the issue is that calibration concerns the tonal response curve in addition to the colors. Some of the trickiest prints I've worked on involved getting the sense of luminance just right. Tweaking the curve a bit too much one way makes the print dull, too much the other way makes it brittle. These issues have to do with tonal relationships and not absolute values. If anyone could visualize this kind of thing by the numbers, I would be impressed, but would have little hope of emulating them.

[bob carnie]

I never substitute numbers over subjective evaluations, I do both, but knowing the values are of immense value as well as my subjective aesthetics.

For example.. when printing silver negatives or silver prints from photoshop, I need to have a very tight handle on where the numbers are on highlight with detail and shadow with detail.
Just because I can see it on screen does not mean it will translate to paper of film.
By knowing for example that highlight with detail above 94 L reading is going to go to paper white, or shadow with detail below 4 L value is going to pure black is important to me.
Every paper that I use has distinctive different end points for these two critical areas.

I want to know exactly that I can lay down tone , if I tried to do this with a subjective evaluation well good luck with that.
A lot of my work these days is basically laying down tones on very different materials and for different print processes, a new area to my shop is making silver negatives for alt processes. Now there is another whole area of complication... We all know that negatives for silver printing is much different for lets say , carbon, gum or pt pd.
I would be in real trouble if I did not know how to translate PS numbers to final media end points.

I hope this explains my tirade on numbers a bit better.





Bob, I think those skills of correcting by the numbers are valuable. I don't think they're completely a substitute for subjective evaluations, just as I don't think a calibrated system is a substite for evaluating the print afterwards. Both just streamline the process.

Part of the issue is that calibration concerns the tonal response curve in addition to the colors. Some of the trickiest prints I've worked on involved getting the sense of luminance just right. Tweaking the curve a bit too much one way makes the print dull, too much the other way makes it brittle. These issues have to do with tonal relationships and not absolute values. If anyone could visualize this kind of thing by the numbers, I would be impressed, but would have little hope of emulating them.