Screen/Printer Color Calibration - How often?

[trjwv]

Since I didn't very many responses to my previous thread on finding any place which calibrates monitors/printers, I am leaning towards purchasing colormunki but was wondering how often a person should calibrate. Is this something that once done and profiles set-up you never have to do again unless you purchase new monitor or printer or would you have to calibrate weekly to keep in calibration? If ofter how or why does your screen/printer get out of calibration once set? Sorry if these questions sound dumb but I know nothing about this topic and have researched some buit still not sure to the answers to these questions.
Thanks

 

[snberk103]

If nothing changes, then weekly to monthly is what I have read. Some places that do highly critical printing professionally may calibrate daily.

One of the things that affects the apparent calibration of your screen is the ambient lighting (ie the lighting in the room that may come from light fixtures and daylight). If you workspace lighting changes due to daylight, you may need to calibrate more often to account for sunny days and nighttime.

I understand that you should calibrate your printer when you change put in new ink tanks (you are replacing older ink from one batch with fresh ink from a 2nd batch).

etc etc

Start with the monitor... I think there is more to go wonky there than the printer. Then use the ICC profiles for the printer. That should get you pretty close easily.

 

[Ruahrc]

I think many of the "weekly to monthly" rules came from the older days of CRTs, where constant re-calibration was needed as the phosphors decayed.

Nowadays with LCDs I think color stability is not as much an issue anymore. I have no empirical data to back this up but it would seem that way to me, given my understanding of the technology (especially so with the recent LED backlit screens).

I recalibrate my stuff about every three months, primarily because I don't really see a benefit to doing it more often and I also am not in a highly color critical situation (just an advanced hobbyist).

Ruahrc

 

[snberk103]

I think many of the "weekly to monthly" rules came from the older days of CRTs, where constant re-calibration was needed as the phosphors decayed.

Nowadays with LCDs I think color stability is not as much an issue anymore. I have no empirical data to back this up but it would seem that way to me, given my understanding of the technology (especially so with the recent LED backlit screens).

I recalibrate my stuff about every three months, primarily because I don't really see a benefit to doing it more often and I also am not in a highly color critical situation (just an advanced hobbyist).

Ruahrc

I understand what you are saying. If I recall, and if I understand correctly, the CRTs started decaying almost from day one. It was a small change each month (perhaps barely noticeable even with precise measurements) but it was steady and constant. More or less a straight line. So one calibrated often to bring the display back towards the target.

Then the LCDs came along. They stay stable for a much longer time. Most of their designed lifespan. But then when they start to go, they go quickly. The frequent calibration is supposed to catch the LCDs as they start their death throes.

I don't know about the LED backlights. Maybe the backlight doesn't matter.

But as I said above, it's not just the monitor that needs calibration, it's the light in the environment as well. Frequent calibration will catch those changes as well. I have a Spyder that is used to calibrate the monitor, but mostly it reminds me to set the lighting in my office to a the "medium" level it was last calibrated at.

Cheers

 

[Ruahrc]

I understand what you are saying. If I recall, and if I understand correctly, the CRTs started decaying almost from day one. It was a small change each month (perhaps barely noticeable even with precise measurements) but it was steady and constant. More or less a straight line. So one calibrated often to bring the display back towards the target.

Then the LCDs came along. They stay stable for a much longer time. Most of their designed lifespan. But then when they start to go, they go quickly. The frequent calibration is supposed to catch the LCDs as they start their death throes.

I don't know about the LED backlights. Maybe the backlight doesn't matter.

But as I said above, it's not just the monitor that needs calibration, it's the light in the environment as well. Frequent calibration will catch those changes as well. I have a Spyder that is used to calibrate the monitor, but mostly it reminds me to set the lighting in my office to a the "medium" level it was last calibrated at.

Cheers

Well the decay of a CRT depended on its usage- if you used the red pixels in it a lot, those phosphors would decay faster than the other ones, and regular calibration helped keep everything on the same level. An LCD does not really experience such usage-specific decay and thus I think the display as a whole is more stable.

Your comment about ambient light is understood, but I am a little up in the air about it. Given that a monitor essentially produces its own light (via the backlight, or CRT illumination, etc.) how much does ambient light really interfere with the actual output of the display? I can see there being some effect, but how much? Would it be noticeable to the naked eye, or even measurable by a device? I would be curious to know. Ambient light can affect your eye's interpretation of the monitor but I think calibrating is meant to set your monitor's output at a known standard, instead of compensating for your eye's subjective interpretation.

IMHO though, a good foundation of a color reliable environment is to be able to recreate a consistent and repeatable lighting situation when you are working. Then, calibrate to that. So for example close any blinds and use interior light only when you work on pictures so that the ambient lighting situation is not changing depending on the time of day. I don't think it is the right solution to the problem to be re-calibrating your monitor every evening because it's darker outside, then re-calibrating it again the next morning when the sun is up.

I'm not sure what you refer to when you say that when LCDs go, they go quickly. As the "colored" elements of an LCD don't really change over the lifespan of the device (they are just color filters tinting the backlight coming through) the only real element that changes is the backlight. This is why I say that CRTs needed more constant calibration due to their variable decay (the same problem would exist in an OLED display, at least with current technology) whereas LCDs stay more stable over time because everything is decaying at a more equal rate. There may be a drop in brightness over time but the nature of fluorescent light is that the spectral output is pretty stable too AFAIK.

I guess I can only add my anecdotes wherein since I have 2 monitors, I can qualitatively judge the difference re-calibrating makes when I do it to one monitor and not the other. Assuming both are equal to start with, if I calibrate my external display, I can see if it is visibly different from the laptop display after it has been calibrated (whereas the laptop LCD is still using the older calibration). Even after a few months, I see little to no difference.

I think in the end 90% of the battle is getting your display calibrated upfront, the other 10% is all in using higher-end spectrophotometers vs colorimters, or doing regular calibrations to keep up with display aging.

Ruahrc

 

[snberk103]

...
Your comment about ambient light is understood, but I am a little up in the air about it. Given that a monitor essentially produces its own light (via the backlight, or CRT illumination, etc.) how much does ambient light really interfere with the actual output of the display? I can see there being some effect, but how much? Would it be noticeable to the naked eye, or even measurable by a device? I would be curious to know. Ambient light can affect your eye's interpretation of the monitor but I think calibrating is meant to set your monitor's output at a known standard, instead of compensating for your eye's subjective interpretation.
...

Despite the majority of the light being transmitted, under normal circumstances, I think the ambient light can add a measurable amount. Logic tells me that if I take my screen outside on a sunny day (I have a matt surface cinema display) I will be barely be able to see the display through the sunlight reflecting off of it.

So the question is, really, at what point does the ambient light level become too little to be noticed by the eye. And not just any eye, but an eye that is trained to detect very small differences in colour. (My optometrist loves when I come in. I've had a cataract removed, and I go in and tell him all the different ways that my two eyes see colour differently).

If I'm trying to colour correct something to within 1% or 2% colour accuracy, then if the ambient lighting adds even 3% of a colour (daylight vs tungsten) I need to account for it.

I'm not really sure about decay rates of CRTs or LCDs - it's just things that I've heard in the past, and that get lodged in the memory banks. But I am sure about LCD panels failing quickly when they do go... it's the florescent backlighting that goes. I think with CRTs you kept tweaking over the lifespan of the unit, until the controls had been maxed out. And that LCDs you shouldn't really have to tweak at all (after the first one).... that when an LCD starts to need tweaking you should be looking at replacements in anycase. I could be wrong... but that's what lodged in the memory banks.

The Spyder I have has a sensor on the back for the monitor. I don't use it often enough, probably; and one on the front for ambient light that measures constantly. It just sits on shelf above the monitor, measuring my office light. When I want to start colour correcting I make sure that the room light is close to what it was when I calibrated the monitor and confirm that with the Spyder.