- This topic has 4 replies, 2 voices, and was last updated 7 years, 7 months ago by
.
-
Posts
-
Hello Florian.
So in terms of temperature, my understanding is that we should calibrate our monitors to match the final application (ie., 5000k for print, and 6500k for Web.) … Then we should match our room light temperature to that of our monitor. (In an Ideal situation).
So if the above is generally true, then wouldn’t editing at 5000K with room light at 5000k look the same on the web to a computer calibrated at 6500k in a room light of 6500?
Wouldn’t the mind adapt to the proportionality and see white as white in both situations? … So why do we need to calibrate for different situations? Doesn’t it just matter that we are calibrated properly at the source, assuming that the destination is also properly calibrated, regardless of the numbers?
I ask because I did a test: I edited an image in ACR on a machine calibrated to 5000k in room light of 5000k then posted it to Facebook… Then I went to another machine calibrated to 6500k in a room light of 6500k and pulled the image back up… The result is that the image looked too cold… In the 5000k scenario I got it to look right, but relative to the 6500k scenario I saw that I had pulled out too much warmth!
Can’t wait to figure this one out once and for all!
So in terms of temperature, my understanding is that we should calibrate our monitors to match the final application (ie., 5000k for print, and 6500k for Web.) … Then we should match our room light temperature to that of our monitor. (In an Ideal situation).
What usually works well is adjusting the room light to a given standard (e.g. “D50″ for print, or”D65” for video, although video grading is usually done in a fairly dim environment without much light), and then match the monitor white to the room light.
So if the above is generally true, then wouldn’t editing at 5000K with room light at 5000k look the same on the web to a computer calibrated at 6500k in a room light of 6500?
Not next to each other, but in isolation due to the way the human visual system works which adapts to a wide range of “white” light, the colors relations to one another and the respective whitepoint are preserved in a way that should give a similar visual impression.
Wouldn’t the mind adapt to the proportionality and see white as white in both situations?
Yes.
So why do we need to calibrate for different situations?
To make it easier to perceive “white” as “white” with respect to the ambient lighting situation and viewing conditions. E.g. a “D65” monitor in a “D50” environment will look exceptionally blue, and a “D50” monitor in a “D65” environment will look exceptionally yellow.
Then I went to another machine calibrated to 6500k in a room light of 6500k and pulled the image back up… The result is that the image looked too cold…
It may take a while to adapt to a whitepoint change or a change in viewing conditions, depending on how drastic the differences are, especially in terms of absolute luminance. Also, you need to make sure the web browser actually color manages the image, and there are other pitfalls i.e. I’m not sure if embedded profiles are stripped if images are posted to Facebook (should be less of a problem if it’s an sRGB image because that is usually the assumed default).
In the 5000k scenario I got it to look right, but relative to the 6500k scenario I saw that I had pulled out too much warmth!
Just make sure to not make direct side-by-side comparisons, i.e. the other monitor should not be in viewing distance and there should not be different sources of “white” light.
Thanks for your explanations! It helps…
Question: I have two brand new light bulbs. both rated at 5000K . One is LED and the other is Florescent. But they each have a different color tint! The LED had a greenish tint, and the Florescent has a reddish tint. (Relative to each other). Which is better to use for the editing environment? LED or Florescent?
They both can’t be right.
Thanks.
Which is better to use for the editing environment? LED or Florescent?
The one with the higher CRI (Color Rendering Index), although take the manufacturer claims with a grain of salt. With a spectrometer, you could check the CRI yourself. Good graphics arts illuminants usually have a CRI of > 90.
Thanks, yes, I found an LED bulb today at a normal hardware store which actually advertises this CRI rating on the package as being “Better Color” . It’s an 83 which I know is below your recommendation of >90, but I have to believe that it’s a lot closer than my other ‘normal’ bulbs are.
In fact, the 5000K rating of this bulb is much closer to 5000K compared to those other “5000K” bulbs which only reached 4300K. (This new bulb reads about 4900K!) And when viewing my screen, which is now calibrated to 5000K thanks to DisplayCAL, I am seeing richer and more realistic color than ever! … It is truly beautiful. I had no idea what I was missing due to mismatched environment\monitor lighting and poor CRI bulbs.
I have also noticed that if I stay in a room light of 5000K for a while, I cease to think the screen looks yellow\orange anymore! … We truly do adjust if we set it up right.
Lets get the word out, this REALLY matters!! (And for a $10 bulb and a little time with DisplayCAL, results are stunning!)
Now to find a 6500K bulb. (Don’t seem to be as popular.) I’ve had salespeople tell me that they are waaay too blue. (I don’t think they understand why I need them.)
Steve.
