- This topic has 7 replies, 2 voices, and was last updated 8 years, 9 months ago by
.
-
Posts
-
As I understand it, the “Verification” function is a closed loop test and only measures the accuracy of a generated profile against itself. That is to say, it measures your display’s response using the ICC profile against what the profile predicts, am I understanding that correctly?
What if I want to assess the profile I’ve created against the sRGB standard? Can this be done by selecting Simulation Profile and choosing “sRGB IEC61966-2.1”. Is that the correct simulation profile for sRGB?
The reason I ask is, I’ve run the verification versus the sRGB simulation profile target, and the ICC profile I created through calibration has some significant errors in certain places, with a few DeltaE values > 4. In contrast, I ran the verification against the sRGB target using my Dell U2414H’s factory calibrated sRGB mode (with no active ICC profile), and the results are much better, with the largest DeltaE at just over 2.
Am I correct in concluding that the monitor’s factory calibrated sRGB mode is providing a more accurate result in this case than the profile generated by dispcalGUI?
Thanks in advance!!!
Attaching verification reports.
- This reply was modified on 2015-07-08 00:49:08 by lockjaw333.
As I understand it, the “Verification” function is a closed loop test and only measures the accuracy of a generated profile against itself.
Correct.
What if I want to assess the profile I’ve created against the sRGB standard? Can this be done by selecting Simulation Profile and choosing “sRGB IEC61966-2.1”. Is that the correct simulation profile for sRGB?
Yes.
Am I correct in concluding that the monitor’s factory calibrated sRGB mode is providing a more accurate result in this case than the profile generated by dispcalGUI?
If the attached report “SRGBmode_vs_sRGB.html” is any indication, then the factory mode is reasonable but a profile could get you even closer (typical for a good quality profile is an average dE 2000 around 0.5).
Note that to check with the display profile, you need to disable “Use simulation profile as target profile”, so ICCvsSRGB.html is actually still evalating against the display in unprofiled state (you can see that in the report under “Target profile” which should be your display profile not sRGB in this case).
Indeed you are right. Attached is a verification report of the created ICC profile vs the sRGB simulation profile with “use simulation profile as target” unchecked. Significant improvement across the board from the factory sRGB mode., although the white point is now closer to 6600K at 6580K.
Would calibrating to a target white point of 6500K help? Are there any negative side effects to choosing a white point target and not using “as measured”?
Thanks!
the white point is now closer to 6600K at 6580K.
I wouldn’t worry about it, the measured whitepoint has a dE 2000 of 0.5 towards the nominal whitepoint of 6500K so the repeatability of display + instrument would already account for the difference.
One additional question.
I’ve read that games don’t utilize ICC profiles. However I can see a visual difference when playing games, as if the ICC profile is being applied. Do games really not use the profiles? Is so, is there any way to force them to be used? Would using borderless window work?
