Monitor calibration quality assessment?
Chris Noble wrote:
This is a survey like question. For those that color calibrate their monitors what calculation method do you most frequently rely on to check quality of the calibration?
1) delta-E 1976
2) delta-E 1994 (CIE94)
3) delta-E 2000 (DE*00)
In addition to this, what criteria do you use to assess pass or fail, average delta-E and maximum delta-E? E.g., average delta-E < 2% and max delta-E < 4%?
Please also comment on your reasoning for choosing one method over the other.
I have Dell UP2718Q and Dell use delta-E 1994 (CIE94) with criteria average delta-E < 2% for their internal quality control prior to shipping monitors. X-Rite calibration software only provides option for delta-E 1976 and delta-E 2000. The latter provides lower delta-E values as it accounts for diminished human perception of color differences which is likely a more pragmatic approach to color accuracy assessment.
Thanks in advance!
My X-Rite i1 Display puck and software has a quality-measurement function that gives me delta-E values, but are these valid anyway? Presumably, some of the quality error is due to the accuracy of the puck, and using the same sensor to calibrate and then to measure accuracy will just cancel out the errors due to the puck itself. Any comments on this? How big an error source is the puck itself, compared to other parts of the signal chain? Does the puck get worse over time? Can the puck be calibrated using a reference card?
Revisiting this, DisplayCal software guidelines does delve into this question. A spectrophotometer likely will perform better than a colorimeter depending on software used to process/filter the calibration data. That said it seems that colorimeters such as x-rite i1 display pro have been designed to minimize errors. At end of the day each and every sensor would need to be calibrated according to some standard. For consumer products this is is a bit of a stretch beyond what is “indicated to customer in the usual just buy me jargon”. Indeed a valid point. I think the crappiness of X-rite software also gave motivation to my original question. Different cal SW tools such as DisplayCal give different impressions with regard to calibration quality according to the various standards, ‘76, vs ‘94, vs *00, etc. A big old can of worms actually.
|Katie at the River by RCGranstra|
|Diagonals by 6bender|
from Urban Decay - Abandoned Buildings
|Frost in the morning by Jill Hancock|
from Alphabet soup II. Letter B for “Bokeh"
|Singapore Merlion by RicksSony|
from Out of state.