Unfortunately, the topic of this EDR's usefulness has recently been called into question again. This is despite the data already posted here, and the substantial amounts of data posted on other forums by myself and several others comparing an i1 Display Pro using both its factory Generic CMF calibration and the EDR, against consumer 10nm X-Rite Spectro's, and high end 5nm Spectro's in the past. So I have once again repeated some more measurements in an attempt to show the benefits of using the WRGB OLED EDR over the factory Generic CMF calibration when access to a Spectro to profile your meter is not an option. Comparison measurements were made against a Jeti 1501 5nm Spectro that had its yearly recalibration/recertificati on in early August 2021. Each set of measurements were made in ColourSpace using the Remote Control function, allowing two profiling windows to drive two probes simultaneously, with the Extra Delay setting timed so that each meter read the same patch at the same time.
The WRGB OLED display was put into an uncalibrated state. The white point was "roughly" adjusted to be somewhere close to a D65 white point, and the internal 3D LUT was reset to ensure the display was in its native widest gamut in an attempt to replicate how the display would be in a standard OOTB condition prior to performing a Pre-Calibration, profile and LUT generation.
This comparison was then run three times, producing six measurement profiles (three pairs). First, comparing the Jeti to the I1d3 using the factory Generic CMF setting. Second, comparing the Jeti to the i1d3 using the WRGB OLED EDR setting. And finally, comparing the Jeti to the i1d3 after performing a FCVM probe match to the Jeti.
The results of these three sets of comparison measurements can be seen below.
In the results shown above, you can see the xyY of each profile which has been directly converted from the XYZ tristimulus values stored in the profile BCS. It also shows the +- xyY error, followed by the dE2000 of the colour difference of each patch read. This error value was calculated by using the reference white of each profile to convert the measurements from XYZ into L*a*b* space, allowing each colour patch measured in each pair of profiles to be compared directly against each other and return a colour difference error value. The dE2000 error value is not comparing to any specific colour space/gamut, it is directly comparing one colour against the other (e.g. White vs White, Red vs Red, etc.).
It is well known among anyone who has done any serious testing with the i1 Display Pro on a WOLED display that the probe has some issues using its Generic CMF factory calibration, especially reading the red primary. So the results there are expected.
The results of the WRGB OLED EDR comparison profile show a clear advantage, especially in White & Red, as well as the Magenta & Yellow secondaries. Again, this is expected due to the EDR significantly improving the way the i1d3 reads the Red primary.
Also as expected, the comparison when using a dedicated FCVM probe match is far, far superior than the other two, producing a highly accurate match, not just in White and the Primaries, but the secondaries also. This gives confidence that when you do perform your profile/LUT, that it will be as accurate as possible.
Clearly, anyone who has access to Spectro would always perform their own probe match. But it is just as clear that not everyone has that luxury, and in this case, the WRGB OLED EDR is far from being considered worthless because it does provide some improvement. Especially in the Red primary and its associated Secondaries.
With the measurements provided above, it also shows why it can be considered useless to use "shared" alternate white points. Whilst I admit I was actively participating in testing and sharing AWP's in the past, the more I looked into it, the more I discovered it can actually cause more harm than good when providing an AWP for someone to use if that person does not have a reference (Spectro) to compare against.
For example, if I had no Spectro myself and was restricted to using the i1d3 in conjunction with the WRGB OLED EDR and targeted the default standard of D65 (0.3127/0.3290), what I would actually be getting is 0.3073/0.3248, which is already somewhere in the ballpark of what others have suggested when using a Spectro. So if you do target that AWP that someone else provided, you would just end up adding more and more blue.
This defeats the whole point of using an AWP, which is to match to a non metameric display at D65!!!
So, this is the primary reason I stopped "recommending" an AWP to anyone some time ago, and always advise that without a Spectro to create a reference meter profile, first try D65 (using the EDR to correct the issues using the Factory Gen CMF calibration), and if that doesn't look right, try performing your own perceptual match, as outlined in the Light Illusion guides pages.
https://www.lightillusion.com/perceptual_match_guide.html
Of course, using an AWP is always going to be a subjective thing and can/will vary from person to person, so use what you want at the end of the day. As long as it looks good to you.
Anyway, the main point of this post was to show there is benefits in using the WRGB OLED EDR over Generic CMF, and that it is not useless, worthless, or any of the other things its been called again recently. And if you don't have a Spectro, it absolutely is a good choice.
Measurement data (BCS) and Excel with results attached at the bottom.
Sorry for the rambling rant, I just get annoyed when I see some of the statements I've seen again recently without any data to back those statements up.
Happy New Year everyone. Let's hope this one is better than last!!
|
Page 2 of 2: « Previous 1