- This topic has 4 replies, 3 voices, and was last updated 6 years, 8 months ago by
.
-
Posts
-
So I just got a new monitor (a Dell U3011) and in the process of calibrating it, I’m running into some issues.
I’m using Resolve, with two monitors connected to the desktop (through a GTX 780) and one more connected through an Ultra Studio Mini Monitor. I’ve calibrated and profiled all three, with a whitepoint of 6500K, however they don’t look the same. The U3011 is far warmer and has a magenta cast, compared to both of the others, even though all the LUTs are applied in the proper places, and the DisplayCal Loader appears to be functioning. This happens in whatever configuration the monitors are in, even when the U3011 is connected to the Ultra Studio it’s always warmer and magenta.
Based on my understanding, after calibration to Rec. 709, they should all look the same or atleast very close (since they’re all different monitors) otherwise there wouldn’t be any point in having standardized color spaces. So my question is, what am I doing wrong, and how do I know which monitor is correct?
The other two monitors are a Dell S2715H and a Dell S2740L. I know these aren’t ideal, which is why I went for the U3011 (Until I can get something from FSI or similar). I mostly care about the monitor on the ultrastudio being accurate, I’m not so concerned about the windows desktop although it would be nice to get the resolve viewer at least pretty close.
Thanks so much any help anyone can provide!
The comparison can only be as good as the verification reports show they can be. What you are probably doing wrong is missing the importance of the first stage of calibration – balancing the colour bars and setting the brightness. Just getting 6500K may not be correct. You also need to look at the de error from the white point and get it as low as possible. This takes time and all of the adjustments on the monitor inter react including the brightness in a way that isn’t predictable when they are close to the desired value.
Since monitors have offered adjustment ranges of 0 – 100 rather than 0-255 hitting an exact 6500K might prove impossible so it’s best to set white point as found and get as close as you can along with minimising or at least obtaining acceptable de errors found via verification, This means carrying out more than one run and takes time and needs a judgement call. If not happy doing this use the software that came with the calibration instrument. Argyll is extremely high end and more comprehensive than what usually comes as standard with some very expensive spectrometers.
Your U3011 is an odd ball for calibration this way. Best read this
http://www.tftcentral.co.uk/reviews/dell_u3011.htm
If I had one and ran windows or I assume mac I’d buy dell’s calibration kit. If you follow the link’s suggestions you would just set brightness in the first stage and accept the colour bars as they are.
In terms of rec709 and sRGB you’d best read this. I’m a photo person so no interest to me.
https://www.image-engineering.de/library/technotes/714-color-spaces-rec-709-vs-srgb
Maybe the monitor has a rec709 preset. If not looks like you should use the sRGB preset. It seems the main difference is the mathematics associated with the media.
Out of interest I have just switched to a much larger than usual monitor. As I suspected for colour critical work there is a maximum image size that is truly useful. All down to viewing angles and eye accommodation. The eye can can only take in so much accurately and to really judge a result the image needs to fit within that area and that depends on how far some one is away from their monitor.
John
–
Hi John, thank you for the detailed response!
I understand how to adjust the monitor settings, and my understanding is that the whitepoint setting is what determines the target for that. I’m not talking about a subtle difference here, the monitors are clearly very different. Attached is a photo that should help demonstrate what I’m seeing.
Also attached are measurment reports from the U3011 and the S2715H. Unless I’m reading it wrong, the color tempature it’s saying they are at is pretty close, and not at all in line with what it looks like. Correct me if I’m wrong, but shouldn’t Rec709 keep a standard whitepoint, and all monitors calibrated to it should ideally look exactly the same?
Jeremy
Based on my understanding, after calibration to Rec. 709, they should all look the same or atleast very close (since they’re all different monitors) otherwise there wouldn’t be any point in having standardized color spaces. So my question is, what am I doing wrong, and how do I know which monitor is correct?
Please see the FAQ on visual whitepoint mismatch (and solutions).
