- This topic has 9 replies, 2 voices, and was last updated 9 years, 8 months ago by
.
-
Posts
-
Hello there.
Just wondering if you could take a look at two attachments.
2 questions:
(1) On the color temp graph, the readings at the low end are bad. .. Do you think if I lower the brightness of my monitor, before calibration, it would help?
(2) On the Gamma graph, the high end is bad… Do you think if I raise the contrast, before calibration, it would help?
Thanks.
(1) On the color temp graph, the readings at the low end are bad
(2) On the Gamma graph, the high end is bad
No, not really. These graphs are only informational.
Correlated color temperature is not suitable for the assessment of whitepoint and grayscale accuracy, because the correlation to visual appearance is much better expressed in a different unit like delta E 2000.
Same goes for gamma: It is normal that gamma fluctuates wildly near black and white, because tiny changes in measured luminance will already effect a relatively large swing in gamma, which again does not have much bearing on the visual appearance.
OK, thanks! …
By the way, please let me know your thoughts on the attached verification report. Maximum settings used just for fun. (11,000 patches and best calibration settings – Approx 4 hours using i1 Display pro) the best calibration I’ve ever achieved! (Inexpensive HDTV!)
Do you think it worked well? … Doesn’t appear to have instrument/device drift after 4 hours.
Looking foreword to your thoughts!
🙂
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Sorry, forgot to ask you something:
I’ve attached another screenshot of the gamut recorded during the above calibration…. It’s reports a 92.2% coverage and 95.2% volume … I’ve noticed that when I reduce the color saturation control on the HDTV before calibration, the gamut goes down from about 97% and 102% respectively… Is this drop significant in terms of being visual?
See second attachment shows a larger gamut produced by increasing the tv’s color saturation control closer to the default position. (If I put it all the way up to the default, the gamut will be filled completely, but the windows desktop becomes unbearably over saturated)
The problem is that the Windows desktop becomes over saturated if I leave the TV’s color saturation at it’s default setting during calibration…So I lower it to even things out, but then I loose gamut!
Thoughts?
Do you think it worked well?
Looks very good.
It’s reports a 92.2% coverage and 95.2% volume … I’ve noticed that when I reduce the color saturation control on the HDTV before calibration, the gamut goes down from about 97% and 102% respectively… Is this drop significant in terms of being visual?
Roughly 5% shouldn’t be too bad.
(If I put it all the way up to the default, the gamut will be filled completely, but the windows desktop becomes unbearably over saturated)
That’s not what the graph shows though. At default saturation, the gamut looks closer to Rec. 709 and thus the correct saturations for Rec 709 and sRGB content. Do you happen to use a particularly saturated desktop wallpaper?
No, I use a fairly plain wallpaper… By ‘Windows desktop’ I meant things like the picture icons in windows picture folders are saturated very red…
Also, I have Firefox setup as fully color managed (using my monitor profile) but things like Youtube videos are over saturated toward red as well…Not huge, but pretty obvious.
p.s. The graph I sent you with the larger color gamut was calibrated at the tv’s color saturation setting of ’43’ (The default is 50) … The graph with the smaller color gamut was calibrated at ’37’ (Where flesh-tones seem to look right.)
So, if I set the tv’s color saturation to default (50) I get 100% gamut (102% volume) but over saturated reds in windows (Thumbnails in picture folders, etc) … But, if I set the tv’s color saturation to 37, I get 92% gamut (95% volume) but proper looking thumbnail etc in windows … (But not the full gamut in photo applications).
What do you think I should do? (More important to keep 100% gamut for photo editing – Just sacrifice a nice looking windows environment?)
Can I fix this?
Thanks for all your help Florian… Greatly appreciated!
Success!
I think it’s best to calibrate at the default color setting of the tv (50) because this is what I got: …(See attached)
3300 patches was a test… Better ‘score’ usually comes at 11,000 patches.(My next, and hopefully final, step!)
Do you know if YouTube and Netflix are passed through Firefox’s color management system? …Doesn’t seem like it. (Red saturated)
What about Chromecast? … Could I plug it into the computer to get color management? (Stupid question, I know.)
See attached…
By the way, the slow-down to a stop thing happened to me again after the calibration and before the profile begins…This time I got a screen shot of the info box at the place where it always stops: (See attached)
Any idea why this keeps happening to me?
ps I noticed that the log doesn’t say anything after the calibration file was created…
This time I got a screen shot of the info box at the place where it always stops: (See attached)
Thanks. So it hangs while generating the profiling patch set, which is curious because that part is purely computational. Try reducing the number of patches on the “Profiling” tab using the slider and see if that changes anything. In case it happens again, don’t cancel the process right away, instead go to C:\Users\Your_Username\AppData\Local\Temp\DisplayCAL-<random_suffix> – there may be several folders like this, only the active one will contain some files, and you can check the modification date to see which one that is. Then, send me the .icm file from in there, please.
Any idea why this keeps happening to me?
No clue yet unfortunately. If I set up the profiling in the same way, the process finishes in a normal fashion.
I think I might have figured it out… I am using a home-built computer. (Asus motherboard with high performance hardware) and I did some overclocking about a year ago. I thought the system was stable, but now I suspect that it’s not. (I actually forgot that I had an overclock running)
So I’ve lowered the clock speed a notch and so far so good. (Your statement ‘it’s purley computational” gave me the idea.
Haven’t done another run yet, but will let you know.
