- This topic has 3 replies, 2 voices, and was last updated 7 years, 8 months ago by
.
-
Posts
-
I’ve been trying to improve the match between soft proofs from darktable and what comes off the printer. The contrast ratio of my monitor was about 600:1 (using the “Report on calibrated display device” tool). So I re-calibrated setting the Black level to that previously reported (0.2643) and the white level to achieve the desired contrast ratio (300:1 – 79.29). This involved greatly reducing the brightness setting on my monitor, as expected.
However, the brightness of the soft proof in darktable did not seem any less. So I ran the “Report on calibrated display device” tool again. This time it reports:
Current calibration response:
09:17:22,677 Black level = 0.0889 cd/m^2
09:17:22,678 50% level = 15.20 cd/m^2
09:17:22,678 White level = 80.05 cd/m^2
09:17:22,678 Aprox. gamma = 2.40
09:17:22,678 Contrast ratio = 901:1so my attempt had failed miserably – the black level was not honoured.
Is this a bug, or is it not possible to constrain both the black and white levels like this?
300:1 seems excessive when it comes to print – even a high quality inkjet print on premium glossy paper should be less than 200:1 – but normally reducing contrast ratio for soft proofing is achieved by using appropriate and accurate ICC profiles. Does dark table honor the printer profile and use colorimetric intent for the softproof, so that paper white and inks are properly taken into account (these options may also be called “Simulate paper white” and “Simulate black ink”, or something along those lines, respectively)? Is the printer profile accurate?
It is perfectly fine to raise the black level via the calibration curves (too) though, when it doesn’t matter that contrast is reduced system-wide, not just when softproofing.
so my attempt had failed miserably – the black level was not honoured.
What instrument are you using? Failing to hit a specific target black level may indicate that the instrument isn’t precise enough to read that low (at least not consistently), or may indicate the display has some dynamic dimming or similar features enabled. Also note that “report on current calibration” will report on whatever is currently in the videoLUT.
Yes, darktable honours the ICC profile. I think my main problem is the location of the print – it’s rather a poorly lit room. Moving it to a brightly lit room shows it to be brighter than I had believed, and much closer to the soft proof view in darktable.
The instrument I have is ColorMunki Display.
I’ve since discovered that my monitor has a “paper” preset, so I’m going to try to recalibrate with the monitor set to that level. Hopefully that will be a more suitable contrast ratio.
Calibrite Display SL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.I think my main problem is the location of the print – it’s rather a poorly lit room.
This may well be. When it comes to achieving a good screen-to-print match, proper lighting is of significant importance.
The instrument I have is ColorMunki Display.
Ok, this instrument should be more than capable to read low light levels.
Hopefully that will be a more suitable contrast ratio.
When you test contrast after calibration using “report on current calibration”, make sure the generated profile is installed and active (i.e. assigned in System -> Color if you’re using Gnome).
