- This topic has 2 replies, 2 voices, and was last updated 1 year, 2 months ago by
.
-
Posts
-
I’ve got an LG 32UN650-W, and am re-calibrating it with DisplayCal 3.8.9.3 and an old Spyder3 Express. My previous calibration looked pretty good to me, but was done a couple years ago, and now I need to soft-proof some photos for printing. The DisplayCal options can be a little overwhelming, so I’m hoping to get a sanity check here.
Monitor specs are here: https://www.displayspecifications.com/en/model/623d20f3
#1 (and most important?) is the Correction drop-down on the Display tab. From the “display tech” info bubble & the display specs above, it appears that this monitor is 95% P3, and so should be considered “LCD PFS WLED”, yes? The Correction drop-down comes up with Auto (None) selected. The options are mostly Intelli Proof variants that don’t seem relevant, but one – Spyder 3 & Wide Gamut S-IPS (generic) – seems like it would be more correct.
#2 Whitepoint – currently set to As Measured, which the uncalibrated report gives as:
16:52:32,061 White chromaticity coordinates 0.3149, 0.3307
16:52:32,061 White Correlated Color Temperature = 6373K, DE 2K to locus = 4.3
16:52:32,061 White Correlated Daylight Temperature = 6375K, DE 2K to locus = 0.5
16:52:32,061 White Visual Color Temperature = 6229K, DE 2K to locus = 4.1
16:52:32,061 White Visual Daylight Temperature = 6392K, DE 2K to locus = 0.5But I’ve read in this guide that “The whites on your screen will look a little warmer at 5,000 and it is the setting to use if you want to softproof for printing.” Is this true? (I’m a VFX guy, I don’t do a lot of print work.)
#3 White level – the same guide linked above says, “90cd is pretty dark for my liking, but is that value that is recommended to start with if you goal is to softproof for printing.” Uncalibrated report gives 215.55 cd/m2 at a monitor brightness setting of 80. Obviously, super bright for print work.
#4 Tone Curve – previous setting was sRGB. The guide suggests, “Always use Gamma 2.2 here.”
#5 Contrast (monitor setting) – current setting is 70, but the guide suggests resetting it to factory default: “The DisplayCal forums are filled with people saying that you need to leave contrast at default or you run the risk of clipping whites and/or blacks at a hardware level and the screen performance will be far less than what it is capable of when you are through calibration.” Uncalibrated report gives: Contrast ratio = 541:
#6 Gamma (monitor setting) – The monitor has three Gamma modes, unhelpfully labeled 1,2,3. The manual is similarly unhelpful: “The higher the gamma value, the darker the image becomes. Likewise, the lower the gamma value, the lighter the image becomes. If you do not need to adjust the gamma settings, select Mode 4.” Current setting is Mode 2, which I think I just picked as randomly pleasing in the past. Uncalibrated report with this setting gives: Aprox. gamma = 2.21
#7 Black Level – Again, the monitor has “high” and “low” settings, and the manual states: “High – Keeps the current contrast ratio of the screen. Low – Lowers the black levels and raises the white levels from the current contrast ratio of the screen.” Current setting is High, which sounds correct to me. Uncalibrated report gives: Black level = 0.3985 cd/m^2, 50% level = 46.52 cd/m^2
Apologies for yet another “help me with settings” thread… As mentioned, it’s a bit overwhelming – even for someone who’s been doing VFX for 30 years, and looks at color all day.
I’ve got an LG 32UN650-W, and am re-calibrating it with DisplayCal 3.8.9.3 and an old Spyder3 Express. My previous calibration looked pretty good to me, but was done a couple years ago, and now I need to soft-proof some photos for printing. The DisplayCal options can be a little overwhelming, so I’m hoping to get a sanity check here.
Useless colorimeter… so whatever you do is unreliable.
Monitor specs are here: https://www.displayspecifications.com/en/model/623d20f3
#1 (and most important?) is the Correction drop-down on the Display tab. From the “display tech” info bubble & the display specs above, it appears that this monitor is 95% P3, and so should be considered “LCD PFS WLED”, yes? The Correction drop-down comes up with Auto (None) selected. The options are mostly Intelli Proof variants that don’t seem relevant, but one – Spyder 3 & Wide Gamut S-IPS (generic) – seems like it would be more correct.
#2 Whitepoint – currently set to As Measured, which the uncalibrated report gives as:
16:52:32,061 White chromaticity coordinates 0.3149, 0.3307
16:52:32,061 White Correlated Color Temperature = 6373K, DE 2K to locus = 4.3
16:52:32,061 White Correlated Daylight Temperature = 6375K, DE 2K to locus = 0.5
16:52:32,061 White Visual Color Temperature = 6229K, DE 2K to locus = 4.1
16:52:32,061 White Visual Daylight Temperature = 6392K, DE 2K to locus = 0.5But I’ve read in this guide that “The whites on your screen will look a little warmer at 5,000 and it is the setting to use if you want to softproof for printing.” Is this true? (I’m a VFX guy, I don’t do a lot of print work.)
Unreliable measurement device, so it’s pointless. Make a visual match with your light booth an then set whitepoint as native. Then ignore whitepoint error in measurement report. But the sensible choice is to get a reliable colorimeter.
#3 White level – the same guide linked above says, “90cd is pretty dark for my liking, but is that value that is recommended to start with if you goal is to softproof for printing.” Uncalibrated report gives 215.55 cd/m2 at a monitor brightness setting of 80. Obviously, super bright for print work.
Maybe unreliable too. If colorimeter was reliable “X lux on booth / PI = Y cd/m2 in display” as an starting point, then visual match both
So with your current device set manually (so white level as measured) then match visually with booth.#4 Tone Curve – previous setting was sRGB. The guide suggests, “Always use Gamma 2.2 here.”
Yes, softproofing is color managed so whatever gamma you put will be undone… so choose a gamma close to monitor native response (~2.2 usually).
#5 Contrast (monitor setting) – current setting is 70, but the guide suggests resetting it to factory default: “The DisplayCal forums are filled with people saying that you need to leave contrast at default or you run the risk of clipping whites and/or blacks at a hardware level and the screen performance will be far less than what it is capable of when you are through calibration.” Uncalibrated report gives: Contrast ratio = 541:
Unreliable colorimeter, so who knows actual contrast value. Get a reliable one like i1d3.
#6 Gamma (monitor setting) – The monitor has three Gamma modes, unhelpfully labeled 1,2,3. The manual is similarly unhelpful: “The higher the gamma value, the darker the image becomes. Likewise, the lower the gamma value, the lighter the image becomes. If you do not need to adjust the gamma settings, select Mode 4.” Current setting is Mode 2, which I think I just picked as randomly pleasing in the past. Uncalibrated report with this setting gives: Aprox. gamma = 2.21
Keep that value, as said before whatever you put on DisplayCAL as target gamma will be undone by color management engine in your application, so if display gamma is close to 2.2. in that OST settings, use 2.2 in Displaycal.
#7 Black Level – Again, the monitor has “high” and “low” settings, and the manual states: “High – Keeps the current contrast ratio of the screen. Low – Lowers the black levels and raises the white levels from the current contrast ratio of the screen.” Current setting is High, which sounds correct to me. Uncalibrated report gives: Black level = 0.3985 cd/m^2, 50% level = 46.52 cd/m^2
“Usually” you rely on application softproof capabilities to simulate paper black level, but with very low contrast displays and if you need to use a fake infinite contrast profile (BPC) as display profile then simulate value may be too fake to be reliable.
So if you had a well behaved 1000:1 IPS display @ D50 with 800:1, and you had a i1d3 and you were using a XYZLUT+matrix profile with no BPC and you were using PS as softproof application, leaving display at its maximum contrast at that whitepoint and using Photoshoop simulate black ink in softoroof wil do the job.
(=do not alter display black level, keep max contrast you can at that whitepoint)If you are not using a softproof engine with black point simulation in printer profile you may rely on manual contrast decrease (increase back level).
Example: Photoshop allows you to disable that black in simulation and rely on manual contrast decrease, but other apps may not, hence you cannot use manual contrast decrease and a display profile with BPC.
Also using a display profile with no BPC may result on clipping on older software when no softproofing.
Apologies for yet another “help me with settings” thread… As mentioned, it’s a bit overwhelming – even for someone who’s been doing VFX for 30 years, and looks at color all day.
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Vincent – thanks for the reply. Yes, the Spyder3 device is quite old. I probably should have mentioned that this isn’t for some pro-photo project. I’ve got no light box, I’m using a consumer print service that doesn’t have icc profiles, and I just need to get close enough that what I’m seeing doesn’t look radically different when printed. The reason? Time & money – we’ve got a week and change to get this done, and the photos aren’t going into a museum, but out to an “open air” art fest, where they’ll be viewed under uncertain lighting conditions, and likely be damaged by the end.
The initial test prints I sent to several labs did look radically different from my monitor, but at least some of that difference was consistent across the labs (darker, less contrast, bit of shift in the reds), so I thought I should start with re-calibrating my monitor. The first thing I discovered was that since the last calibration, the monitor colors had shifted a bit. (Either that, or I’d done something to cause DisplayCal to be loading the wrong LUT.) Fixing that already resulted in a closer match to the test prints.
So I suppose I’ll go with the linked guide’s numbers as a starting point, and eyeball it from there, per your suggestion.
Oh – and apologies for the “non-seriousness” of the whole situation. I know that it can grate on one’s nerves when someone asks how they can do something half-assedly, rather than the right way. But that’s just the nature of this project. With more time, cash, and a different exhibit location, we’d likely have done it differently. Maybe we’ll get that chance in the future.
