- This topic has 13 replies, 2 voices, and was last updated 3 years, 5 months ago by
.
-
Posts
-
Hello!
Sorry if this is a dumb question. I’m kinda new to calibration business. I’ve been researching a lot of stuff about it for the past months, but never actually calibrated something yet (soon will).
I was thinking of getting an IPS Black monitor from dell. It has a contrast ratio of 2000:1
My question is: How important are contrast ratios for image work with photoshop? Does photoshop convert the colors back to 1000:1 after calibration? (kinda like if I get a wide gamut screen, color managed applications will clamp it back to sRGB to show correct colors)
If I get a 4000:1 contrast ratio monitor, or 3000:1, or 2000:1, or 1000:1 will it matter for calibration/work?
I really don’t understand how these different contrast ratios affect color work, if maybe I need to work within a specific contrast ratio depending on my work or what. I know I should get a screen that is at least 100% sRGB. If it’s a wider gamut, then I should either clamp it to sRGB or calibrate it so color managed applications show me the sRGB spectrum…. but what about the different contrast ratios? How should I treat them?
Also my drawing tablet is huion’s RDS-160-K. It’s advertised to have a Quantum Dot display. Do I choose DisplayCal’s Samsung QLED Q9? or a different correction?
I ask this because I’ve seen posts around that people are selecting something different despite their monitor being advertised to have Quantum Dot. Why is that?Last question: Can I just plug Calibrite Colorchecker Display into DisplayCal out of the box or do I need to download something from Xrite before using DisplayCal? (Like drivers or some sort of configurations).
Sorry if these are dumb questions.
Thank you very much!Calibrite Display SL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.Photoshop (or any other color managed software) do as profile says.
Usually calibration software (displayCAL) fakes “infinite contrast” on display profile (black point compensation), hence form certain L* to black profile data is not accurate and color managed app works as if display has 0cdm2 at RGB 0.
Usually advance color management software (Photoshop) fakes response when dealing with ICC display profiles with actual black brightness (actual contrast), a black point compensation, hence they fake display response from certain near black L* to 0 to avoid clipping: you’ll want tonal separation rather than black clipping.
A display with actual 1000:1 can work with a faked infinitye contrast profile… and usually this should be your first option: single matrix+curve+blacpoint compensation. If display had 500:1 or something like that due to agressive uniformity correction and warm white, the L* where it istarts to be fake may be a value to high…Also my drawing tablet is huion’s RDS-160-K. It’s advertised to have a Quantum Dot display. Do I choose DisplayCal’s Samsung QLED Q9? or a different correction?
As with WLED PFS there are “flavors” or QLED blacklights and depening on where it is place green peak (wavelegth) may give you P3 green or AdobeRGB green. That TV CCSS is P3 if my memory doe snot fail. You have WLED sampels of AdobeRGB green QLED backlights in some Asus PG models.
If you can choose, choose the closest one.
Depending on your particular unit of i1d3 colorimeter using one CCSS not accurate for your backlight may render bigger error than other i1d3.Last question: Can I just plug Calibrite Colorchecker Display into DisplayCal out of the box or do I need to download something from Xrite before using DisplayCal? (Like drivers or some sort of configurations).
i1d3 are HID devices and require no drivers to work with DisplayCAL. If you have installed Xrite software close i1Tray app and XRGamma executables to avoid DIsplaYCAL fighting with them for VCGT in GPU… or even better block them to do not run on startup/user logon.
From what I understood overall is that I shouldn’t worry about the contrast ratio of the monitor. I calibrate the monitor and if I want to work in sRGB I just set photoshop to sRGB and it will show me an sRGB image on my calibrated screen (as long as my screen can display at least the sRGB). And the contrast ratio will not be affected. If it’s 1000:1 or 3000:1 it will still stay the same in sRGB media creation. Is that right?
I was worried that getting a higher contrast ratio monitor would be pointless in media creation (photoshop). That it would only be usable in media consumption (games, movies and such).As with WLED PFS there are “flavors” or QLED blacklights and depening on where it is place green peak (wavelegth) may give you P3 green or AdobeRGB green. That TV CCSS is P3 if my memory doe snot fail. You have WLED sampels of AdobeRGB green QLED backlights in some Asus PG models.
If you can choose, choose the closest one.
Depending on your particular unit of i1d3 colorimeter using one CCSS not accurate for your backlight may render bigger error than other i1d3.What I gather from this is that I should test different corrections and stick with the one that gives me the best results… right?
That TV
You have WLED sampels of AdobeRGB green QLED backlights in some Asus PG models
I don’t understand what you said here. This is neither a TV nor an ASUS. Were you talking about my Huion or were you just giving some examples?
Another question: If I set the two monitors to screen mirroring what would happen to the color profiles? Would both calibrated monitors stay calibrated, or would their profiles drop from use, or would one of their profiles be repeated on the other monitor (due to them sharing the same image in mirroring)?
Thank you very much!
From what I understood overall is that I shouldn’t worry about the contrast ratio of the monitor. I calibrate the monitor and if I want to work in sRGB I just set photoshop to sRGB and it will show me an sRGB image on my calibrated screen (as long as my screen can display at least the sRGB). And the contrast ratio will not be affected. If it’s 1000:1 or 3000:1 it will still stay the same in sRGB media creation. Is that right?
I was worried that getting a higher contrast ratio monitor would be pointless in media creation (photoshop). That it would only be usable in media consumption (games, movies and such).Bellow sone L* value in dark greys it won’t track sRGB TRC (or gamma 2.2) because it cannot do it (1000:1)… but the error is small.
As with WLED PFS there are “flavors” or QLED blacklights and depening on where it is place green peak (wavelegth) may give you P3 green or AdobeRGB green. That TV CCSS is P3 if my memory doe snot fail. You have WLED sampels of AdobeRGB green QLED backlights in some Asus PG models.
If you can choose, choose the closest one.
Depending on your particular unit of i1d3 colorimeter using one CCSS not accurate for your backlight may render bigger error than other i1d3.What I gather from this is that I should test different corrections and stick with the one that gives me the best results… right?
No. You must use the one that matches your display spectral power distribution. If you know that it is a QLED, measure priaries and choose the QLED flavor that is able to produce such primaries.
That TV
You have WLED sampels of AdobeRGB green QLED backlights in some Asus PG models
I don’t understand what you said here. This is neither a TV nor an ASUS. Were you talking about my Huion or were you just giving some examples?
Examples of P3 QLED and P3 red + AdobeRGB green QLED. Depending on your native green primary coordinates choose one of those 2 samples.
Another question: If I set the two monitors to screen mirroring what would happen to the color profiles? Would both calibrated monitors stay calibrated, or would their profiles drop from use, or would one of their profiles be repeated on the other monitor (due to them sharing the same image in mirroring)?
Thank you very much!
AFAIK by limitation mirroring/cloning applies only one profile. You need to use extended desktop.
Bellow sone L* value in dark greys it won’t track sRGB TRC (or gamma 2.2) because it cannot do it (1000:1)… but the error is small.
So…. do I care if my monitor is 1000:1 or higher for image creation? Like… should I strictly use 1000:1 screens or can I get a nicer screen with a higher contrast ratio?
I don’t understand those terms ????No. You must use the one that matches your display spectral power distribution. If you know that it is a QLED, measure priaries and choose the QLED flavor that is able to produce such primaries.
How do I measure the primaries? Do I need to use a spectrophotometer? I don’t have the means to get one. Also in databases no one tested any Huion tablet.
Do you mean to say that not all QLED screens are “Samsung QLED Q9”?also
What I gather from this is that I should test different corrections and stick with the one that gives me the best results… right?
No. You must use the one that matches your display spectral power distribution. I
But wouldn’t using a wrong correction give me worse results?
Bellow sone L* value in dark greys it won’t track sRGB TRC (or gamma 2.2) because it cannot do it (1000:1)… but the error is small.
So…. do I care if my monitor is 1000:1 or higher for image creation? Like… should I strictly use 1000:1 screens or can I get a nicer screen with a higher contrast ratio?
I don’t understand those terms
There will be an error in darker greys for 1000:1, but it should not matter. look for a display with good color uniformity rather than 2000:1
No. You must use the one that matches your display spectral power distribution. If you know that it is a QLED, measure priaries and choose the QLED flavor that is able to produce such primaries.
How do I measure the primaries? Do I need to use a spectrophotometer? I don’t have the means to get one.
Uncalibrated display report in tools menu, even without correction.
Also in databases no one tested any Huion tablet.
Do you mean to say that not all QLED screens are “Samsung QLED Q9”?Green chanel spectral power distribution is a peak. Depending on main wavelegth of the peak you get a green primary with different coordinates (p3 vs Adobergb or beyond that).
So measuring color coordinates of native green primary will get you a hint about which QLED it is using.also
What I gather from this is that I should test different corrections and stick with the one that gives me the best results… right?
No. You must use the one that matches your display spectral power distribution. I
But wouldn’t using a wrong correction give me worse results?
Against what? You do not have a reference spectroradiometer to compare your measurements.
If you mean worse results validating factory calibration … is an useless test, like cheating on solitarie. Factory calibration can be very off.
There will be an error in darker greys for 1000:1, but it should not matter. look for a display with good color uniformity rather than 2000:1
But if I can get a good color uniformity monitor with 2000:1 shouldn’t I get that over another one that is equally as good in color uniformity but is 1000:1?
My doubt really was if after calibrating and setting things to sRGB if contrast ratio would make me not able to see the extra contrast. Basically if in photoshop it would end up showing as a certain contrast ratio instead of the monitor’s native contrast ratio. (kinda like how after calibrating and profiling a wide gamut monitor then setting photoshop to sRGB the only colors I see are sRGB)Green chanel spectral power distribution is a peak. Depending on main wavelegth of the peak you get a green primary with different coordinates (p3 vs Adobergb or beyond that).
So measuring color coordinates of native green primary will get you a hint about which QLED it is using.Alright I see now. When my colorchecker arrives I’ll see that. Thanks!
Against what? You do not have a reference spectroradiometer to compare your measurements.
If you mean worse results validating factory calibration … is an useless test, like cheating on solitarie. Factory calibration can be very off.
I meant deltaE of the colors and such. I thought that after calibrating, I would validate the calibration, and if the correction was wrong it would show up messed up in the report.
Like if I select a wrong correction and there’s no way to see if I choose a wrong one, then would I just proceed happily thinking that I got good results, but actually everything is wrong?
If I select a correction and that gives me a certain dE, then I try with another and that gives me a better dE. Then doesn’t that mean that the second correction is better than the first one?
Am I missing something?There will be an error in darker greys for 1000:1, but it should not matter. look for a display with good color uniformity rather than 2000:1
But if I can get a good color uniformity monitor with 2000:1 shouldn’t I get that over another one that is equally as good in color uniformity but is 1000:1?
Most uniform monitors are 1000:1/1500:1…
My doubt really was if after calibrating and setting things to sRGB if contrast ratio would make me not able to see the extra contrast. Basically if in photoshop it would end up showing as a certain contrast ratio instead of the monitor’s native contrast ratio. (kinda like how after calibrating and profiling a wide gamut monitor then setting photoshop to sRGB the only colors I see are sRGB)
Explained on previous messages. Photoshop uses blackpoint compensation. Calibration software usually fales contrast stored in profile.
Against what? You do not have a reference spectroradiometer to compare your measurements.
If you mean worse results validating factory calibration … is an useless test, like cheating on solitarie. Factory calibration can be very off.
I meant deltaE of the colors and such. I thought that after calibrating, I would validate the calibration, and if the correction was wrong it would show up messed up in the report.
No, it will be useless. In that scenario with a wrong CCSS : you will be calibrating with a “wrong meter”, getting some coordinates that are ok according to “wrong meter” because it aimed to them and will see validation according to that “wrong meter”. That test shows nothing and its totally useless.
Like if I select a wrong correction and there’s no way to see if I choose a wrong one, then would I just proceed happily thinking that I got good results, but actually everything is wrong?
That’s what will happen. Usually you can get a hint if whitepoint has a pink-green color tint… but you’ll need a correct reference to compare against (or one that you rememeber… but memory would be tricky)
If I select a correction and that gives me a certain dE, then I try with another and that gives me a better dE. Then doesn’t that mean that the second correction is better than the first one?
Am I missing something?That test is totally useless and means nothing.
If you know that it is using QLED, if you are certain about it: it is a QLED 100% sure, just measure green primary: P3-> use bundled CCSS for Samsung QLED TV, AdobeRGB -> use a CCSS from a Asus PG QLED.
Depending on i1d3 meter, on your particular unit, both CCSS may get same coordinates and it won’t matter but since this depends on firmware data and some math calculations explained in many threads… usually we want to use the correct CCSS to avoid uncertainties.Most uniform monitors are 1000:1/1500:1…
Vincent that is not an answer to my question.
If two monitors are identical, but one is 1000:1 and the other is 2000:1. I calibrate and profile both. If I open photoshop will I see a difference in contrast due to one being 2000:1 and the other 1000:1, or not?For everything else that you said…. thanks. I will live paranoid now LOL
How wrong will the colors be for choosing a wrong correction? Is it a small amount if I don’t get it exactly right?
Huion says it contains quantum dot. DisplayCal has 1 option for quantum dot. I thought I could just choose that and be done. Now I have to do all these extra steps which will probably increase the chance of me messing something up and getting things even more wrong.Most uniform monitors are 1000:1/1500:1…
Vincent that is not an answer to my question.
If two monitors are identical, but one is 1000:1 and the other is 2000:1. I calibrate and profile both. If I open photoshop will I see a difference in contrast due to one being 2000:1 and the other 1000:1, or not?2000: are typical VAs, common IPS are 1000-1500:1.
High end IPS, ColorEdge from Eizo, widegamut, are 1000-1500:1 A coloredge will win against any VA for image editing due to uniformity. There is no equal quality in a 2000:1You may say, OK … I DO NOT WANT extremely good uniformity, then you should not care about contrast.
If you aim to sRGB only or maybe P3 screens, no adobergb, BenQ PD series are reasionably good for their price, or dell U series (u, not UP… if you need an UP go to an Eizo Coloredge)
For everything else that you said…. thanks. I will live paranoid now LOL
How wrong will the colors be for choosing a wrong correction? Is it a small amount if I don’t get it exactly right?It depends on colorimeter firmware. It will affect whitepoint. I’ve explained it prevously.
Huion says it contains quantum dot. DisplayCal has 1 option for quantum dot.
It has many … but online.
I thought I could just choose that and be done. Now I have to do all these extra steps which will probably increase the chance of me messing something up and getting things even more wrong.
blame huoin, if they had provided a SPD of their displays you’kl know which one, but… measure native green 255 and you’ll know too.
