Are monitor calibrators a meme?

When working on colour critical applications with a >95% RGB gamut monitor it might be worth investing in one.

>>59401086
Maybe you should install Gentoo first

>>59401086
>installing proprietary software so you can use your proprietary hardware

>>59401086
They're fucking amazing. Even on shit srgb monitors they tremendouslyimprove image iq.

Just get last gens model and use DisplayCal

It's very nice.

>>59401086
No, they're great but if you have a only one screen it's a waste to buy one, borrow one or rent one
>>59401273
>not using ArgyllCMS and Displaycal
It's like you want shit quality, also most of them are dead simple to interface with and don't need any blobs to operate
>muh hardware
There's the ColorHug if your freedom demands it

>>59401086

They're more exciting before you get em. After you get em, you use them once or twice and then they sit in a drawer somewhere forever.

>>59401273

I got an old i1 Display2 and though the hardware is proprietary, it just werks out of the box with DisplayCal.

On the topic of calibrators, is there any point to spending ~300$ on the latest one compared to getting a cheap 10+ year oldie?

>>59401953
Odds are that the filters on the 10 year oldie have degraded to shit, a lot of Display2's are useless nowadays since their plastic filters have drifted a lot
But the ones that used glass filters are said to stay fine, there's people with DPT94's that claim that theirs has drifted less than 2 dE's
Also, newer ones can profile wide gamut screens, which older ones can't, and they're much faster and quite more accurate, specially on blacks

>>59402062

How can you tell if it's a plastic filter or not?

Where to get a cheap old model/used/refurb? Regular stores dont seem to carry them and hardly even carry new ones

>>59401953
Most old ones don't work properly with modern LED backlighting. There's a huge difference depending on what types of LED's are in your monitor.

Used one of the recent Spyders on my previous screen and all it did was tint it a weird colour and fuck with my shadows. After having prints done multiple times in a professional lab that doesnt touch my edits, I found that by default my monitor was more or less accurate to what I saw when I recieved my prints. Maybe if you have a super expensive monitor you'd appreciate it, but as a colour blind retard I do not

speaking of monitor calibrators, how do I best calibrate my monitor WITHOUT a monitor calibrator?

>>59404063
I had problems with a spyder 3 on my older mva monitor, there was some setting in the calibrator, use native monitor black or whatever and putting a black t-shirt over the monitor when calibrating to kill ambient light helped a lot. But it was still hit or miss calibrating with it. My next Dell IPS monitor with 98% AdobeRGB worked a lot better with it.

>>59401086
No

>>59401086
Just get one isf calibrated.
They are done much better than you can do with $50 toy from amazon.

>>59404225
any respectable monitor is calibrated when it leaves the factory
the thing is they change over time, if you really need to be autistic about colour accuracy even at least once a month

>>59404365
>the thing is they change over time
Care to explain how because Im not following...

displays suffer from color drift and should be calibrated every so often if you're doing color intensive work

>>59404732
meant to quote >>59404570
as the monitor ages or is exposed to different elements, it affects the electronics that display the image

>>59404752
You will have to provide me some links to read on this stuff since google isn't turning up anything and I am very intrigued what it takes to perform calibrations on regular basis.

>>59404143
Be sure you have plenty of daylight in the room rather than doing it when it's dark or lights are on.
Windows 7+ has a simple configuration utility, you can use that.
Another option is manually tuning primaries on the monitor. You have to keep in mind that at different brightness levels the display may exhibit different color temperatures and there's nothing you can do there. Pick a brightness level that closely fits your most common working condition. Afterwards open up some grayscale charts and try to tune the primaries so the colors come out actually gray, not yellow-gray or blue -gray. Without a callibrator the only thing you can do is eyeball it and do multiple adjustments back and forth.

>>59404871

Usually takes couple of hours.

Have two identical monitors but one has huge amount of hours and lamp shifted towards yellow and is less bright, was real bitch to get it right but now when you look at it it's really hard to tell any difference from one with less hours.

>>59404783
you would need a colorimeter, which is a special tool for measuring light wavelengths. if you're just a regular user there's really no need for one, but if you work in photography or video production, proper calibration is necessary.

sometimes monitors have them built in

http://www.eizo.com/products/coloredge/cg318-4k/

or you can buy a separate one

http://spyder.datacolor.com/display-calibration/

>>59405130
Yes I get that,what I was asking is how do we measure what factors distort and how much they influence on calibrated pictures so we are to know that one monitor is scheduled for recalibration.
Thanks for responding anyway.

>>59405312
it's hard for me to answer since I'm not familiar with all the English terms

https://forums.adobe.com/thread/285096

https://www.quora.com/Why-do-color-profiles-drift-on-some-monitors-and-not-on-others

>buy expensive monitor calibrator
>run it on my monitor
>now it's dark and slightly tinted red

never again

>>59404871
>Windows 7+ has a simple configuration utility, you can use that.
it doesn't work on LCDs

>>59405595
>never again
then enjoy your incorrect colours.

>>59401086
only worth it if you work in some artistic field where colour accuracy is important.

>>59401086
On HDTVs they help a LOT to correct the huge blue shift of vivid/dynamic mode and red and green boosts the manufacturers use to fake skin tones and grass colors.
If you want your TV to conform to the standards you need one of those.
protip: don't buy one of them, rent it.

>>59405660
so "correct" is supposed to be shitty tinted dark stuff?
wow

>>59405541
Alright so I have CCFL Alpha IPS screen with factory isf calibration.
I tested the profile on AVSHD 709 few weeks back and passed.
How long of a time frame do I have until decay of the back light starts?

>>59405747
you either did it incorrectly or you're used to a colour shift and think the correct one is off.

>>59404063
>recent Spyder
>it tint[ed the screen] it a weird colour and fuck with my shadows.
Nothing wrong. Those "fucked up" shadows are the correct ones.
>by default my monitor was more or less accurate to what I saw when I recieved my prints
Let me explain you: the monitor has its specific color rendition capabilities, and the printer has a different one. Although the computer has the monitor profile installed the results are unpredictable because it doesn't know how the printer outputs color.
That is why you must also profile your printer so the PC knows how do printer and screen handle color. Only then you can match the prints to the screen and even preview the result(soft proofing).
You can do both by renting or buying an i1Pro or a ColorMunki Pro, those things can calibrate and profile monitors and printers. Argyll supports both devices so they should work on Linux.

>>59405784
No i had this problem with a spyder 3 as well. Half of the time i calibrated on my old MVA panel monitor i had color tint in the darks and sometimes crushed blacks, with some trying i got it to produce neutral greys but it definitely depends on the monitor you're calibrating. I never had problems on IPS panels.

>>59406087
The spyder sensitivity at low brightness was (and it's still) bad so it can explain the wrong darks.
The i1Display pro is better at this situations.

>>59406187
been thinking about getting that to calibrate my current monitor Dell UP2716D, since it's the only one that can calibrate the hardware LUT afaik.
But seems this one has been calibrated really well from the factory so i haven't had the urge to dish out 250€ for the x-rite. (I'm a Photographer)

>>59406405
rent it. Colorimeters drift over time and become useless after 5 or 6 years (the i1D3 is better in this aspect, it should last more because filters are sealed). If you rent it you leave the responsibility of checking and replacing aged devices to the company renting them.
I'd suggest you to use an i1Pro because their accuracy and because it's a spectrophotometer it doesn't get fooled by WLED, RGB backlights or OLED panels.

>>59406580

I doubt there's a any company in my country that rents that type of equipment.

Unless you need it for work where color accuracy is required it is a meme.

I want to calibrate my screens but not buy another device

will buying then returning with amazon work fine?

Where do I buy a calibrator for my monitor calibrator?

File: index.jpg (14KB, 208x208px) Image search: [Google]

14KB, 208x208px

>>59401086
im colorblind will it work?

>>59401868
>Colorhug
Looks pretty good, well priced too.
How accurate is it compared to the other "professional" brands?

>>59405638
But it does even though it gets the gamma wrong
>>59409502
Unfortunately it's shit unless you use a correction matrix for your monitor made with a spectrophotometer or a non shit colorimeter
>>59408312
eBay or contact Minolta or another of the lab tier spectro manufactures

>>59401086
No. The difference between an uncalibrated and a well-calibrated monitor is absolutely massive.

They're the single best image quality enhancement you can get, in terms of price/result.

>>59401953
>On the topic of calibrators, is there any point to spending ~300$ on the latest one compared to getting a cheap 10+ year oldie?
They're faster, more consistent and come with more rigorous factory testing/calibration.

The cheaper ones are usually hit and miss or slow as fuck

>>59404143
It's real easy anon, you just

1. Buy a colorimeter
2. Calibrate
-or-
1. Rent a colorimeter
2. Calibrate

>>59404365
>any respectable monitor is calibrated when it leaves the factory
If you're getting a monitor expensive enough to be factory-calibrated, the price of a colorimeter is nothing in comparison.

P.s. your $500 dell isn't factory-calibrated

>>59412374
Yeah but how do I calibrate a colorimeter?

>>59406087
>Spyder3
found your problem. The spyder3 is tuned to shitty sRGB W-CCFLs and won't work on any other type of display.

Should have gotten an i1 or Spyder 4, they let you customize the sensitivity curves to a variety of different backlight/display types.

>>59412400
With a spectrophotometer

>>59409502
iirc it's complete hit and miss. Lack of any professional QA on the product means they can either be spot-on or total garbage

>>59408287
Why don't you try it, tyrone?

>>59401086
They're not if you're using a proper display with good color stability, uniformity and linearity, and capable of hardware calibration (i.e. writable LUT). Trying to software-calibrate your typical "battlestation thread" supermarket monitor is pointless, though.

>>59412428
>Trying to software-calibrate your typical "battlestation thread" supermarket monitor is pointless, though.
works fine 4 me

>literally not a single monitor comes calibrated to BT.1886 out of the box
They all use shitty pure power 2.2 gamma or something, which looks totally wrong

>>59411157
>Unfortunately it's shit unless you use a correction matrix for your monitor made with a spectrophotometer or a non shit colorimeter
>>59401868
>>not using ArgyllCMS and Displaycal

Nice expertise there anon, have a fedora tip.

>>59401281
>Even on shit srgb monitors they tremendouslyimprove image iq.

Calibrators/profilers DO NOT improve image quality unless you define quality as ICC conformance.

>>59401953
>I got an old i1 Display2 and though the hardware is proprietary, it just werks out of the box with DisplayCal.

This. Got used Display LT instead, same hardware. BTW, is aging sensor window plastic a meme?

>>59404016

True enough, there are correction matrices in DsiplayCal though:

- for Apple Cinema Display (WLED)
- wide gamut S-IPS
- wide gamut S-PVA

And a whole load of corrections for Intelli monitors (probably not useful).

>>59408459

You can still correct the tone response.

>>59404016
>>59412619
For the record, these are the available modes I get from my i1 Display Pro

>Generic non-refresh display
>Generic refresh display
>LCD W-CCFL IPS
>LCD W-CCFL wide-gamut IPS
>LCD RG Phosphor IPS
>LCD RGB-LED IPS
>LCD W-LED IPS
>OLED
>Plasma
>Projector

Since I own an RG Phosphor-type display myself, this is very handy for calibrating it.

>>59412619
Color accuracy is defined as the second most important assessment on image quality according to ISF, with contrast being first
And contrast can be severely subjectively improved by decent gamma

>>59412675
Although note that in principle, the i1 exports its internal sensor sensitivity curves so you can combine it with measured primary curves of any display you want and get an accurate correction matrix out of it.

>>59412682

Calibration for colour space X won't improve image quality for displays with no internal LUT: you do not get any additional percepted information from monitor if you are using VCGT to calibrate it.

Profiling will improve accuracy only for colour managing programs.

Colour accuracy is not a common value for consumers because of how different uncalibrated displays of the people whom they are communicating with are.

>>59412798
Your GPU has 1x3D LUTs which are used for calibrating the tone response

>>59412821
>1x3D
3x1D ofc

>>59412798
>Profiling will improve accuracy only for colour managing programs.
You mean like image viewers, browsers, video players, image editors etc?

>>59412798
You do get improved gamma, color balance and color temp from VCGT
And a lot of programs nowadays are color managed, every major browser comes with color management enabled by default
>Color accuracy is not a common value for consumers
Even phones nowadays come with sRGB modes and decent accuracy

>>59412798
>Colour accuracy is not a common value for consumers because of how different uncalibrated displays of the people whom they are communicating with are.
Movies are mastered on uncalibrated displays? News to me

>>59409502
>>59411157
That's not the case with the colorhug2, it uses a much higher quality sensor from JenColor that's much more reliable and has much better compatibility.

Just be sure to turn any corrections off, the maker of it put some correction matrix on it despite JenColor recommending against it.

(https://groups.google.com/forum/#!msg/colorhug-users/NB7tdh4ojIw/Boc1lItOw5QJ)

Biggest thing I've noticed (even on cheap laptop TNs) is how black blacks can be while maintaining colour vibrancy. On my XPS15 and Qnix the difference is insane.

>>59412821

Those are called VGCT's and they are 8 bit -> 8 bit i.e. if you are using them to correct your monitor you are loosing information.

You should only use VGCT's for video gayms because they have no profile support.

Internal monitor LUTs may be 8bits -> 12 bits or better but it is a recent development.

>>59412839
>You mean like image viewers, browsers, video players, image editors etc?

Yes, given that they are not shitty - they should explicitly amange it themselves.

Calibration is considered to be changing hardware settings to better conform to various colour spaces while
profiling is describing the device behaviour to create ICC profile to use in ICC-aware programs.

>>59412842
>You do get improved gamma, color balance and color temp from VCGT

Of course you do but you'd better set the colour temp (what is balance lol) in monitor (even the shittiest of them have it).

See above,

>>59412854

Movies being mastered on calibrated displays does not mean that they are not adapting the distributed copy for uncalibrated displays.

>>59412893
You don't lose significant information with proper dithering, and some VCGT's are 10 bit or even 12 bit

>>59412893
>I like my blacks brown and my whites yellow

To each their own m8

>>59412893
>and they are 8 bit -> 8 bit
Actually it depends. Some are 10 bit, some are 12 bit, some are 16 bit, some are a mix; and all of them usually pass through dithering before reaching the display as well

>>59401086
>Are monitor calibrators a meme?
Colorphile memes. Just like Audiophiles.

>>59413358
shitty analogy
audiohphiles avoid objective measurements at all costs, which is the whole point of monitor calibration. so it's basically opposites

>>59412908
>and some VCGT's are 10 bit or even 12 bit

Unless it is loaded into monitor, it is not. Displays are accepting 8 bits, not 10.

>>59413017

Can you link some material? I am interested.

>>59413516
>Can you link some material? I am interested.
I've seen it posted on devtalk, search for the update that broke argyllcms because argyllcms couldn't handle the higher bit depth that the nvidia drivers exposed

>>59413516
>Displays are accepting 8 bits, not 10.
Many displays, especially the more expensive ones, can accept 10 bit input just fine

Also, at least the nvidia drivers do soft-dithering; i.e. they can do color adjustments at a higher precision and then dither down to 10/8 bit for the display.

You can in theory even run a 10-bit framebuffer on an 8-bit monitor with (temporal/spatial) dithering.

>>59412893
>Movies being mastered on calibrated displays does not mean that they are not adapting the distributed copy for uncalibrated displays.
Yes so explain to me how uncalibrated displays are fine for consumers again

>>59412893
>Yes, given that they are not shitty - they should explicitly amange it themselves.
Not sure if you understood my point; my point was that all of these applications are examples of things that benefit from a color profile

You made it sound like it's reserved to specialized applications only, which is simply not the case

>>59412893
>You should only use VGCT's for video gayms because they have no profile support.
Video games override the VGCTs anyway

>>59414757
There's some utilities that can force the VCGT, and ReShade can bolt on a 3DLUT on some games
I don't play games though so I don't know how much shit you have to do to keep it working

>>59413516
>>59412893
>>59412798
>>59412619
>>59401953
nice reddit spacing retard

>>59414745

Most of people cannot even sense the insanity of modern photography processing, why would they care about calibration? Resolution and contrast is probably what people care about, not much more than that.

>>59414751
>You made it sound like it's reserved to specialized applications only, which is simply not the case

Simpler programs WILL benefir from it but, for example, built in MS photo viewer does not support LUT profiles, only matrix, and Chrome does not support LUT too.

>>59414757
>Video games override the VGCTs anyway

DisplayCal has a tol for that.

>>59415669

Any better way of delimiting paragraphs? Manually editing each post in notepad to insert nbsp?

Aaaand Chrome won't assume that untagged image is in sRGB and won't convert it BTW. Most images of WWW are untagged.

>>59401086
If you're a professional they're useful.

>>59415781
>built in MS photo viewer does not support LUT profiles, only matrix, and Chrome does not support LUT too.
Well they're also examples of shit programs. Firefox does it fine.

Btw, chrome actually has support for ICCv4 profiles including LUT-type, but they specifically disabled it by default in their build script. You can change it to enable it.

>>59416988
>Firefox does it fine.

Same as Chrome (>>59415840) unless set up correctly.

>Btw, chrome actually has support for ICCv4 profiles including LUT-type, but they specifically disabled it by default in their build script. You can change it to enable it.

ArgyllCMS (which is what DisplayCal uses) does not produce v4 profiles at all, my profile is certainly v2. Yes it can have LUT too.

File: tmp_27579-14391667296341222949792.jpg (17KB, 250x250px) Image search: [Google]

17KB, 250x250px

>>59401272
my nigerian

>>59418442
>unless set up correctly.
this is /g/, are you implying changing a few settings is difficult?

>>59418721

Nope, I am implying that colour management is far from being mainstream.

>>59418930
/g/ is far from being mainstream

>>59418934

Sky is far from being Earth. What's your point?

>>59401272
>Proffesional video and photo software supported os
>Gentoo

pick one noob