Monitor calibration advice

Eddzz!!

Eddzz!!

Messages
1,120
Name
Eddy
Edit My Images
Yes
Hey all,

A few days ago I bought myself a second-hand Spyder3 Pro monitor calibration tool. I decided to plug it in yesterday and calibrate my Dell u2412m monitor. I ran the first calibration following the instructions on screen... On this first occasion the calibration finished and I thought that the picture was far too warm. I did a little research online and read advice that I should disable the ambient light setting on the tool as this can do more harm than good, so I did. Ran the calibration once more and, whilst the picture was cooler than before, still seemed incredibly warm and dim (this is with a white point of between 5800K and 6500K). It may be that I'm just used to a cooler and brighter picture, but obviously I will favour accuracy. I usually have my monitor quite bright, but the calibration tool forces me to turn down the brightness far below what I would usually set it at. Does it sounds like I'm doing anything wrong? Is it a case of just "getting used" to the warmer calibration? Let me know.
 
I guess the ultimate test is to view other 'calibrated' photos from other photographers and see how they compare on your monitor. You're quite welcome to look at my Flickr albums which have been processed using a calibrated screen.
 
Easiest way is to process a photo on the screen with the calibration as set by your calibration tool. Send it off to your photo printing service and see what comes back.

My prints were always coming back too dark. After calibrating the screen it was less bright but prints started coming back looking properly exposed.

If you do not print photos then do what Barry suggests above and/or email photos to friends and see what they look like on their monitors.

Also the ambient light needs to be constant, I use a fairly dark room away from any direct sunlight and always have the lights switched on.
 
What are your settings? What is your room lighting?

If you are used to a bright screen you will be surprised how dark a correctly calibrated screen is. It's for this reason most people's images appear dark on a calibrated screen, as most people have their screens FAR too bright.

Why not edit a photograph on it now it's calibrated and post it on here to check?

I've just taken an image of my currently messy desk. The room is lit with 6500k lighting.. the screen is calibrated and I've included a grey card. I have made sure the grey card on the screen matched reality, and I'm using a very high end screen that's hardware calibrated with a i1 Display pro.

How does this look on your screen?


View attachment 37960
 
Last edited:
chuckles said:
The card looks a bit 'dog-eared' - sorry, couldn't resist. ;)
Click to expand...

Oh it's had a hard life! It's actually not that old. I get through grey cards at a scary rate.
 
Spot on:plus1:.

Something else to consider is the light that is hitting your screen where it is sited. Try to block off any light that is directly hitting the screen from above and from either side that way variable light changes are minimalised and you create a more stable light enviroment. I have tried propriety screens but nothing has worked better than a two foot deep shelf above the monitor which overhangs the screen by about 18" and a couple of tight fitting black cards either side. That shelf is bloody handy:). I would recommend you read this...

http://www.colourcollective.co.uk/dzyt652t/Practical_Colour_Management.pdf

I paid for mine but that is a freebee and if you need to know about the subject then it is a pretty good guide. Trouble is, if you are like me, you eventually understand and get the plot but then six weeks later forget it all:confused:.

I lost over a year to obsessing over colour management stuff as a newbee, most of which I never use now or cant remember. Colour is very important to what I do but it is all done over the net, so a waste of time in the real world/end. Everyones monitor is different in 99.9% of cases so if thats where your images are going, i dont think its worth putting too much effort into trying to make that perfect. if you are printing your images of course they have to be to a very high standardbut for the net/ websites ....

What I have done in the past is to show images of popular mass market product packaging (xox boxes in yellow red and green I think it was:naughty:) that was edited/created on a calibrated system that the customer can compare to the real thing at home on their own monitor, ie a benchmarch if you like, that they can use to see how their monitor compares to a professionaly calibrated one. I stopped this in fear of breaching any copywrite but it was a good solution to the problem for a moment.

If I was a younger man I could see myself getting right into this, the growth in high quality image colour accuracy will, I think be huge and the opportunities vast.

Ats wot eye fink
 
Pookeyhead said:
What are your settings? What is your room lighting?


How does this look on your screen?
Click to expand...

I don't really know a way in which to describe how it looks... The grey card is a fairly dark grey on my calibrated monitor?

I can tell you one thing... The white of the forum background appears to have a warm yellow-ish tinge to it.
 
Eddzz!! said:
I don't really know a way in which to describe how it looks... The grey card is a fairly dark grey on my calibrated monitor?

I can tell you one thing... The white of the forum background appears to have a warm yellow-ish tinge to it.
Click to expand...

It'll only look correct if the room lighting is correct too.
 
Okay... Here is an image edited by myself this evening on my "calibrated" monitor to what I think is a pleasing result. Tell me what you guys think.

From San Marco by Edward Reynolds, on Flickr
 
Last edited:
Eddzz!! said:
I don't really know a way in which to describe how it looks... The grey card is a fairly dark grey on my calibrated monitor?

I can tell you one thing... The white of the forum background appears to have a warm yellow-ish tinge to it.
Click to expand...

The grey card on my screen is the same as a 18% grey card in real life.. I spot metered off the card, so it was rendered 18% grey (or possibly closer to 16% for all you pedants out there :)), so that is how it should appear in the image..which it does here... so yes, it should be a reasonably dark grey.

The white of the forum background is pure white here... no trace of yellow, if anything... slightly cool, with a slight hint of magenta: The result of the CCF backlighting not being spectrally pure and being recorded by the camera in a way the eye does not, and the fact that the polarisation filter in the monitor panel will always cause colour shifts when you photograph a monitor... but it's only slight here... to all pretence and purposes, whites are white when viewed here - absolutely no trace of yellow.

st599 said:
It'll only look correct if the room lighting is correct too.
Click to expand...

Room lighting is 6500K balanced in the room the image was taken. If you mean it will only look right for anyone viewing the image if they have correct lighting, that is true... but highlights are definitely NOT yellow in that image.

Eddzz!! said:
Okay... Here is an image edited by myself this evening on my "calibrated" monitor to what I think is a pleasing result. Tell me what you guys think.

From San Marco by Edward Reynolds, on Flickr
Click to expand...


Have you gone for pleasing, or accurate? It has a strong cyan/yellow cast here, which gives it an overall green feel. However, as I have no image of the same place, taken at the same time, I have no standard to judge it by. Remember, "pleasing" is subjective - accurate is not.

Make sure you are using controlled lighting the same as mine (6500K), I use one of these... http://www.bltdirect.com/energy-sav...watt-daylight-bayonet-cap-60-watt-alternative which are actually pretty damned good. Take a test shot, including a 18% grey card. Edit it so the grey card on your screen is identical to the real card when held in the same place with the same lighting in real life (best to do it at your workstation as I have done) and then post that. If your screen is accurate, the grey card in your shot should be identical to the one in mine. (But even this will only work if we both use the same room lighting, so if you have 3400k "normal" room lighting yours will still look different to mine).

This is the only way to be sure - if we both photograph something of a known, accurate standard, edit it so what's on our screens is the same as our eyes perceive it, then compare results. If our screens are calibrated the same, the images will be the same. Even those with uncalibrated screens that are not accurate will perceive them as the same - inaccurate... but the same.

(Do not auto white balance by using the dropper tool in lightroom by clicking it on the grey card - This WILL give an accurate result, but it will do so regardless of how accurate your monitor is. Do it VISUALLY by judging the grey card on your screen against the actual grey card when held against your monitor and making manual adjustments to get them looking the same - this way you are testing the accuracy of your screen's calibration.)

By posting an image of something I can't compare it to here (like the above image of San Marco), it becomes subjective. You have to use a calibrated standard of some sort, so a scene with a grey card in it is ideal, as you can then adjust the image so that what is on your screen looks identical to the actual card when held next to the monitor (assuming the lighting of the scene has not changed).

Also... remember to correctly convert the image to sRGB using "Edit/Convert to Profile" in Photoshop before posting - do not use "Assign Profile".

That sounds like a pain in the arse, but it's the ONLY way to be absolutely sure that it's accurate. If you want to test whether one person's screen is the same as someone else's, then you either A) Have to be there to compare the two directly, or B) control every aspect of a test image's production to establish a standard by which to judge the output. Any differences in room lighting and test object being photographed will make the results meaningless.

I'd be giving that old Spyder 3 some suspicious looks if your highlights are yellow in my image. How old is the screen?
 
Last edited:
This is possibly the most frustrating thing I've ever tackled with photography lol. I can't photograph you a grey card because I don't have any! The Spyder always attempts to calibrate to a white point of 6500K. If I let it take a reading of ambient light it wants to calibrate to 5800K which is warmer still! The whites on screen definitely have a yellow tinge to them, no doubt. The monitor isn't old either. It's a Dell U2412m. Maybe about three years old? Seriously stuck and on the verge of giving up with it.
 
I've f*cking had enough of this now... Calibrated again and my whites have gone from yellow-ish to quite red-ish. I'm clearly doing something wrong but there is no telling what.
 
Eddzz!! said:
This is possibly the most frustrating thing I've ever tackled with photography lol. I can't photograph you a grey card because I don't have any! The Spyder always attempts to calibrate to a white point of 6500K. If I let it take a reading of ambient light it wants to calibrate to 5800K which is warmer still! The whites on screen definitely have a yellow tinge to them, no doubt. The monitor isn't old either. It's a Dell U2412m. Maybe about three years old? Seriously stuck and on the verge of giving up with it.
Click to expand...

Ideally, you should calibrate to whatever the ambient light temperature is, so if it's saying yours is 5800, then that's what it should be, BUT..... this only holds true within a small variance from what is considered "daylight". If you do that at night with normal warm room lighting, and it measures it at 3400K, then calibrating your screen to 3400K will look massively warm and yellow.

The best way, although it's a hassle, is to just control the room lighting. Get a black out blind for daytime use, and then use a 6500K daylight balanced light source in the room. That way you can control this aspect fully. You know your room lighting is 6500K and you can calibrate to a white point of 6500K. Sorted.

The other issue is that you seem to be having problems with that spyder3.
 
Eddzz!! said:
I've f*cking had enough of this now... Calibrated again and my whites have gone from yellow-ish to quite red-ish. I'm clearly doing something wrong but there is no telling what.
Click to expand...


I suspect that Spyder. Even if it's not accurate... there's no reason it shoudl be giving you different results each time you calibrate.

6500K
100cd/m2
ICC v2.0 profile.
Switch off all that ambient light measuring crap too.

If it's still giving you grief, then I strongly suspect the calibrator is the problem. The U2412M calibrates well. My wife has one, and I use the i1 Display Pro with the X-Rite software it comes with, and the results are identical to when I use the same calibrator on my Eizo with the Eizo Color Navigator software.

Pity you're so far away.... I'd just come and calibrate the damned thin for you! LOL
 
Last edited:
Pookeyhead said:
I suspect that Spyder. Even if it's not accurate... there's no reason it shoudl be giving you different results each time you calibrate.

6500K
100cd/m2
ICC v2.0 profile.
Switch off all that ambient light measuring crap too.

If it's still giving you grief, then I strongly suspect the calibrator is the problem. The U2412M calibrates well. My wife has one, and I use the i1 Display Pro with the X-Rite software it comes with, and the results are identical to when I use the same calibrator on my Eizo with the Eizo Color Navigator software.

Pity you're so far away.... I'd just come and calibrate the damned thin for you! LOL
Click to expand...

Thanks for all the help...

I would say that under daylight the whites are better... 6500K and there is a slight magenta cast on whites (more noticeable on greys). The brightness is calibrated to 120cd/m2 (that was determined by the calibrator). Not sure what you mean by ICC v2.0 profile? I have the ambient light setting turned off, as this causes no end of issues with the changing light in my office.
 
That changing light will cause you issues anyway as it will play merry hell with your colour acuity. If this is a home office, get black out blinds and a decent light source. The lamp I linked to above somewhere is ideal.
 
Pookeyhead said:
That changing light will cause you issues anyway as it will play merry hell with your colour acuity. If this is a home office, get black out blinds and a decent light source. The lamp I linked to above somewhere is ideal.
Click to expand...

Guess I'll go hunting for a 6500K light bulb then! What was it you meant by ICC v2.0 profile? Also, would you recommend I use 100cd/m2 white point despite my software recommending 120cd/m2?
 
I have the same model and when first calibrated the monitor brightness was reduced quite a lot . I can't say that the whites changed that much and when I recalibrate it doesn't tend to change much either . Not really sure how you can test the device to ensure it's working correctly .
 
Eddzz!! said:
Guess I'll go hunting for a 6500K light bulb then! What was it you meant by ICC v2.0 profile? Also, would you recommend I use 100cd/m2 white point despite my software recommending 120cd/m2?
Click to expand...

Don't bother hunting for one... use the link I posted further up from BLT Direct. They're cheap and reliable, and the colour temp is accurate.

v2.0 profiles just play nicer with Windows is all. If you can see no option to change this in the Spyder software, and considering it is a Spyder3, then it's almost certain to be V2.0 already.

I use 100cd/m2 because it's a better match for the brightness of my room lighting. As a rough guide:

Dark room with no lighting = 80cd/m2
Dim to average lit room = 100cd/m2
Average to reasonably lit room = 120 cd/m2

Brightly lit room or bright daylight = Don't edit in such conditions.

If the spyder is recommending 120 after measuring your ambient lighting, then do as it suggests. If it is recommending for no apparent reason, then consider your room lighting and use the rough guide above.

Fit one of those little 11w lamps into a desktop lamp... something like this.... http://www.bltdirect.com/eglo-basic...Q253il2leCS2GAAjaxo2VlRUMCPzubd5GQBoCesTw_wcB

...and fit the bulb, you'll have a reasonable room lighting. I'd not have it lighting the desk directly, but have it aimed at a white wall and use the reflected light instead. That is a ES screw mount, but you can get those lamps in that too.
 
BigJohn said:
Not really sure how you can test the device to ensure it's working correctly .
Click to expand...

Unless you have another monitor and/or calibrator available, you can't really. All you can do is go to the lengths I recommend above.

@Eddzz!! If the screen is new, or relatively new, I would be suspecting the Spyder3. My wife's U2412M calibrates superbly, which is one of the reasons I always recommend it to those who want a fantastic screen on a budget.
 
If that's a Kodak grey card you have David, they're not guaranteed neutral grey. Mine is a bit green, though it's 18% reflectance. The white side is guaranteed neutral for white balancing. A lot of so-called grey cards are not neutral and don't actually claim to be (check the wording).

I find the background in Lightroom a very useful neutral grey reference.

Edit: it would be interesting to know what the background used here is. I use the 'Evening' setting - scroll down to bottom/left the page to change.
 
Last edited:
HoppyUK said:
---clip---

I find the background in Lightroom a very useful neutral grey reference.

---clip---
Click to expand...

How does that help? Surely it's dependent upon the monitor settings (which is what is being calibrated) or am I being thick here? :(
 
chuckles said:
How does that help? Surely it's dependent upon the monitor settings (which is what is being calibrated) or am I being thick here? :(
Click to expand...

Maybe not that useful! I use it as a visual reference to compare things that I know should be white, or neutral grey, or black. It's easy to see if there's a slight tint one way or the other.

It doesn't matter if your monitor is calibrated or not - if it's the same hue, then it's neutral.
 
Last edited:
HoppyUK said:
If that's a Kodak grey card you have David, they're not guaranteed neutral grey. Mine is a bit green, though it's 18% reflectance. The white side is guaranteed neutral for white balancing. A lot of so-called grey cards are not neutral and don't actually claim to be (check the wording).

I find the background in Lightroom a very useful neutral grey reference.

Edit: it would be interesting to know what the background used here is. I use the 'Evening' setting - scroll down to bottom/left the page to change.
Click to expand...

Would you recommend a particular grey card? I need to get some anyway...

Also @Pookeyhead, here are a couple of shots edited on my calibrated monitor. They looked fine on my monitor and they look just as good on my uncalibrated monitor at work. Both I applied a slight magenta tint to.

http://www.reynoldsphoto.co.uk/travel-landscape/dmub5q1nr5v47ddedfe97v22gucafv

http://www.reynoldsphoto.co.uk/travel-landscape/21vv7x416pg5nvf8r1y94tuilppbus
 
Last edited:
Would you recommend a particular grey card? I need to get some anyway... <snip>[/QUOTE]

I use a Macbeth, as it is often known - actually an X-Rite Color Checker Classic. It's a universally used reference, good for calibrating the camera, too. I keep meaning to get the little Passport version that includes a grey card http://www.amazon.co.uk/Pantone-X-R...2647951&sr=1-3-catcorr&keywords=macbeth+chart

For something bigger, I also use a Lastolite TriBalance, and would quite like one of their smaller Ezybalance 12% jobbies.
Click to expand...
 
Last edited:
Eddzz!! said:
Would you recommend a particular grey card? I need to get some anyway...

Also @Pookeyhead, here are a couple of shots edited on my calibrated monitor. They looked fine on my monitor and they look just as good on my uncalibrated monitor at work. Both I applied a slight magenta tint to.

http://www.reynoldsphoto.co.uk/travel-landscape/dmub5q1nr5v47ddedfe97v22gucafv

http://www.reynoldsphoto.co.uk/travel-landscape/21vv7x416pg5nvf8r1y94tuilppbus
Click to expand...

I use a Macbeth, as it is often known - actually an X-Rite Color Checker Classic. It's a universally used reference, good for calibrating the camera, too. I keep meaning to get the little Passport version that includes a grey card http://www.amazon.co.uk/Pantone-X-Rite-ColorChecker-Passport/dp/B0036ZBUOG/ref=sr_1_cc_3?s=aps&ie=UTF8&qid=1432647951&sr=1-3-catcorr&keywords=macbeth+chart

For something bigger, I also use a Lastolite TriBalance, and would quite like one of their smaller Ezybalance 12% jobbies.
 
HoppyUK said:
I use a Macbeth, as it is often known - actually an X-Rite Color Checker Classic. It's a universally used reference, good for calibrating the camera, too. I keep meaning to get the little Passport version that includes a grey card http://www.amazon.co.uk/Pantone-X-Rite-ColorChecker-Passport/dp/B0036ZBUOG/ref=sr_1_cc_3?s=aps&ie=UTF8&qid=1432647951&sr=1-3-catcorr&keywords=macbeth+chart

For something bigger, I also use a Lastolite TriBalance, and would quite like one of their smaller Ezybalance 12% jobbies.
Click to expand...

Well... That's rather expensive!
 
Eddzz!! said:
Well... That's rather expensive!
Click to expand...

Well, yes ;)

But it's kinda important to get the key reference right, that everything else hangs on. Don't know if the cheaper clones on Amazon are as good, quite likely, but how do you know? My X-Rite Classic gets well used and abused but is still as good as new. The colour patches are mounted on a robust plastic backing and the actual patches seem quite tough too. Plenty of life left in it. It would imagine that the little Passport version would last many years, so not that expensive in the great scheme of things.

It's easy to get carried away and obsessive about this kind of thing, though if strict colour accuracy is important (as opposed to just looking good/attractive) then you really need to go through the whole routine with a calibrated camera for starters, right through to monitors and viewing light, and to printers etc.
 
Last edited:
HoppyUK said:
If that's a Kodak grey card you have David, they're not guaranteed neutral grey. Mine is a bit green, though it's 18% reflectance. The white side is guaranteed neutral for white balancing. A lot of so-called grey cards are not neutral and don't actually claim to be (check the wording).
Click to expand...

It's close enough. I've got quite a few gray cards here, some Kodak, some others... they are extremely close. Whether it is neutral is not really as important as both parties using the SAME grey point for this exercise.

HoppyUK said:
I find the background in Lightroom a very useful neutral grey reference.
Click to expand...

As someone else has pointed out, that would be useless as a reference if your monior is not calibrated.
 
Eddzz!! said:
Would you recommend a particular grey card? I need to get some anyway...

Also @Pookeyhead, here are a couple of shots edited on my calibrated monitor. They looked fine on my monitor and they look just as good on my uncalibrated monitor at work. Both I applied a slight magenta tint to.

http://www.reynoldsphoto.co.uk/travel-landscape/dmub5q1nr5v47ddedfe97v22gucafv

http://www.reynoldsphoto.co.uk/travel-landscape/21vv7x416pg5nvf8r1y94tuilppbus
Click to expand...


Both those links have an empty frame here.

However, the images on your website seem fine to me. Then again, without a known reference, the colour balance can be subjective. They appear fine for luminance and gamma however, and so seem neutral in colour.
 
Pookeyhead said:
It's close enough. I've got quite a few gray cards here, some Kodak, some others... they are extremely close. Whether it is neutral is not really as important as both parties using the SAME grey point for this exercise.
Click to expand...

Yes, close enough I'm sure - and you'd have noticed if it wasn't ;) The point really is that grey cards are primarily intended as an exposure reference, and may not therefore always be absolutely neutral (my Kodak grey card certainly isn't) and white is often recommended for white balance. But if you can kill two birds so to speak, that's cool.

As someone else has pointed out, that would be useless as a reference if your monior is not calibrated.
Click to expand...

Using the monitor background as a neutral reference is maybe more useful to me for the kind of technical work I often do. But it is certainly handy to know if you have a grey card reference in your shot as you can see in an instant if it's right and it's independent of screen calibration (within reason).
 
HoppyUK said:
Using the monitor background as a neutral reference is maybe more useful to me for the kind of technical work I often do. But it is certainly handy to know if you have a grey card reference in your shot as you can see in an instant if it's right and it's independent of screen calibration (within reason).
Click to expand...

You are mistaken. If you balance to a grey card, and the forum background is neutral, then they will look the same regardless of your monitor calibration. Both are being viewed on your monitor, so if your monitor is out, it's a useless reference.
 
Pookeyhead said:
You are mistaken. If you balance to a grey card, and the forum background is neutral, then they will look the same regardless of your monitor calibration. Both are being viewed on your monitor, so if your monitor is out, it's a useless reference.
Click to expand...

We're talking at cross purposes and you are reading too much into what was simply intended as a useful tip. I am not mistaken, not in the way I meant - as you have just explained yourself! It's not a calibration tool, but a handy reference, or a double-check if you like nothing more, precisely because both are being viewed on the same monitor.
 
HoppyUK said:
We're talking at cross purposes and you are reading too much into what was simply intended as a useful tip. I am not mistaken, not in the way I meant - as you have just explained yourself! It's not a calibration tool, but a handy reference, or a double-check if you like nothing more, precisely because both are being viewed on the same monitor.
Click to expand...

The OP is trying to decide if his monitor is correctly calibrated. Establishing whether his grey card is the same colour as the forum background (which it's not) won't help him.
 
Pookeyhead said:
The OP is trying to decide if his monitor is correctly calibrated. Establishing whether his grey card is the same colour as the forum background (which it's not) won't help him.
Click to expand...

Yes yes, okay.

Instead of arguing over nothing, go and check that your unverified grey card is in fact neutral grey. That's quite important.
 
The background colours (I know it may be moot now of course :))..... are

#b8b8b8 (184,184,184) and
#dddddd(221,221,221)

:)
 
Marcel said:
The background colours (I know it may be moot now of course :))..... are

#b8b8b8 (184,184,184) and
#dddddd(221,221,221)

:)
Click to expand...

Thanks Marcel. Moot maybe, but still interesting.

In English, is that equal red, green, blue - ie neutral grey?
 
Eddzz!! said:
Well... That's rather expensive!
Click to expand...

Eddzz!!, please be aware that suddenly after calibration, it is normal that you might see a color tint, as your eyes need some minutes to get used to the new color temperature.

Ambient light compensation is a bad thing if you don't really look after your room light, but is perfect if you're taking care of it. Therefore, I would suggest the following:
Always work in dark room conditions when working on photos. So darken down your room and do a calibration with the ambient light compensation activated and accepting its suggestions. Do that with these calibration settings: Gamma 2.2, 6500K, 120 cd/m2. Gamma won't change, the rest might change upon the suggestions of the ambient light sensor.

Then try again. How about that?
 
You must log in or register to reply here.
Back
Top