Monitor: Dell or Eizo?

@Pookeyhead I'm not using any of these screens myself. I'm on a 27" imac at the moment, but consisting the pros and cons of upgrading to a decent system late in the year.

In aware IPS glow is the nature of the beast, I've just read a lot of people complaining the 2713h had poor QC variations from panel to panel, and that back light bleed was also an issue with them. Had a lot of people saying NEC take things more seriously and you're less likely to get a lemon. And have been told the PA272W is the way to go. Of course that's a lot of cash compared to the Dell, so was wondering if the differences are going to be that noticeable. (Also heard people complaining about red glow on the NEC from the different back lighting).

I wouldn't worry about the doom sayers on the internet. If there's a problem with the Dell screen, their after sales service is second to none... certainly better than Eizo's! I've read nothing from serious sources about backlight bleed, and the one I used certainly had none. There was some slight unevenness in illumination at the edges, but you'll only really notice this if you sit staring at a grey screen or something. It's the sort of thing a review would be obliged to mention, and therefore set up all manner of panic on the interwebz, but nothing you'll actually have a problem with in real life :)


Not used the PA272W... it's a fairly new screen. As with the other PA screens such as the outgoing PA271W (which I have used), there's a difference between the versions available in US & Canada and Europe. In the US, you can purchase the PA series with an upgrade kit (Spectraview upgrade kit) that allows for hardware profiling, but can not in the UK. This is locked out at Firmware level. I'm still trying to confirm that the PA272W is subject to this, but this is what NEC say about the issue with the older PA271W.

NEC said:
Also, importantly for calibration the PA271W regular edition and PA271W-SV have a firmware difference which allows for different levels of calibration. The PA271W only allows you to use the European software to calibrate at a software (graphics card LUT) level alone, or at a hard/software level which also controls limited monitor settings such as brightness and the RGB channels automatically. It does not however control the monitors 14-bit LUT and the option for "hardware calibration" within the software is greyed out.

So it's very likely that the same will be true for the PA272W. Especially as the US website advertises the PA272W-SV version, which is the one with the firmware allowing hardware calibration, but the UK website makes no mention of it whatsoever. Looks like the UK version is software calibration only. I have read (can not confirm this personally) that getting hold of the US Spectraview software will allow hardware calibration of the PA series. You'll have to do your own research, or risk that yourself.

It's still a great screen either way, but to be honest... the ability to hardware profile is not to be underestimated. Hence the Dell + i1 Display Pro combination is so tempting.
 
Run the eyedropper tool (or whatever it is called) where the banding is and see what the RGB values are. If it is in the image, there will be a significant RGB change over the banding point. If there isn't a big change, it's your monitor and how it is setup....

Thanks for the advice Andy (re the banding mentioned in my post 34). Actually I think some of the banding might be something to do with this https://forums.adobe.com/message/3475331#3475331.

Also discussed here https://forums.adobe.com/thread/1054238.
 
Yeah.. this is why profiling at 8bit software is not as good as using a monitor that can be hardware profiled at 12 or more bits. Banding is usually a fact of life if you software profile at 8bit.
 
Yeah.. this is why profiling at 8bit software is not as good as using a monitor that can be hardware profiled at 12 or more bits. Banding is usually a fact of life if you software profile at 8bit.

In the first link above they seem to be saying it's more to do with Photoshop settings (OpenGL) rather than monitor profiling. If profiling is also a culprit I don't know what I can do about it (I've profiled this CG241 with a ColorMunki Photo).

PS. What sort of over-processing would cause banding? (your post 36)
 
Last edited:
I know.. but you'll always get some banding if you software profile... you're pushing things around within an 8 bit space with no room to spread out, so you're just bunching up the level in some place, and creating holes in the histogram in others. The banding is not on the actual file though, so it's a good trade off for colour accuracy... but 8bit software profiling does cause banding. This is why high end screens allow you to program the monitor's LUT at a much higher bit depth.
 
Last edited:
Thanks David.

Which high end screens are you referring to? Is your 30" Eizo one of them?
 
Thanks David.

Which high end screens are you referring to? Is your 30" Eizo one of them?


Yes.. but many are able to be hardware calibrated. Certain Dell screens can be too, namely the U2413, U2713H, UP3214Q.


[EDIT].. forgot to say the Dell screens can only be hardware calibrated with the x-rite i1 Display Pro calibrator and Dell's own software (manufactured by x-rite)
 
Last edited:
My CG241 supposedly has 12bit hardware calibration (link). I use the ColorMunki Photo with ColorNavigator but I don't know if this combination is calibrating the LUT to 12bit or 8bit.
 
Incidentally, if you want to check if your display/workflow is band free, download this TIFF file.


https://dl.dropboxusercontent.com/u/23953768/gradient test.tif

Open in PS. Don't use Windows' built in image viewers.. they suck.

Make sure it is loading as 16bit (in PS - Image/Mode), and make sure you are viewing the image at 100% (In PS press Ctrl+1)

If form some reason it has opened as a 8bit file, converting it to 16bit will NOT work.

Viewing at any other zoomed size could result in Photoshop's aliasing adding banding.. especially if using Open GL/GPU acceleration.

Once open as a 16bit file, and viewed at 100% (1:1 pixel mapping) this should be an utterly smooth transition from white to black wit absolutely no banding or artefacts.. just as if it was sprayed from a real life airbrush. Any banding is a result of something manipulating your 8 bit output.

If you have software calibrated to adjust your GPU's LUT, unless you are extremely fortunate, you will have some banding... don't worry about it.. it can't be helped. This is a mathematically linear gradient, and in real life, such things don't really exist, so it doesn't mean you'll get banding in reality... except with the possible exception of some skies perhaps. Any banding you see however, will NOT be on the file itself.... unless you have over processed your images.. it will just be on your display.


This is why if you have banding on an image, getting someone else to check the file may give a false positive, because they may also have a manipulated 8 bit display. The only real test is to ask someone with a hardware calibrated screen, or a true 10bit workflow, or print it.
 
Last edited:
My CG241 supposedly has 12bit hardware calibration (link). I use the ColorMunki Photo with ColorNavigator but I don't know if this combination is calibrating the LUT to 12bit or 8bit.


Yes.. all Eizo Color Edge display have programmable LUTs. and using the color Munki with Eizo's Color Navigator will be programming the monitor's LUT... ONLY if you plug the Color Munki into the monitor's USB ports... NOT the computer's.
 
Last edited:
Incidentally, if you want to check if your display/workflow is band free, download this TIFF file.


https://dl.dropboxusercontent.com/u/23953768/gradient test.tif

Open in PS. Don't use Windows' built in image viewers.. they suck.

Make sure it is loading as 16bit (in PS - Image/Mode), and make sure you are viewing the image at 100% (In PS press Ctrl+1)

If form some reason it has opened as a 8bit file, converting it to 16bit will NOT work.

Viewing at any other zoomed size could result in Photoshop's aliasing adding banding.. especially if using Open GL/GPU acceleration.

Once open as a 16bit file, and viewed at 100% (1:1 pixel mapping) this should be an utterly smooth transition from white to black wit absolutely no banding or artefacts.. just as if it was sprayed from a real life airbrush. Any banding is a result of something manipulating your 8 bit output.

If you have software calibrated to adjust your GPU's LUT, unless you are extremely fortunate, you will have some banding... don't worry about it.. it can't be helped. This is a mathematically linear gradient, and in real life, such things don't really exist, so it doesn't mean you'll get banding in reality... except with the possible exception of some skies perhaps. Any banding you see however, will NOT be on the file itself.... unless you have over processed your images.. it will just be on your display.


This is why if you have banding on an image, getting someone else to check the file may give a false positive, because they may also have a manipulated 8 bit display. The only real test is to ask someone with a hardware calibrated screen, or a true 10bit workflow, or print it.

I've just tried the gradient TIFF (it opened at 16bit) and I zoomed to 100% - I can't tell if it's banding-free or not now (if there is banding it's very subtle).

Here's a screengrab resized down to 800x to fit the forum limit - trouble is the process of resizing itself may have introduced banding, I can't tell that either!

6286-1407420287-a025f8bc3f1faeacb9f26fcc5f779ab6.jpg
 
Last edited:
Yes.. all Eizo Color Edge display have programmable LUTs. and using the color Munki with Eizo's Color Navigator will be programming the monitor's LUT... ONLY if you plug the Color Munki into the monitor's USB ports... NOT the computer's.

I always plug the ColorMunki into the Eizo's up/down USB ports (not into the PC itself) - so it should be programming the monitor's LUT - but is it doing it at 12bit (as per the press release) or 8bit?
 
Last edited:
Note to Paul (the OP): sorry for hijacking your thread!
 
Yes.. all Eizo Color Edge display have programmable LUTs. and using the color Munki with Eizo's Color Navigator will be programming the monitor's LUT... ONLY if you plug the Color Munki into the monitor's USB ports... NOT the computer's.
Does the same apply to the Dell monitors? I've not found anything in the blurb to say one way or the other ...
Note to Paul (the OP): sorry for hijacking your thread!
No probs :)
 
I've just tried the gradient TIFF (it opened at 16bit) and I zoomed to 100% - I can't tell if it's banding-free or not now (if there is banding it's very subtle).

Here's a screengrab resized down to 800x to fit the forum limit - trouble is the process of resizing itself may have introduced banding, I can't tell that either!

6286-1407420287-a025f8bc3f1faeacb9f26fcc5f779ab6.jpg


I can see banding.. but as you said... this is now an 8bit JPEG... and a resized 8bit JPEG.


You can't take a screen grab of this to post on here, because everyone will be seeing it on THEIR screen, not yours, and as soon as you JPEG it A) you're introducing artefacts, and B) it's 8 bit. :)
 
Last edited:
I always plug the ColorMunki into the Eizo's up/down USB ports (not into the PC itself) - so it should be programming the monitor's LUT - but is it doing it at 12bit (as per the press release) or 8bit?


12bit. The screen's LUT is 12 bit, so the 8bit output from your GPU is being modified at 12bit level, so there's room to push things around in your RGB histogram.


Does the same apply to the Dell monitors? I've not found anything in the blurb to say one way or the other ...
No probs :)

only if you use the supplied Dell software, and again, uise the Monitor's USB ports.
 
Last edited:
@Pookeyhead thanks David, I am using the Dell supplied s/w and the I used the monitors usb but not with the knowledge it affected the calibration, only because it was easier to plug the calibrator in than one of the base unit :)
 
Then you're hardware profiled :)
 
Back
Top