How to identify the type of pixel defect on an IPS panel

Jeroen1000

Limp Gawd
Joined
Sep 17, 2010
Messages
266
Everytime I try to do this for someone I get lost at explaining what is happening.

I'm using another thread as a guidance as I've lost my own info: http://hardforum.com/showthread.php?t=1700892

The problem:
Crake8 said:
On solid color checks, one pixel: On solid black background, it is red; on red, it is a pale grey, on blue, it is a pale white; on white, I do not see it - it is white; on green, I do not see it - it is green.

As far as I can tell this statement should be correct:
BearOso said:
PSA: On a transmissive-by-default panel like IPS or PVA, a bright pixel is a dead pixel. "Common knowledge" about stuck and dead pixels should be inverted when not discussing TN.

If your pixel is bright at any time when it's supposed to be dark, Dell should replace it. If you get a bad support rep who refuses you then ask for his/her supervisor, as their guarantee should definitely apply to what you describe.

(trying to) Explaining it all:


I'm assuming that the for the troublesome pixel in question, the RED subpixel is broken => it remains always on => it keeps transmitting RED even when is it supposed to be OFF

In order to display white all three subpixels must be ON (=transmitting light?). In additive color mixing this produces white. All pixels are capable of turning on in above scenario so displaying white still works fine.

In order to produce black all three subpixels must be OFF (blocking light?). However, the RED one remains in the ON state here. So when displaying black, the defective RED pixel stays lit.

And then I run into issues:

When displaying RED one would assume the defective pixel turns RED like all the others. But the poster states it turns into a pale grey.


When displaying GREEN one would assume the pixel would turn YELLOW. As GREEN + RED = YELLOW.
Yet the poster says GREEN displays just fine. This is not possible when the RED pixel is always ON.

When displaying BLUE one would assume the pixel would turn MAGENTA. As BLUE+ RED = MAGENTA.
Yet the poster says BLUE turns out to be a pale WHITE.

All kidding aside, I did have fun with the coloured text, but does anyone know why he is seeing these things? There must be some logic that works.

Cheers,
Jeroen
 

Mr.Pixel

Limp Gawd
Joined
Oct 30, 2010
Messages
138
Kind of a basic question, but are you sure the person describing the colors is not color-blind in some way?

If not, perhaps some of the pixel wires might be incorrectly connected in an unusual way.
 

Jeroen1000

Limp Gawd
Joined
Sep 17, 2010
Messages
266
I'm kind of hung up on that TN logic is the inverse for IPS and MVA when it comes to pixel defects.
No chance that has any to do with it?

Maybe my assumption that the RED pixel is the faulty one is wrong...
Maybe I can hunt down a dead pixel screen at work lol
 

Mr.Pixel

Limp Gawd
Joined
Oct 30, 2010
Messages
138
Actually, IPS panels should block light by default when "off", and wikipedia seems to agree with this statement. However, it doesn't matter how a pixel is stuck dark or bright as all the additive color properties still work like they should.

Your logic with the colored text appears fine, and that's how the pixel should theoretically behave if the red subpixel is stuck bright and the other 2 work as expected. I asked if the person was colorblind because if they can't differentiate between red and green that would explain most of the discrepancy between your theory and their observations. For example, to someone with red-green colorblindness RGB 255,0,0 would appear to be like RGB 0,255,0, possibly explaining why green "appears" fine.

It is possible for defects in the panel to send a signal (or even a partial signal) to an incorrect subpixel, but that type of defect is rarer than the usual stuck bright/dark types.
 
Top