Dell U2410

[Solstice]

sRGB mode isn't hot for calibrating sadly. Do a factory reset and re-do the calibration process, also check to see if the colour controls or 'enhancements' in Catalyst Control Centre are turned off.
This monitor should have no problems getting close to D65 white, 2.2 gamma and DeltaE < 2, should get a refund if it won't, obviously defective unit.

 

[Sailor_Moon]

Custom mode shares the main problems with the Standard mode. It has a gamma curve around the 2.0 curve. It gave me the possibility to set the RGB Gains but it produces very bad color accuracy (dE94.max above 6.0)

That's because the display behaves nonlinear in this mode. A matrix profile with the actual colorimetrical values would be defective so your calibration software makes a "forced normalisation" which leads to these high deviations in a simple profile validation.

I can notice every dark bar on the 7 bit grayscale now. (8 bit is too much for a 8+2 bit display. Hahhh, ok. But I can live with this...)

We should add that the panel is still a 8bit version with internal FRC (which makes no difference in a 8bit workflow compared to a 10bit panel (whose FRC stage would be bypassed with 10bit feed)).

Btw, what probe and software have you used? The DELL has a WCG-CCFL backlight what makes calibration with a colorimeter tricky (link).

- I want a gamma 2.2 instead of the sRGB curve

In sRGB mode a sRGB gradation would be good. Our display hasn't reached that characteristic (see grey line in the linked graphics; it remained the same in AdobeRGB mode). But apart from that factory calibration for sRGB and AdobeRGB mode was quite accurate and showed good neutrality (link).

Best regards

Denis

 

[janos666]

sRGB mode isn't hot for calibrating sadly. Do a factory reset and re-do the calibration process, also check to see if the colour controls or 'enhancements' in Catalyst Control Centre are turned off.

I spent two days to find the best settings. I can't find any other acceptable solution expect this calibrated sRGB preset.

Btw, what probe and software have you used?

The hardware is an EyeOne Display 2 with Basiccolor 4.1.20 software.

In sRGB mode a sRGB gradation would be good.

Yes I would be very happy about it in a perfect world. It would be nice if there would be any other preset which I can use with gamma 2.2 and/or developers/artists would follow the sRGB standard. PC Games are usually made with gamma 2.2, so there will be a noticeable difference when I play on a real sRGB monitor. Pictures are usually edited on displays which were calibrated to gamma 2.2
sRGB is a PC standard but very few people use the real sRGB tonal response curve. It is nice from Dell but it is not good for me today.

 

[Sailor_Moon]

sRGB is a PC standard but very few people use the real sRGB tonal response curve. It is nice from Dell but it is not good for me today.
Reply With Quote

Are you sure that your display has a sRGB gradation? If you look at the results of our test display you will see that gradation couldn't be perfectly described trough a gamma of 2.2 but was even farther away from a sRGB characteristic.

The hardware is an EyeOne Display 2 with Basiccolor 4.1.20 software.

I don't think that Basiccolor implements any generic corrections like iColor Display (Quato). The filter characteristic of the EOD2 matches the CIE standard observer slightly better than the DTP94 (but inter-instrument agreement is worse) but corrections are still necessary.

Best regards

Denis

 

[janos666]

Are you sure that your display has a sRGB gradation?

I have much smoother LUT curves (near to zero) and more noticeable shades on the 8 bit grascale (near to black) with sRGB calibration. But this could be an error...

EDIT: Another question: Do I really need a wide-gamut compatible (or compensated) probe to calibrate the sRGB preset (it has an sRGB gamut...)?

I could produce a nice LUT curve with some manual iterations but the contras ratio dropped to an 540:1 value again. I started from 255-255-255 RGB values and default 50% contrast and I balanced the RGB Level graph in HCFR by setting the RGB values in the service menu and the contrast in the user menu. Calibration LUT curves are not perfect but they look nice enough (and I could use gamma 2.2 now).
I can distinguish the first two bars on the 7 bit grayscale (but not on 8 bit and the overal gradient is not too nice...), sRGB gamut coverage is better now, and 100% white looks really white now (instead of greenish/bluish shade). So, it wouldn't be that bad if contrast could be higher. A 540:1 ratio is somehow lower than I expected from this display.

But this is only the sRGB preset. I think I should spend another few hours to make it through with the Standard preset. (And the changing between them would be too difficult for everyday usage. - Not only a preset with a hotkey and an ICC profile but brightness, contrast setup in the OSD...) But the emulated gamut coverage is good enough now and I want to use the sRGB color space anyway. So, I will keep this state until I can examine a HP LP2475w.

May be I could make a good display of this crap if I could overwrite the internal LUT (manually).

 

[janos666]

So, I learned that every available colorimeter need a correction matrix for the specific wide-gamut display but it will do a better job with displays than the spectrometer itself which should be used to create the correction matrix because they can't do correct measures on dark areas.

Does anyone have a correction matrix for the EyeOne Display 2 and U2410 combination?

Does anyone can tell me how can I make it for myself?
I think I can borrow a spectrometer for this job. But this won't happen before I have to make the choise about my U2410.

I know that this will change the white balance. But will it effect the average gamma curve?
I think that the RGB levels will keep their strange characteristics: short chaotic start, long nearly-coherent middle stage and a chaotic end.
I tried to use the Generic S-IPS compensation in iColor and white balance changed a lot but LUT curves still have the chaotic ending (the same attitude with different arrangement). 100% white is better now but it is still not perfect. May be because I set the RGB Gains to max and the LUT had to change the 100% white. But I have to do so if I don't want to loose too much contrast. (Setting the white point with RGB Gains isn't optimal because RGB levels have a chaotic behavior near to 100%...)
I could erase the chaotic end stage by setting the contrast (in the OSD) to 37% but that reduced the contrast ratio as well...



I think I get it. Both instruments produce sensor data with three 3D color coordinates (XYZ for RGB colors) which is a 3x3 matrix. So, I should create a 3x3 matrix which transforms the bad one to the right one. And I think I should apply my own values here:

}
"eye-one display 2 for Wide Gamut S-IPS (generic)" =
{
Name = "eye-one display 2 for Quato 220 excellence"
Device = "eye-one display 2"
Display = "Wide Gamut S-IPS (generic)"
ReferenceDevice = "eye-one Pro Rev.D"
MatrixXYZ = "3 3 1482250784 1.020387 -0.033715 0.012371 0.020156 0.988765 -0.000819 -0.025479 0.045097 1.043297 "
}

Does anyone have an excel or mathcad file for this job? It was some years ago when I learned math.

Should I use the data from the full native gamut (max RGB values in the OSD) or after white point balancing (with the hardware RGB gains)?
I think I should use the full RGB values and let the LUT to do it's job. I found it as a good idea with 10+ bit/color connection (at least on the achievable contrast ratio side and with this display).

 

[Sailor_Moon]

I know that this will change the white balance. But will it effect the average gamma curve?

No.

I think that the RGB levels will keep their strange characteristics: short chaotic start, long nearly-coherent middle stage and a chaotic end.

It's not that I don't believe your measurments but I'm still wondering because our U2410 and U2711 were quite neutral in sRGB mode and gradation could be ~ expressed through a Gamma of 2.1 (both U2410 we had tested). Could be of course variances regarding factory calibration.

I think I get it. Both instruments produce sensor data with three 3D color coordinates (XYZ for RGB colors) which is a 3x3 matrix. So, I should create a 3x3 matrix which transforms the bad one to the right one. And I think I should apply my own values here:

Yes. The generic correction in iColor works in most cases already quite good when using a DTP94. They have averaged over some screens and probes of one type.

Does anyone have an excel or mathcad file for this job? It was some years ago when I learned math

If you send me the reference data and the measurment of your EOD2 (primaries) I can provide you with the matrix that you then can use in the way you have already described. You can try to use the generic correction in iColor too but the poor inter-instrument agreement of the EOD2 is a limiting factor here. The only affordable solution for the reference measurement is the EyeOne Pro (although sampling intervall is quite coarse; fogra recommends <= 5nm but there is no affordable device with that charateristic). Maybe you can borrow it.

Should I use the data from the full native gamut (max RGB values in the OSD) or after white point balancing (with the hardware RGB gains)?

The conditions for both measurements must be equal. We have for example simply used the "untouched" standard mode.

but it will do a better job with displays than the spectrometer itself which should be used to create the correction matrix because they can't do correct measures on dark areas.

If you already have an EyeOne Pro available I would just use it. Yes - it has some drawbacks in dark measurements that I wouldn't call severe. You don't have to fiddle with additional corrections and have a device with good inter-instrument agreement and long-term stability.

Best regards

Denis

 

[janos666]

It's not that I don't believe your measurments but I'm still wondering because our U2410 and U2711 were quite neutral in sRGB mode

I made these measurements with HCFR from the sRGB preset with RGB values at 255 in the service menu (contrast 50 ; brightness 5 in the OSD, clean VGA LUT).
HCFR does not have any correction matrix (but it supports some kind of instrument calibration for the HCFR probe with reference measures. May be it is possible with i1 devices as well, if we ask the developers...) I made these measures before I know about the wide-gamut issue...
I tried to optimize them with the RGB gains but the levels will keep their shape (more or less, but they won't be constant near to the start or end points), so I can set the 100% white and loose a lot of contrast or I can set it to a handmade optimised state. But I couldn't achieve better results with handmade iterations than VGA LUT can do with RGB gains at 255. (I am using DP with 10 bit/color signal...)

I could erase the chaotic end stage by setting the contrast to 37, so I could set the WP with the RGB gains. But it produced a very low contrast ratio as well. (~500:1)

I tried to use the iColor software with Generic corrections but I still have this chaotic end stage on the LUT curves. But I could make some optimisations with the RGB gains according to the LUT curves, so I think I achieved the best state I can. (I reduced the Green gain only because it is the only constant one and it was the lowest LUT curve.)
White looks real white nearly enough (LUT curves ends near to 1;1 point) and I have a contrast ratio about 600:1 which is acceptable. At least until I can replace this display with something else...

If you send me the reference data and the measurment of your EOD2 (primaries) I can provide you with the matrix

Thanks but I need some faster solution. I will go to the person with the spectrometer or he will come to me for some hours, and I want to make sure about the final results. (Or may be he will give it to me for some days, I don't know...) And I want to be sure about the results.
Is it a simple equation like: E * C = R which can be easily solved by MathCAD? (E - EyeOne sensor data, C=Correcton matrix, R - Reference sensor data?)

But I have to decide about this display until tomorrow. I won't have any spectrometer until then. So, I assume that I will be able to set the correct white balance with a spectrometer or a software compensated colorimeter. This U2410 (which I have now) is a crap anyway, isn't it?

What do you think about the HP LP2475w (the final revision, of course)? Would it be a better choice for me?
- I want some H-IPS display with good overdrive. I really like the good response time and I don't want black crush again (with S-PVA panels). But I want good contrast ratio (>700:1).
- I want gamma 2.2 with as good gradients as I can achieve in this price range.
- I want ~100% sRGB gamut coverage. This could be a native sRGB gamut or a wide gamut which fully includes it. I can use CMS supported softwares for pictures and videos. But I need a hardware emulated sRGB space for PC games.
- 48 or 72 Hz (official or unofficial) support would be cool.
Does the latest LP2475w revision supports sRGB emulation (or 72Hz)? I heard that it does (both) but I heard that U2410 is a very good display as well.

So, may be a healthy U2410 could be better for me if LP2475w can't offer sRGB emulation.

Can I use my uncorrected colorimeter to test some U2410 and tell if they has good RGB level (or average gamma) characteristics?

The conditions for both measurements must be equal. We have for example simply used the "untouched" standard mode.

Yes but which would be better? Should I make the measures with RGB gains at 255 or with optimized gains (after I set the gains to achieve D65 white)?
I think I will do it with RGB255 and I will use the LUT for calibration. I think it could be a good practice when you have 10 bit/color output. (This can produce the best contrast ratio with averaged deltas...)

If you already have an EyeOne Pro available I would just use it.

I want to calibrate my display in 1-2 months period, so I don't want to borrow expensive instruments all the time (I may get it for free once, but monthly...)
May be I will hunt for a cheap, used spectrometer. (So I can use them together or I can decide which one should I keep...)


Does warranty covers my problems?
I can produce much worse measures with default factory settings (I mean the the fectory menu...).

 

[Sailor_Moon]

I made these measurements with HCFR from the sRGB preset with RGB values at 255 in the service menu

Can you send me the HCFR file? The analysis of HCFR is in many respects problematic. Keep in mind that measurement of tonal values with very low brightness is inaccurate with a colorimeter too.

Thanks but I need some faster solution. I will go to the person with the spectrometer or he will come to me for some hours, and I want to make sure about the final results.

I'm @home the whole day. Should be no problem.

Yes but which would be better?

Equal - but if you use standard mode I could compare the results with our measurement.

Does the latest LP2475w revision supports sRGB emulation? I heard that it does but I heard that U2410 is a very good display

I've never seen a newer LP2475w but I don't think that they have implemented a sRGB emulation mode.

Can I use my uncorrected colorimeter to test some U2410 and tell if they has good RGB level (or average gamma) characteristics?

Regarding to the false whitepoint the measurements should be correct altogether. This is my experience with the DTP94. If I chromatically adapted the measurements to the same whitepoint they were quite identical. You can at least compare neutrality and gradation "out of the box".

Best regards

Denis

 

[janos666]

HCFR - Uncalibrated_sRGB_full255

I'm @home the whole day. Should be no problem.

Is it a secret? But it doesn't matter, I would thank your help anyway.

I've never seen a newer LP2475w

And which one would you prefer according to your experiences with the given test subjects?
I think I will do some measures on a different U2410 A01 (may be with a borrowed spectrometer), just to be substantial. (If I will have enough time and material...) I can live without it but sRGB gamut would be nice for PC games.


Does the white balance correction affects the measurement of the primary color coordinates?
I extracted the XYZ coordinates from the ICC file (which was created by iColor with generic correction) to plot them on the CIE chart and it is close to the HCFR measures. (But I have made those HCFR measures before the calibration, which can affect the primaries a bit...)

 

[Sailor_Moon]

Is it a secret?

Certainly not, but I have implemented the (simple matrix) calculation in our test software - so it's now only one click. After seeing your results I must admit that there is something wrong.

The second picture shows the DeltaC (chroma difference) regarding the actual whitepoint ("Range" is the largest distance vector between the measurements in the a*b*-plane). Deviations are very high (detection threshold ~0,5). Could be a severe problem with the factory calibration or a problem with the probe. It has nothing to do with the limitations of a colorimeter when measuring a display with WCG-CCFL or LED backlight. To be sure you should really measure again with an other probe.

I extracted the XYZ coordinates from the ICC file (which was created by iColor with generic correction) to plot them on the CIE chart and it is close to the HCFR measures.

There should be a difference.You have to perform a chromatic adaptation to the actual whitepoint (even though it is wrong) because the measured primaries were adapted to D50 for storage in the ICC profile. That is also why you must be very carefull when comparing with the black sRGB triangle in HCFR. It is based on primaries relative to D65.

Best regards

Denis

 

[janos666]

Yes I know that there is a problem with this display. I already wrote an E-mail to [email protected] (right after my first measures). I have found this address when I searched for my local (hungarian) contact informations (and it belongs to a name which I can see on the box as well). There is no answer yet. But I don't think it would be easy to convince them that this is a warranty problem and they should give me another U2410. And that one could have a bad subpixel, ect.
There was some difference but not too big. But yes, I didn't transform it.

EDIT: I recalculated the values and it indicates a much more accurate sRGB emulation. The primaries where not perfect but close enough expect the blue. That color ran out of the HCFR diagram. But this must be an error. I copied the values manually and I used a different chromatic adaptation method than iColor used.
So, the colorimeter correction affects the gamut measurements and this display has a much better sRGB emulation than I thought.

I start to think I will replace it with another U2410.

 

[Motaa]

Claiming "Luckiest U2410 Gambler Ever"

Went on eBay and found some U2410s for $370 + $30 shipping OBO. Offered $300 + $30 shipping and it was automatically accepted.

Grit my teeth waiting for it and wondering if I had made the right choice. Got the monitor delivered today:

Manufactured: May 2010
Revision: A02
No dead pixles. Perfectly even back light. No scratches. Looks like they took it out of a Dell box and put it into another one without cables or papers.

Hoping the monitor is new enough that transferring the warranty into my name will be easy/ possible.

 

[janos666]

Is there an A02 revision?
Is it a real hardware revision or only a new firmware? Can I update it at home?
(I wouldn't do it with this bad one, it won't solve my problem. But if I receive an A01 replacement...)

 

[janos666]

Today, I got a replacement and I was happy until I found a bad sub-pixel. I swear that it wasn't there until we finished with the paperworks. But it doesn't want to disappear again...

It is located near to the center of the screen and it is easily noticeable on the 100% white image (a pixel with magenta color).

This could be only a bad joke.

UPDATE:

I found another sub-pixel error. There is a dark blue pixel on the 0% black test image. This is two traditional sub-pixel error (until now).

But there is more... I can see a plenty of little flickering dots on the 0% black test image. It cannot be noticed from the usual viewing distance but when I take a closer look to the lower-left corner it looks like the sky with the stars.


By the way, this crap has significantly higher contrast ratio (780:1 vs 480:1) and the RGB Level graph is not perfect but much better (it is not an EIZO but nothing to bitch about...) and white is more neutral after a calibration with iColor + generic WCG correction.
So, the last one was sick but this one is much worse with the pixel errors...


When will they ship the A02 revision? May be I can live with these pixel errors for some days. (But not too long...)

 

[janos666]

I just get my third U2410 (second replacement) today.
It is an A02 revision (according to the little mounted paper on the plastic card) The last two was A01, so I think it is some kind of improvement.
It looks much better with it's factory default settings than the last two. The only problem is the huge contamination between the layers. It covers at least four pixels.
So, I am waiting for my next U2410.

I am already tired but I would be very angry if they don't came to me for free.

 

[Yawgmoth]

Hey guys i am considering this monitor Dell u2410 for my next purchase.

I was wondering if it is totally necesary to calibrate this monitor with a calibration tool & colorimeter? Or will it work just fine out of the box?

I don't do Graphic or photoshop or any of that... I only do pc Gaming, Ps3 gaming ( with the monitor) and Movies. Thats it.

What do you say... is calibration needed?.

 

[ilkevinli]

I think every monitor needs to be calibrated

Hey guys i am considering this monitor Dell u2410 for my next purchase.

I was wondering if it is totally necesary to calibrate this monitor with a calibration tool & colorimeter? Or will it work just fine out of the box?

I don't do Graphic or photoshop or any of that... I only do pc Gaming, Ps3 gaming ( with the monitor) and Movies. Thats it.

What do you say... is calibration needed?.

 

[Sailor_Moon]

What do you say... is calibration needed?.

Only reasonable if you have an adequate probe respectively probe-software combination (link). Otherwise I would use the display in AdobeRGB or sRGB mode.

Best regards

Denis

 

[mvalprckeneda]

The 2 2410s I got were pretty close out of the box....calibration brought them as close as possible. I was impressed how close to 6500k they were from Dell.

I used a Sypder 3 Elite. Very happy with it.

 

[Anti-Hero]

Anyone got a good suggestion for a monitor stand for 3 U2410's in landscape mode? Being able to flip them to portrait mode would be a plus, too.

 

[Yawgmoth]

Only reasonable if you have an adequate probe respectively probe-software combination (link). Otherwise I would use the display in AdobeRGB or sRGB mode.

Best regards

Denis

Thanks denis.. So maybe its not a matter of life and death for me to calibrate.. maybe i wont notice a huge difference, i dont know.. i guess ill use it in sRGB mode or adobeRGB as u mention.

 

[Synomenon]

Can we update our U2410's to A02 ourselves / has anyone leaked the update like they did with A01?

 

[shawnoen]

Anyone got a good suggestion for a monitor stand for 3 U2410's in landscape mode? Being able to flip them to portrait mode would be a plus, too.

Ergo Tech makes a desk one that will work with bigger displays like this 24" Dell. They have one that will hold all 3 at one actually.

I'm using Ergotron LX arms for my 2410s...

 

[janos666]

Can we update our U2410's to A02 ourselves / has anyone leaked the update like they did with A01?

My U2410 has the new M1F193 firmware version but I can't find any software which is able to save it to an image file. I will gladly give it to you if you provide me a software.

And if you are already there (to ask/steal it, or I don't know... ) I would ask for a utility which can read and write the internal LUTs. It would be much more useful than this firmware update.

I can't see any firmware related improvements and Dell Support refused to give me a change list.
It has to be a minor update to resolve the side effects of the A01 update or fix for other rare issues. (I heard that it degraded the component video input quality, or something like that.)

The Custom mode is still useless.

The contrast ratio is higher but they say that it depends on the manufacturing date. (It has to be a low-level driver panel attribute.)


Edit:
I think I found an improvement. The internal frame rate conversion is much better with this firmware. Blu-Ray movies looks good in the 1080p24 HDTV mode (with 1:1 pixel mapping in the OSD).
I still prefer the 1920x1200@60Hz mode with a good video renderer which takes care of correct V-sync, but this HDTV mode was absolutely useless with the A01 revision and it is usable now.

 

[Dave L]

I bought a used U2410, but it didn't come with a power cord. What are the specs (amps and volts) of the power cord that originally comes with the monitor?

Thanks.

 

[Imbroglio]

looking to pick up a u2410 to go along side my benq fp241wz.

The only thing is, is that I'd like to get it from dell.com (for warrenty issues) for ~$450 or less if possible. I've seen them go on sale from time to time and even seen some dell coupons on sites like slickdeals.net and what not, in my price range. my question is, I was wondering if there was a specific place to keep on eye out for promo codes/sales for this monitor (or dell.com in general)? or should i just stick to monitoring slickdeals/fatwallet/etc for the next round of sales?

 

[needmorecarnitine]

almost every deal comes up on slickdeals.

Set up a deal alert for u2410 to pvt message you or email you when it comes up.

 

[Lollipop]

For weeks I have debated between the U2410 and the ZR24w with the Dell usually coming out on top in familiar advice and reviews. I ordered the U2410 today after shopping both companies many times to test both their CS and sales staff. I refuse to give my business to rude companies and HP acted both lazy and entitled. The monitor will be here on or before the 20th with free next day shipping. Got it for just shy of $500 drive out and without any coupons. That&#8217;s almost the same price as the HP after tax and shipping. &#8230;and for more features. And I 'm in TX too. Just find a nice rep and haggle "nicely". Just keep trying until you find a rep you can schmooze well with. Hopefully I&#8217;ll get a good one, but if at first you don&#8217;t succeed&#8230;

...kick 'em [H]arder!

 

[janos666]

I think I found the controller chipset which is used by this display: PDF Data Brief These features looks familiar, isn't they?

Does somebody know about a display which uses the same controller and the manufacturer let it to be hardware-calibrated?

 

[bustaplz]

A few weeks ago, my company bought 7 U2410 monitors and had them shipped to our main office with ATTN: (ME) as I was to set them up for various users. I stood in front of that stack of boxes labeled U2410 for 5-10 minutes before I decided that my job was probably worth more. Probably. Maybe?

 

[janos666]

I found an interesting line in the document which I linked above.

Deep color and wide gamut support: 12-bit HDMI input at YCC 4:4:4

Does it mean that U2410 is able to receive 12 bit/color signal through HDMI connection? It mentions YCC here and I am interested in the RGB format.
DP connection is working with 10 bit/color RGB (right now for me with HD5850) and I don't have a HDMI cable to test it with RGB format. Should I buy a HDMI cable? Can somebody check it with moninfo?

 

[hmz]

Ergo Tech makes a desk one that will work with bigger displays like this 24" Dell. They have one that will hold all 3 at one actually.

I'm using Ergotron LX arms for my 2410s...

Link to those arms?

 

[hmz]

Rev A02 already? I didn't know that. I guess I need an update.

 

[Wazooty]

Janos, did you order directly through Dell? I think I'm about to get one and a native RevA02 would be nice...

Also, if you look at the brand new Rev A01s being sold on ebay, some places offer a $20 pixel check. I asked them about it and mentioned the tinting issues some people are having and:

"Yes, we offer the $20 pixel/tint check as a premium service. We would check for any issues. If any issues were found we would set that one aside for us to RMA later and then start checking another brand new unit until a perfect one was found. "

Seems like a solid deal considering how often people can have problems.

I am rather iffy on if they'd actually check for tinting or if they're just saying it because they can but...better QC than none at all.

And they're still covered by dells warranty.

Hmm

 

[janos666]

Wazooty, I want to suggest you to change one of these options:
- Pay the tip (<35$) for a suitable seller but check it by yourself and test it very carefully (I suggest you to take some measures with a WCG capable instrument too, if it is possible).
- When the first option is not possible, then you should order it from Dell. You will have much better warranty. (It is hard to find the order number when you buy it elsewhere. I think it is not even legal. And it is also very hard to find a "perfect" specimen.)


I bought my first U2410 from a seller who offered the pixel and tinting check for a little tip. This was a new A01. We watched it and we found that there is no pixel error or hard tinting, but it had very low contrast ratio (~500:1) after the calibration (I figured it out at home).
I managed to get a replacement from the seller (because I paid a tip for an overseen specimen and I got a retarded one). My second U2410 (a new A01 from the same 2010 Marc series) had some stucked pixels. (We didn't have time to check it properly.)
My third one came from Dell because the seller didn't had any damage-free U2410 on stock (yes, we unpacked some boxes ). So, he gave me the order number and Dell replaced my new A01 with a refurbished A02 which was born as an A00.

But I think that this is the best combination. The old ones from 2009 have better contrast ratio, and this one has the latest firmware.

On the other side. I am still not satisfied, because Standard and Custom modes are useless, so I have to use the emulated sRGB mode all the time. And it won't change with a new firmware. (May be with some internal LUT hacking. But I gave up the hope for that.)
I could buy a real sRGB display instead of using a WCG display in sRGB emulation mode.
I also had to buy a new (and expensive) calibration instrument for this back-light (to calibrate the sRGB mode -> you can't say that it is not funny and pathetic at the same time...)


A question for everyone: What do you think? Will I be able to request an U2411 (replace this with the next 16:10 24" model) from Dell with a reasonable price compliance if I talk them about my problems with the Standard mode?
I think they assume that the strange gamma curve is a feature and not an error. So, my display in the Standard mode is perfect in their eyes, even if it is useless for everyone else.

 

[Wazooty]

Might I ask which modes are wide gamut and which aren't?

Obvuiously sRGB and adobeRGB aren't, but is game mode native wide gamut (which explains why the game mode sRGB trick works..?)

Also, I take it a device like the spyer 3 pro is best used in a wide gamut mode? (assuming custom...)

 

[janos666]

As I know (and I played a lot with OSD settings and calibration softwares):

sRGB and AdobeRGB presets has the corresponding gamut (emulated by hardware) and they have a reasonable native tonal response (around the ~2.1 pure-power curve).
I guess they use 3DLUTs but I am not sure about that, or about the bit-depth of these LUTs -> A 8 bit 3DLUT is 48Mb, and we need at least two of them (If I ware there, I would use 3DLUT for the Standard mode as well if the chipset already supports it...). But a good quality gamut conversion requires a lot of calculations. So, real-time processing can't be optimal and very unbelievable.

Custom mode is a mess. It does real-time image manipulation with a lot of user defined parameters. It is completely useless because the native tonal response is strange and it looks like the image is damaged by calculations even with the initial settings (without changing the Huye or Saturation controls from the default values).

Standard mode uses the native gamut (a really wide gamut which fully covers the sRGB or AdobeRGB gamuts) but it has a stupid native tonal response (around the ~1.75 pure-power curves). It is better than the Custom mode but I have heavy banding issues when I force it closer to the pure-power 2.2 tonal response curve.

Cool and Warm Presets are the same as the Standard mode but with offset white points (different color temperature presets in the service menu by 0-255 RGB Gains).

I am not sure about the other presets. I played a bit with Video modes too but they are useless. I can't remember the Multimedia Preset but I think it is an offset-Standard mode, or another useless crap.



In my conclusion, you have to use the emulated modes (sRGB or AdobeRGB) and you have to handle this display differently than you usually would.
The best way is to set the RGB Gains (in the service menu ; after taking a backup photo of them!!!) to 255-255-255 for 6500K, sRGB and AdobeRGB presets.
Now, you should calibrate the display with ArgyIICSM + DispcalGUI softwares in High quality mode. This software will need some exercise but current bundle and other popular softwares won't handle it well enough.
(Of course, you should set up a 10 bit connection and you will need a good calibration instrument too).

In this case (after some practice), you can achieve an acceptable gray-scale with good contrast ratio or a nearly perfect gray-scale with nearly acceptable contrast ratio (it will close to old S-IPS contrast ratios).


The emulated gamut is far from perfect (but instruments are not perfect either) but close enough. I think you can't achieve much better simulation with softwares. But native sRGB displays won't have perfect coverage either.
On my current U2410, the blue is very close, the red is a little over-saturated and green has a little bigger offset from the sRGB primaries.



I don't think you should even try to use a standard spyder 3 pro on this display. If you want a WCG-CCFL or LED back-light display, you need a special colorimeter (equipped with WCG compatible color filters) or you will need some display model specific correction matrices for your given colorimeter (stored on the instrument itself, or applied by the software).
I don't know the Spyders but you can ask X-rite to change the measure head on an i1d2 (and may be you can find some out-of-box WCG compatible EyeOnes somewhere) but I would suggest you to get a ColorMunki. I bought one some days ago and I think it is a great device (much better than an EyeOne or Spyder. So, it worth it's price.)



I think that U2410 is a really great hardware (I didn't make a mistake when I thought it after I have read the specifications.) But there is some serious problems with it's factory settings.
I don't know what caused these issues. May be they made a mistake or may be it was designed (they didn't want to make a nearly perfect display with this price tag.)

I would give my old colorimeter or my old S-PVA display to the person who can provide me with a software which can read and write the internal LUTs.
It should be possible and relatively easy to hack it. People hack a lot of devices with very hard security (like network routers, surfboard modems, game consoles, ect).
I think we could conjure a nearly perfect display with this kind of tools.

But I would be happy with a much easier solution: A preset mode which bypasses the internal LUT and ignores any image manipulation algorithms. May be we would be able to calibrate it from the scratch with 10 bit connection.



But take this post as an overstatement. It is a good display anyway, even with it's issues. Most of the displays have some kind of issues, even the much more expensive ones.
It would worth it's price if you can use your old colorimeter. But you can't, so the final price is almost doubled. (But you will need a new instrument for any kind of WCG, LED, ect. displays...)

Or you can live without a proper calibration. The factory state of the sRGB and AdobeRGB modes are not so bad. WP is very close to D65 and the tonal response is strange around the dark shades only (but you have the full 8 bit grayscale...).



Some subjective overall opinion:
After I finally received my colormunki and I figured out the optimal Dispcal settings, I started to test the result with my favorite pictures and Blu-Ray movies. Pictures are good, of course, it is an IPS and there is some additional fine-tuning with color managed softwares.
But Blu-Ray movies are excellent! I spent a hour to watch various scenes from my favorite movies because the perceived image is very good! The not-so-perfect near-black grayscale is much better than any S-PVA with it's black-crush effect.



Another thing: Can someone check the bit-depth of the HDMI connection?
My DP connection is working with 10 bit but I read that "12 bit YCC through HDMI" thing and may be it works with RGB too. I think it would be able to eliminate any banding in sRGB mode. (I don't have a HDMI cable to test it.)

 

[janos666]

Here is my last calibration result for the sRGB preset.
I achieved it with RGB Gains at 255-255-255 in the service menu, 10 bit DP connection, and "absolute" gamma scaling in Dispcal.
The calibration itself took ~23 minutes and the profiling took another ~15 minutes (but you can choose much more iteration points for the profiling ).

Current calibration response:
Black level = 0.14 cd/m^2
White level = 122.66 cd/m^2
Aprox. gamma = 2.20
Contrast ratio = 873:1
CIE chart from the emulated sRGB gamut

 

[Sailor_Moon]

In my conclusion, you have to use the emulated modes (sRGB or AdobeRGB) and you have to handle this display differently than you usually would.
The best way is to set the RGB Gains (in the service menu ; after taking a backup photo of them!!!) to 255-255-255 for 6500K, sRGB and AdobeRGB presets.

What were the XYZ tristimulus values of the white point?

(equipped with WCG compatible color filters)

If we have a colorimeter that (through its filters) reaches the curves of the CIE standard obsever, no correction for any sprectrum is needed. These devices exist but unfortunately not in the consumer sector (only in german, sry). The actual colorimeters aren't "72% NTSC CCFL compatible" out of the box but their internal correction is based on reference measurements against such a display. The higher the deviation of the filter solution is, the stronger must this correction be (and the deviations of measurements of other spectra than that of the reference display will be higher too). So we can't speak of "WCG compatible filters/ non WCG compatible" filters. There are just two classes: Very few solutions which reach the CIE standard observer (this probe should work without any correction regardless of the current spectrum, but price and software compabibility would be a problem) and more solutions which don't. The latter ones always need a form of correction for every spectrum.

I don't know the Spyders but you can ask X-rite to change the measure head on an i1d2 (and may be you can find some out-of-box WCG compatible EyeOnes somewhere) but I would suggest you to get a ColorMunki. I bought one some days ago and I think it is a great device (much better than an EyeOne or Spyder. So, it worth it's price.)

The internal correction can be altered. NEC should use this solution for "their" EOD2 and hopefully also preselect the probes because inter-instrument agreement is quite poor. HP does so for their LP2480zx calibration bundle* (an OEM EOD2) which delivers good results compared to an EyeOne Pro**. A good solution - aside from using a spectrophotometer (that has some limitations too but is suitable) - would be usage of Quatos Silver Haze Pro Bundle (DTP94 + iColor Display). The DTP94 has a good inter-instrument agreement and Quato has calculated generic corrections for WCG-CCFL spectra (92-96% NTSC). Works very satisfying for the U2410 (link).

I am not sure about the other presets. I played a bit with Video modes too but they are useless.

Behaviour should be similar to the U2711. There are some combinations that are usable but many are not.
http://www.prad.de/en/monitore/review/2010/review-dell-u2711-part19.html

Custom mode is a mess. It does real-time image manipulation with a lot of user defined parameters. It is completely useless because the native tonal response is strange and it looks like the image is damaged by calculations even with the initial settings

Yes, the custom mode is very nonlinear. That' s the cause for the high deviations in a profile validation because the software will "force normalize" the colorimetric values. A workaround would be usage of standard mode and setting RGB-gain via service menu. But you have already mentioned the "bad" gradation in this mode. The absence of a true gradation control is a pity. At least numeric gamma values should have been implemented.

Best regards

Denis

*
Corrections could be also carried out in software here - I can't check this because the probe is limited to the HP software

**
Of course no ideal reference because of constraints regarding sampling intervall and inter-instrument agreement, but it shows the direction