Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: geneo on March 10, 2019, 01:04:31 pm
-
I have tried everything under the sun and the results are still the same. delta-E of white point is fine.
-
I have tried everything under the sun and the results are still the same. delta-E of white point is fine.
Maybe some details about the calibration conditions, spectrometer type, etc.
Sent from my iPhone using Tapatalk
-
Why are you using dE(94) for this verification?
What profiling and profile verification package are you using?
If this were a dE(76) or dE(2000) number and the verification procedure were valid, to answer your question, no, the results are not very good.
-
Why are you using dE(94) for this verification?
What profiling and profile verification package are you using?
If this were a dE(76) or dE(2000) number and the verification procedure were valid, to answer your question, no, the results are not very good.
I get the same ranges under dE(94). I had the same question to ask but wanted to leave the OP detail his means of doing the calibration. Maybe he's into smth.
With dE(00) the results are just fine, at least with my PA271Q.
-
I think the OP should advise what results he gets with dE(00) or dE(76). If those numbers look good, he has no problem.
-
No, something is very wrong; here's my report just produced.
-
I have tried everything under the sun and the results are still the same. delta-E of white point is fine.
There is smth fishy about those figures. I see you're using SpectraView for calibration (actually for HW LUT that's the only option, afaik).
The dE(94) is what SpectraView is using for displaying "Color Tracking" report so that's fine. What I don't fine OK is that for me average is 0.47 and the d(max) is 1.68 - darks included.
Could you please detail what spectrometer you're using? And also... could you share the other tabs results in SpectraView, e.g. Color Gamut, Curves.
-
I think the OP should advise what results he gets with dE(00) or dE(76). If those numbers look good, he has no problem.
The formula is fixed in this product on the main report using Delta E CIE 94 but this one report might be using Delta H (ΔH). Will H from NEC might be able to comment if he see's this or I can ping him to take a look here.
-
I think the OP should advise what results he gets with dE(00) or dE(76). If those numbers look good, he has no problem.
Hi Mark,
I'm afraid there's no such dE(00)... Or I couldn't find it myself so far! :o
-
Hi Mark,
I'm afraid there's no such dE(00)... Or I couldn't find it myself so far! :o
Correct. At least according to the manual. Further, the numbers the OP is getting is off the charts no matter the formula used. Toggling between the two if possible would never account for the high values we are seeing.
-
I have tried everything under the sun and the results are still the same. delta-E of white point is fine.
No sure what is that "Photo Editing"... I hope though you're calibrating the PA271Q color gamut to "Native (full)". That's the only thing I could think it would trigger such large dE(94) values. Of course, aside from a defective monitor and/or spectrometer.
-
To answer some questions.
The is with the NEC i1 display pro 3. I also have an older i1 I have from an older spectraview monitor - same result. I have tried different graphics cards, cables, monitor settings, spectraview versions all with the same result. The green curve is also off from factory,
-
Correct. At least according to the manual. Further, the numbers the OP is getting is off the charts no matter the formula used. Toggling between the two if possible would never account for the high values we are seeing.
OK, as I don't use Spectraview I wouldn't have known that. Any idea why they use dE(94)?
-
How long have you had this issue?
I suspect you'll need to give NEC a call.
You might give MultiProfiler a try.
-
To answer some questions.
The is with the NEC i1 display pro 3. I also have an older i1 I have from an older spectraview monitor - same result. I have tried different graphics cards, cables, monitor settings, spectraview versions all with the same result. The green curve is also off from factory,
Not trying to tease you... but I would have expected PERFECT with NEC's own variant of MDSV Sensor 3. Weird, anyway! Still, are you sure you're calibrating to Native (full) gamut? Just asking...
-
How long have you had this issue?
I suspect you'll need to give NEC a call.
You might give MultiProfiler a try.
I have been in contact with NEC for this last week. They have kicked it up to engineering but they seem to be having some difficulty getting back to me.
I have had it less than 2 weeks.
-
Not trying to tease you... but I would have expected PERFECT with NEC's own variant of MDSV Sensor 3. Weird, anyway! Still, are you sure you're calibrating to Native (full) gamut? Just asking...
Yes, me too. I have been using an NEC LCD2490WUXI prior to this.
-
No sure what is that "Photo Editing"... I hope though you're calibrating the PA271Q color gamut to "Native (full)". That's the only thing I could think it would trigger such large dE(94) values. Of course, aside from a defective monitor and/or spectrometer.
Photo Editing is monitor full gamut. I have used a spectrometer (i1) that I know was good with the same result.
-
No, something is very wrong; here's my report just produced.
Thanks. Is that from a PA271q as well? I am getting some weird communication from NEC support about this monitor.
-
One thing not mentioned is monitor 'tilt' and how flat the sensor may be to the surface and the room ambient lighting.
Tilt the monitor back and put a black t-shirt over it to keep out stray light. It'd be interesting to know if that made any difference.
-
One thing not mentioned is monitor 'tilt' and how flat the sensor may be to the surface and the room ambient lighting.
Tilt the monitor back and put a black t-shirt over it to keep out stray light. It'd be interesting to know if that made any difference.
Thanks,
I do always tilt the monitor back (you really almost have to) and I do the calibration in a dark room. I used the same procedure on my older spectraview monitor with no issues.
-
For photo editing the contrast value is way to high.
-
Thanks,
I do always tilt the monitor back (you really almost have to) and I do the calibration in a dark room. I used the same procedure on my older spectraview monitor with no issues.
I agree with Andrew that this begins to look more and more like a product issue that you need to get to the bottom of with NEC Solutions Inc. Curious to know what exactly have they advised so far?
-
For photo editing the contrast value is way to high.
I have tried lower contrast as well with the same result.
-
I have tried lower contrast as well with the same result.
They have said the white point is good but there is something wrong with the gray scale, and kicked it up to engineering. That was last Tuesday. I hadn't heard back so called Friday. I was told the engineers were out of town to JP so couldn't communicate with them because of time difference. I asked about putting this on their DOA policy and shipping me a new monitor and was told that a replacement monitor may not resolve the issue with this series of monitor (???), but that they would have some plan of action by the end of Friday - probably shipping me a monitor. I never heard back on Friday, so plan to call them again on Monday.
-
For photo editing the contrast value is way to high.
It's too high for a print but that doesn't mean it's too high for other kinds of image viewing. And this has absolutely nothing to do with the dE errors of gray (or the amount of gamut).
Doesn't appear to be the instrument. Could be a software bug or an issue with the hardware. But no settings available for the user should produce such high gray deltas.
-
They have said the white point is good but there is something wrong with the gray scale, and kicked it up to engineering. That was last Tuesday. I hadn't heard back so called Friday. I was told the engineers were out of town to JP so couldn't communicate with them because of time difference. I asked about putting this on their DOA policy and shipping me a new monitor and was told that a replacement monitor may not resolve the issue with this series of monitor (???), but that they would have some plan of action by the end of Friday - probably shipping me a monitor. I never heard back on Friday, so plan to call them again on Monday.
OK, not bad - it's on their radar and they made a commitment to continue helping you. Maybe they think it could be a software issue in which case a bit premature to exchange the hardware. I'm sure you know - stay on top of them till you get satisfaction, now that you know these results are unacceptable.
-
L2 was intimating maybe it was something wrong with the series, hence my post. I am hoping that was just some miscommunication with NEC engineering
BTW, looked back at the lower contrast calibration I did: 250:1 same result.
-
I wanted to point out that I recently purchased two new PA271Q monitors and had the same issue that the OP describes. The greyscale Delta E's were huge whereas the white point values were excellent. I repeated the calibration process numerous times on both monitors and at different white points, luminance levels and contrast ratios, and could never get the average greyscale Delta E to be under 2 using SpectraView II. In all cases I used a Gamma Curve value of 2.2 and the Color Gamut was set to Native (Full). The attached file shows the best greyscale result I could achieve after numerous calibration attempts. The only way I could get a reasonable calibration was when I used DisplayCal. I ended up returning both of the PA271Q monitors. To me this seems to be an issue with the PA271Q and/or the SpectraView II software. It should be noted that I was using the latest version of SpectraView II (Version 1.1.39 Build 181031). I also calibrated with the previous version of 1.1.38 and got the same results.
-
I should also mention that I used the MDSVSENSOR3 that came with the PA271Q for all my calibration attempts using the SpectraView II software.
-
It's a weird software bug, that's occurs randomly in some PA271Qs for no obvious reason. It will be fixed soon.
-
It's a weird software bug, that's occurs randomly in some PA271Qs for no obvious reason. It will be fixed soon.
Ah, thanks. That is good to hear I reckon. That makes sense from what L2 was hinting at. They are cross shipping me a new monitor in any case, I should be getting it tomorrow. If it exhibits this issue I will just sit on it then.
-
Hi Marcin,
Were you able to talk directly with NEC regarding the software bug? The reason I ask is I have not seen anything from NEC regarding this issue. Thanks!
-
My report shown earlier is from a PA271Q, Mac Mojave, custom target, no problems FWIW.
-
Hi Marcin,
Were you able to talk directly with NEC regarding the software bug? The reason I ask is I have not seen anything from NEC regarding this issue. Thanks!
All I can say is that I managed to get rid of the issue by myself, so it seems not to really be a big problem - and AFAIK it will be addressed by NEC soon.
-
All I can say is that I managed to get rid of the issue by myself, so it seems not to really be a big problem - and AFAIK it will be addressed by NEC soon.
Why not share with the others / OP the solution for the benefit of all, if there is a solution?!
-
Why not share with the others / OP the solution for the benefit of all, if there is a solution?!
Problem is that I'm not sure how I fixed it...
-
Problem is that I'm not sure how I fixed it...
We've all been there, Marcin!
-
We've all been there, Marcin!
LOL. Well I received the replacement. It is a bit better but not excellent or even great. This is at 500:1 contrast, full color gamut.
@Czornyj Hope you are right. Everything else about the monitor is fine.
-
LOL. Well I received the replacement. It is a bit better but not excellent or even great. This is at 500:1 contrast, full color gamut.
@Czornyj Hope you are right. Everything else about the monitor is fine.
This is still not ok not at all ok, I would dare to say, for this type of professional monitor. Can you share the summary page as well? One question though: why 500:1 and lot leave it to the native contrast? Just curious...
Else, I have never tried calibrating it via DiplayCal but that would be an idea to test and see if smth is not wrong with the monitor internals (3D LUT) since DisplayCal can only calibrate it via graphic card LUT adjustments.
When you received your replacement have you received also a new spectrometer?! I still suspect there's an issue with the NEC's version colorimeter!
I don't know what colorimeter DigitalDog is using but I am using a EFI ES-2000 spectrophotometer. Would it be possible to rent another spectrometer to try again?
-
We've all been there, Marcin!
Not all of Us.
-
LOL. Well I received the replacement. It is a bit better but not excellent or even great. This is at 500:1 contrast, full color gamut.
@Czornyj Hope you are right. Everything else about the monitor is fine.
He is and to remind others of what he wrote:It's a weird software bug'
-
When you received your replacement have you received also a new spectrometer?! I still suspect there's an issue with the NEC's version colorimeter!
I don't know what colorimeter DigitalDog is using but I am using a EFI ES-2000 spectrophotometer. Would it be possible to rent another spectrometer to try again?
It’s not sensor releated - I have a couple of i1Display Pros and i1Pro2 sensors, the result was the same with all of them. I’m 1000% sure it’s a software bug - there’s strange anomaly in green TRC shape after calibration that causes the above mentioned behavior. The good news is that it’s „curable”, so now we only have to wait for new firmware/SVII revision.
-
It’s not sensor releated - I have a couple of i1Display Pros and i1Pro2 sensors, the result was the same with all of them. I’m 1000% sure it’s a software bug - there’s strange anomaly in green TRC shape after calibration that causes the above mentioned behavior. The good news is that it’s „curable”, so now we only have to wait for new firmware/SVII revision.
What versions of SpectraView are you and OP using? I had to do a Win10 install and updated SpectraView to version 1.1.18 (I see that 1.1.19 is now available). I don't see any problems when I calibrate my monitor (P232W) with an NEC branded colorimeter. Is this also linked to which NEC monitor one is using? Pretty strange behavior.
-
What versions of SpectraView are you and OP using? I had to do a Win10 install and updated SpectraView to version 1.1.18 (I see that 1.1.19 is now available). I don't see any problems when I calibrate my monitor (P232W) with an NEC branded colorimeter. Is this also linked to which NEC monitor one is using? Pretty strange behavior.
You meant 1.1.38 and 1.1.39?
I am on 1.1.39. I tried 1.1.38 too. Also tried older i1 sensor with same results. Also see an anomaly in green - a larger correction.
So I will hold onto this one. and contact L2 to see what the engineers have to say.
-
What versions of SpectraView are you and OP using? I had to do a Win10 install and updated SpectraView to version 1.1.18 (I see that 1.1.19 is now available). I don't see any problems when I calibrate my monitor (P232W) with an NEC branded colorimeter. Is this also linked to which NEC monitor one is using? Pretty strange behavior.
It doesn’t matter - I tested various versions on various platforms with various sensors with the very same results.
-
Not all of Us.
You just couldn't help yourself, could you? Again.
-
You just couldn't help yourself, could you? Again.
Correcting you again, indeed. :o
At least my posts previous to yours here were based on actual usage and experience with the product under discussion.
Or you finally got a PA271Q despite all the flickering (https://forum.luminous-landscape.com/index.php?topic=82256.msg664461#msg664461) issues you find with NECs (among other so called 'issues')?
-
I have tried everything under the sun and the results are still the same. delta-E of white point is fine.
I just calibrated my unit again, still results as expected (shown below).
Let's try something different. Try downloading my custom settings (target file) found here:
http://digitaldog.net/files/ARsPhotoEditing.zip
Unzip,Load in SpectraView and calibrate. How is the report?
-
I just calibrated my unit again, still results as expected (shown below).
Let's try something different. Try downloading my custom settings (target file) found here:
http://digitaldog.net/files/ARsPhotoEditing.zip
Unzip,Load in SpectraView and calibrate. How is the report?
It won't help, the calibration target is irrelevant, you can choose different wtpts, brightnesses, TRCs, CRs - there’s always a weird offset in gTRC that causes validation errors.
https://www.dropbox.com/s/qu3pc6yeafldyks/Zrzut%20ekranu%202019-03-17%2021.51.55.png?dl=0
-
I just recalibrated my monitor last week and get pretty much the same results that Andrew did. I did make an error in the software number as was pointed out by a couple of users. I'm on ver 1.1.38
-
It won't help, the calibration target is irrelevant, you can choose different wtpts, brightnesses, TRCs, CRs - there’s always a weird offset in gTRC that causes validation errors.
https://www.dropbox.com/s/qu3pc6yeafldyks/Zrzut%20ekranu%202019-03-17%2021.51.55.png?dl=0 (https://www.dropbox.com/s/qu3pc6yeafldyks/Zrzut%20ekranu%202019-03-17%2021.51.55.png?dl=0)
I was wondering if a combo of settings would make it work, as it does on this end with the latest version of the Mac software 1.1.39.
Something changed to make your copy work right?
-
I just calibrated my unit again, still results as expected (shown below).
Let's try something different. Try downloading my custom settings (target file) found here:
http://digitaldog.net/files/ARsPhotoEditing.zip
Unzip,Load in SpectraView and calibrate. How is the report?
That is what I would expect to see. I calibrated with your target and here is what I get. So now we have a direct comparison to affected and non-affected monitors.
I wonder if it was some part supply change. What vintage is your monitor?
Also attached my Green curve. It is not as off as the first one, which may explain why this is a little better.
-
I got mine last October.
-
I was wondering if a combo of settings would make it work, as it does on this end with the latest version of the Mac software 1.1.39.
Something changed to make your copy work right?
I tried everything, and suddenly something helped. BTW - can you put an USB stick to first upper USB port on the left side, export the settings of your monitor and upload it here (the .nsvs file)?
-
The Set button doesn't seem to be working when selected. Press set to export monitor settings I see, but set doesn't do anything. I've tried all the USB ports. :(
-
The Set button doesn't seem to be working when selected. Press set to export monitor settings I see, but set doesn't do anything. I've tried all the USB ports. :(
There's one single USB port above them all (not one of 3 USB3 blue ports on the bottom) - when you put an USB stick it displays export/load settings automatically.
-
Here's mine:
https://www.dropbox.com/s/2ev4z5oo3ictlpi/pa271q_86100058TB_a_01.nsvs?dl=0
-
There's one single USB port above them all (not one of 3 USB3 blue ports on the bottom) - when you put an USB stick it displays export/load settings automatically.
Yeah, found it after pulling the cap off. The file is zipped:
http://digitaldog.net/files/pa271q_87100388TA_a.nsvs.zip
-
Here's mine:
https://www.dropbox.com/s/2ev4z5oo3ictlpi/pa271q_86100058TB_a_01.nsvs?dl=0
Wish they were human readable.
Do you have yours self calibrated or using factory calibration?
What version of the firmware?
-
Wish they were human readable.
Do you have yours self calibrated or using factory calibration?
What version of the firmware?
- copy it on USB stick
- put the stick to the USB port above all ports on the left side
- load settings to the display
- calibrate the display again
- if it won't help reset the display using Multiprofiler, recalibrate
- and make factory reset from menu, recalibrate
- and update to latest FW v1.2, recealibrate
- and update SVII to 1.1.39
- and calibrate it again
At the end of the day something fixes it, I'm not sure what and how.
Here's an affected display I took off a while ago:
https://www.dropbox.com/s/3jtho7ij6m1rwci/Zrzut%20ekranu%202019-03-18%2000.56.45.png?dl=0
Same display after applying the above mentioned shamanic rituals (still have no idea what really helped):
https://www.dropbox.com/s/rvq0eu1c8lfb1wm/Zrzut%20ekranu%202019-03-18%2000.56.54.png?dl=0
-
Seems like the steps I made on the first monitor (updated to firmware 1.2 too). And am on SVII .39 as well. But wwill give it a go later.
-
Seems like the steps I made on the first monitor (updated to firmware 1.2 too). And am on SVII .39 as well. But wwill give it a go later.
Try to load our .nsvs files, and resetting it. It doesn't work all the time and it does't make sense, but eventually it may fix it.
-
Correcting you again, indeed. :o
At least my posts previous to yours here were based on actual usage and experience with the product under discussion.
Or you finally got a PA271Q despite all the flickering (https://forum.luminous-landscape.com/index.php?topic=82256.msg664461#msg664461) issues you find with NECs (among other so called 'issues')?
Wrong again. I never found flicker issued with NECs. I asked for other people's possible experience with flicker, be it NEC or otherwise, before buying an LED/LCD monitor, remember? But of course you don't or pretend not to. Facts never stood in your way.
-
I just calibrated my unit again, still results as expected (shown below).
Let's try something different. Try downloading my custom settings (target file) found here:
http://digitaldog.net/files/ARsPhotoEditing.zip
Unzip,Load in SpectraView and calibrate. How is the report?
Dear Andrew,
Here comes trouble and I am puzzled! :O After updating my NEC PA271Q to the latest available 1.202 firmware I my calibration has gone bananas!
While in the past most of my calibration result where on the 0.4 to 0.6 dE(94) margins for "Average" and 1.6 to 2.4 dE(94) for "Maximum" with the biggest delta in "Dark Values" (else the Avg was 0.23 and Max 0.45) now my dE(94) values have rocketed to 1.13 to 1.28 for Avg and 1.95 to 2.10 Max dE(94).
I was expecting the firmware update to correct the deviation caused by the "Include Dark Values" but instead the whole range of values (L* 10+) have gone awry!
I even downloaded your "ARsPhotoEditing.tgt" file for calibration and still my values are out of their normal ranges. Now the "dark values" are not an issue anymore but the "light values" have become an issue!
I am not sure if I should be happy now (with more dE(94) in all spectrum vs larger dE(94) in the darks)!?!! I'm confused!
I have attached comparisons among few past profiles and the newest one after the firmware update.
What's wrong? What to do?
-
Wrong again. I never found flicker issued with NECs. I asked for other people's possible experience with flicker, be it NEC or otherwise, before buying an LED/LCD monitor, remember? But of course you don't or pretend not to. Facts never stood in your way.
I remember nearly all your nonsense posts including (your text from a URL already provided):
Given the potential for problems and the relative easy fix for the P232W I still don't understand why NEC doesn't plan to upgrade that monitor with a much higher backlighting PWM.
This after people told you they DIDN'T (http://www.photo.net/digital-darkroom-forum/00c06Z) flicker! Another example of providing facts you can't accept! ;)
Indeed, you still don't understand.
You own a PA (specifically a 271Q), how's the flickering affecting you? You don't? Maybe that's why there are no facts or a lick of text from your keyboard about solving the OP's issues in this thread. :o
-
I was expecting the firmware update to correct the deviation caused by the "Include Dark Values" but instead the whole range of values (L* 10+) have gone awry!
According to the OSD, I'm running R1.001A. Never updated the firmware nor had any indication to do so.
I am not sure if I should be happy now (with more dE(94) in all spectrum vs larger dE(94) in the darks)!?!! I'm confused!
At this point I wouldn't worry too much as NEC is aware of this and unless images look poor, I'd move on for the time being until we get more information.
I was calibrating and saw the screen saver kick in (stopped it after a second) and saw in that report with one really bad spike but everything looked fine otherwise. I did recalibrate after setting the screen saver off (don't need it with this unit) and that one high dE spike it disappeared. IF the overall report in the main reporting screen is sound, I think you're good to continue until NEC figures out this software issue.
What is the firmware you're running?
-
What is the firmware you're running?
I am now running v1.202... available on NEC's site. Anyway I will raise a ticket to NEC to have the problem on their radar. To me this is not ok at all. Not to mention that am not impressed at all about the quite substantial differences between two consecutive calibration.
Dear Andrew, I still have two questions for you. One might sound dumb but... here they are:
1) Why would one go for a fixed contrast ratio (like you're using 300:1) instead of going to the monitor native highest? Not to mention that almost every time I would choose "Monitor native" I would get results varying from 800:1 to 1400:1... really weird.
2) How could I test color accuracy of the SpectraView generated profile, other that by using the existing "Validate Calibration", which I found to be pretty much useless? i1Profiler keeps telling me "No active profile" when trying to invoke the QA.
Thanks a lot.
-
I am now running v1.202... available on NEC's site. Anyway I will raise a ticket to NEC to have the problem on their radar. To me this is not ok at all. Not to mention that am not impressed at all about the quite substantial differences between two consecutive calibration.
Dear Andrew, I still have two questions for you. One might sound dumb but... here they are:
1) Why would one go for a fixed contrast ratio (like you're using 300:1) instead of going to the monitor native highest? Not to mention that almost every time I would choose "Monitor native" I would get results varying from 800:1 to 1400:1... really weird.
2) How could I test color accuracy of the SpectraView generated profile, other that by using the existing "Validate Calibration", which I found to be pretty much useless? i1Profiler keeps telling me "No active profile" when trying to invoke the QA.
Thanks a lot.
1. To better match the contrast ratio of a print when calibrating the display for a specific print to display match. See:
http://blog.xritephoto.com/2011/07/x-rite-i1display-pro-advanced-features-contrast-ratio-with-coloratti-andrew-rodney/
SpectraView does this of course and better, allows you to build as many targets for specific uses as you desire.
2. The validation IMHO is only useful to determine when the process doesn't work correctly and you want numbers to back it up. It isn't useful in gauging overall accuracy for a number of reasons. Primarily the issue is you're using the same software and more importantly the same instrument, not a higher grade reference device, to gauge accuracy.Lets say I measure my foot with a home depot wooden ruler and find it's 10.7 inches. Clearly my foot isn't an adequate measuring device compared to that 1 foot wooden ruler. But how accurate is that ruler? It too should be of sufficient accuracy for the task. There are devices like a 50 cent wooden ruler that may be accurate to +/- 1/10 of an inch. For measuring a 100 foot fence, that may be all that's needed in terms of measurement accuracy. For measuring a component that will be used on a nuclear power plant 1/10000 of an inch may just be adequate enough for the specifications of the job. But my 10.7 foot clearly isn't accurate. So when we talk accuracy, we need to take into account the instrument or method of measuring the accuracy. And how accurate we need the results of the measurement to the reference. With dE, 1 or less is invisible. We are using the same instrument to evaluate the display accuracy and that's not worthless but not ideal either.
What is useful is trending where we use the same instrument and number of color patches over time, multiple times. Now we can see consistently in the calibration. And we can get a good idea how often we need to calibrate the display due to the dE of drift over time.
-
1. To better match the contrast ratio of a print when calibrating the display for a specific print to display match. See:
http://blog.xritephoto.com/2011/07/x-rite-i1display-pro-advanced-features-contrast-ratio-with-coloratti-andrew-rodney/
SpectraView does this of course and better, allows you to build as many targets for specific uses as you desire.
Wow, that was quick... and very informative! ;) Thanks for the tip.
2. The validation IMHO is only useful to determine when the process doesn't work correctly and you want numbers to back it up. It isn't useful in gauging overall accuracy for a number of reasons. Primarily the issue is you're using the same software and more importantly the same instrument, not a higher grade reference device, to gauge accuracy.Lets say I measure my foot with a home depot wooden ruler and find it's 10.7 inches. Clearly my foot isn't an adequate measuring device compared to that 1 foot wooden ruler. But how accurate is that ruler? It too should be of sufficient accuracy for the task. There are devices like a 50 cent wooden ruler that may be accurate to +/- 1/10 of an inch. For measuring a 100 foot fence, that may be all that's needed in terms of measurement accuracy. For measuring a component that will be used on a nuclear power plant 1/10000 of an inch may just be adequate enough for the specifications of the job. But my 10.7 foot clearly isn't accurate. So when we talk accuracy, we need to take into account the instrument or method of measuring the accuracy. And how accurate we need the results of the measurement to the reference. With dE, 1 or less is invisible. We are using the same instrument to evaluate the display accuracy and that's not worthless but not ideal either.
What is useful is trending where we use the same instrument and number of color patches over time, multiple times. Now we can see consistently in the calibration. And we can get a good idea how often we need to calibrate the display due to the dE of drift over time.
I might be stubborn in my belief that once the NEC calibration results can be so "diverse" there should also be another "impartial" way of checking the actual results. Of course, visually is difficult to assess their quality hence my worries.
Not quite comfortable as of now with the results and the shifts... :(
-
Wow, that was quick... and very informative! ;) Thanks for the tip.
I might be stubborn in my belief that once the NEC calibration results can be so "diverse" there should also be another "impartial" way of checking the actual results.
Well if the goal is calibration to match a print soft proofing, if you see a good match, you're set! ;)
-
Apparently there is no fix at this time so I am to return to retailer :(
Any suggestions for a monitor?
-
Similar Problem. Not really happy with my new NEC PA271q... First NEC after a few Eizos and never had such a bad calibration. All my displays before that were much closer to Rodney's NEC delta.
I tried a few things, but couldn't improve much.
First Image shows, calibration before Firmware and SpectraView update.
Second shows calibration to Rodney's target. Much better, but still not close to his NEC.
Last one is my target again, also better, but still not great.
Perhaps I'm expecting to much, but the PA271q got so great reviews, so I'm surprised about it's perfromance.
-
I was on the cusp of getting a PA271q over an Eizo monitor, and it looks like I should either not do it or hold out to see if there is a solution. I'd love to save some money, but both monitors are in a price range where I want predictable results without fiddling. If this is not a glitch, I guess I will be paying some more for an Eizo.
I'll keep monitoring this thread for a while to see if there is a systematic fix from NEC, but if not I guess that is it.
-
I was on the cusp of getting a PA271q over an Eizo monitor, and it looks like I should either not do it or hold out to see if there is a solution. I'd love to save some money, but both monitors are in a price range where I want predictable results without fiddling. If this is not a glitch, I guess I will be paying some more for an Eizo.
I'll keep monitoring this thread for a while to see if there is a systematic fix from NEC, but if not I guess that is it.
I was told there would be no resolution in the immediate future, and that it was a rare situation (which makes no sense). Hope that is wrong.
I am looking at Eizo too. My two issues there are that I have seen some reports of not so stellar support and their monitors of the same size with hardware LUT and sensor are in the $2500 range. I don't want to pay that much and not have good support. NEC support seems good.
-
I was told there would be no resolution in the immediate future, and that it was a rare situation (which makes no sense). Hope that is wrong.
I am looking at Eizo too. My two issues there are that I have seen some reports of not so stellar support and their monitors of the same size with hardware LUT and sensor are in the $2500 range. I don't want to pay that much and not have good support. NEC support seems good.
Not sure how rare it is, happily mine is fine and I suspect if you got one that isn't correct, NEC would replace under warranty.
Next, there is the possibility the Eizo using the same panel would suffer the same issues....
-
Not sure how rare it is, happily mine is fine and I suspect if you got one that isn't correct, NEC would replace under warranty.
Next, there is the possibility the Eizo using the same panel would suffer the same issues....
As I stated in this thread, they did replace the monitor under warranty and it was marginally better, so I got a full refund since there appears to be no fix on the horizon.
As to my comment about rare, I got two in a row, and they should have two to investigate.
-
As I stated in this thread, they did replace the monitor under warranty and it was marginally better, so I got a full refund since there appears to be no fix on the horizon.
As to my comment about rare, I got two in a row, and they should have two to investigate.
I'd push them to keep sending them until you get one that behaves like mine. Maybe mine is the rare one but all I can tell you is what I see in my report which is what is expected.
-
I'd push them to keep sending them until you get one that behaves like mine. Maybe mine is the rare one but all I can tell you is what I see in my report which is what is expected.
Honestly, it is not worth that kind of hassle for me - I value my time.
-
Honestly, it is not worth that kind of hassle for me - I value my time.
So you end up with what?
-
So you end up with what?
I see you are the kind of person who has to tell other people how they should think and act and who has to have the last word. Well what I have is my current NEC and I will probably have an Eizo as a replacement in the near future and I won't have wasted any more days of my time
And FWIW, NEC suggested the second replacement monitor would probably not resolve the issue, and it did not, so there is more here going on than you know. I also expect you have an earlier one or was very lucky.
-
I see you are the kind of person who has to tell other people how they should think and act and who has to have the last word.
Wrong. Do whatever you want. Point is, facts are, there are PA271q's that run fine. You can either decide to get one, as owed to you or go elsewhere. I could care less. Get an Eizo, don't be shocked if you find the same issues.
NEC suggested the second replacement monitor would probably not resolve the issue, and it did not, so there is more here going on than you know.
Right, I'm the only PA271Q owner who's display as shown operates (https://forum.luminous-landscape.com/index.php?topic=129474.msg1099343#msg1099343) as it should. :P
-
I see you are the kind of person who has to tell other people how they should think and act and who has to have the last word. Well what I have is my current NEC and I will probably have an Eizo as a replacement in the near future and I won't have wasted any more days of my time
And FWIW, NEC suggested the second replacement monitor would probably not resolve the issue, and it did not, so there is more here going on than you know. I also expect you have an earlier one or was very lucky.
You hit the nail on the head!
-
You hit the nail on the head!
If only you used yours.... :P
As you once did earlier:
A topic you are watching has been removed by Frans Waterlander.
Regards,
The Luminous Landscape Forum Team.
-
who has to have the last word.
-
who has to have the last word.
Who had to remove all his silly posts due to embarrassment?
EDIT: I'm not one to strip away mistakes like you, I just fix em.
-
No embracement here. I removed those threads because I got sick and tired of the hostile replies by certain people, and you know who you are.
-
No embracement here. I removed those threads because I got sick and tired of the hostile replies by certain people, and you know who you are.
Good move on your part as so few where taking you seriously, including me of course. Hostile is your code for being corrected or being pushed back for never accepting an answer you didn't predetermine you wanted to hear that were wrong. But then there's this history here and over on PhotoNet where posts you didn't create and can't delete still exist to illustrate this posting agenda.
-
Blah, blah, blah...
-
I suppose this exchange couldn't have been a private one for the sake of keeping this thread focused on the real issue and clean of testosterone, could it?
-
I suppose this exchange couldn't have been a private one for the sake of keeping this thread focused on the real issue and clean of testosterone, could it?
Indeed, Franz is again just hijacking a thread about a product he's got zero experience with. Let's do keep on topic and not feed trolls.
NEC is aware of the issues. I've spoken with the US product manager and they are working on why some panels exhibit this behavior but again, not all do. So anyone who sees the same issues should contact NEC and get the unit replaced. Now if someone does this more than once and doesn't wish to continue until they receive satisfaction, they have to do what they feel is necessary above and beyond accepting units until they work as they should.
-
Indeed, Franz is again just hijacking a thread about a product he's got zero experience with. Let's do keep on topic and not feed trolls.
NEC is aware of the issues. I've spoken with the US product manager and they are working on why some panels exhibit this behavior but again, not all do. So anyone who sees the same issues should contact NEC and get the unit replaced. Now if someone does this more than once and doesn't wish to continue until they receive satisfaction, they have to do what they feel is necessary above and beyond accepting units until they work as they should.
Man what a boor. You have already confirmed my initial impression, so I am sure you will have to reply to this, but you are on my blocked user list now, so I will not see it.
-
Man what a boor. You have already confirmed my initial impression, so I am sure you will have to reply to this, but you are on my blocked user list now, so I will not see it.
But I answered your question (https://forum.luminous-landscape.com/index.php?topic=129474.msg1099343#msg1099343) awhile ago:
Would you find this delta-E acceptable on a new PA271q? I have tried everything under the sun and the results are still the same. delta-E of white point is fine.
You didn't find that boring but sure, block: ignorance is said to be bliss: Enjoy.
-
So here is a question: Is this problem that a screen must manifest straight out of the box, or is it the case that the screen may be fine initially but develop this problem over time?
-
So here is a question: Is this problem that a screen must manifest straight out of the box, or is it the case that the screen may be fine initially but develop this problem over time?
From my experience and what I could glean from support, Straight out of the box or you are good.
-
So here is a question: Is this problem that a screen must manifest straight out of the box, or is it the case that the screen may be fine initially but develop this problem over time?
Straight away, it's an issue with the panel.
-
Rodney in your honest opinion, what do you think of mine? Would you accept it as "good enough" or not?
-
Rodney in your honest opinion, what do you think of mine? Would you accept it as "good enough" or not?
To be honest, I don't think that report is all that critical.
-
wow, so nothing really happened. Next time I'm going with Eizo again.
-
wow, so nothing really happened. Next time I'm going with Eizo again.
Don't be so certain Eizo isn't affected. Might not see anything like the reports shown that illustrate the issue in their software but that doesn't mean it's not happening.
-
wow, so nothing really happened. Next time I'm going with Eizo again.
Ended up getting an Eizo cg2730. No issues.
-
Latest FW update solves the issue:
https://www.nec-display.com/dl/en/dp_soft/pa_fm_update/PA3/download.html
-
That's good news for owners. Wish they could have told me back when.
-
Since I've never had this issue, I have to wonder if I should bother with the new firmware update???
-
This is the update: "The picture uniformity and low luminance stability are improved".
Does this fix the large Delta E problem? Has any one actually applied this update and seen an improvement in their calibration results?
-
This is the update: "The picture uniformity and low luminance stability are improved".
Does this fix the large Delta E problem? Has any one actually applied this update and seen an improvement in their calibration results?
Yes it does - at least in case of 5 units that I tested.
-
Since I've never had this issue, I have to wonder if I should bother with the new firmware update???
I wouldn’t bother until it happens ;)
-
I've tried 3 different USB drives. All of them are recognized but both "Update Firmware" and "Import monitor settings" are ghosted.
PA271Q_R1302.pac is on the USB with nothing else on the drive. Unplugged everything else that was attached, including display feed.
- Load PA271Q_R1302.pac onto usb drive
- Turn off power to PA271Q
- Plug in usb drive to the top SENS/MEM port #1
- Turn on monitor
- Menu automatically opens to USB Flash Drive / Function Selection, but Update Firmware is ghosted. Bottom of menu says, "USB Drive: Connected
- ??
Any suggestions? It is recognizing the usb drive but not the firmware file. I am on a Mac. Do I have to do this from a pc? Should I try reset? Current firmware is R1.001A, Spectraview Engine 5.00
Dave
-
Czornyj should/will chime in; I had a similar issue in the past with the USB stick when he requested I copy some data for him. Seems the trick is inserting it into one of the ports in the display itself. If memory serves, it's on the right side, at the top, behind a rubber goodie protector???
-
Czornyj should/will chime in; I had a similar issue in the past with the USB stick when he requested I copy some data for him. Seems the trick is inserting it into one of the ports in the display itself. If memory serves, it's on the right side, at the top, behind a rubber goodie protector???
Thanks Andrew, I think that is the one I'm using. It does recognize the USB drive; just not the firmware file that I put on there.
I've been turning off the monitor from the front. Going to try the little on/off switch on the left side in the back. If that doesn't work I will reformat a USB drive and try again.
Dave
-
Thanks Andrew, I think that is the one I'm using. It does recognize the USB drive; just not the firmware file that I put on there.
I've been turning off the monitor from the front. Going to try the little on/off switch on the left side in the back. If that doesn't work I will reformat a USB drive and try again.
Dave
If you're on Mac, I'd check if USB is FAT32 formatted.
-
If you're on Mac, I'd check if USB is FAT32 formatted.
Hi Marcin,
Yes, checked:
-
Retried a few other usb drives. Tried downloading the firmware again. No luck.
It just won't recognize the file, and I didn't change the name at all. Hard to tell from this image, but "Import monitor settings" and "Update firmware" are both ghosted out.
-
Tried exporting monitor settings and that worked, and the menu "import monitor settings" is no longer ghosted. So it has no problem reading files on the usb drive. There must be something wrong with the firmware file I have on there.
-
Here is a link to my monitor settings and the firmware file I've been trying to get the monitor to recognize:
https://www.dropbox.com/sh/icxhdj9gnfwaidh/AABfdx-g1ZudnHrIGVZLlnFxa?dl=0
Thank you for the responses. I will call NEC tomorrow when they open. My monitor dE's are borderline. Not as bad as some, but not near Andrew's. Screenshot below using Andrew's target settings.
Dave
-
Good news and bad news. After speaking with NEC, they ran some tests and determined the file needs to be extracted and copied onto the USB drive from a windows machine:
Testing here confirmed the pac file needs to be installed on the USB drive from a Windows PC. For whatever reason, and albeit FAT32 is universal there is something occurring to the file when moved/copied to the USB drive on a MAC.
I can't explain that, but after booting up an old pc, downloading the file and dumping it onto a USB drive, the monitor recognized the file. I successfully updated the firmware.
The bad news is results are no better.
Dave
-
Good news and bad news. After speaking with NEC, they ran some tests and determined the file needs to be extracted and copied onto the USB drive from a windows machine:
I can't explain that, but after booting up an old pc, downloading the file and dumping it onto a USB drive, the monitor recognized the file. I successfully updated the firmware.
The bad news is results are no better.
Dave
That's unfortunate Dave. At least it is not terribly off.
-
I really wouldn’t worry about this.
The manual states:
Values under 3 are generally considered excellent for display calibration. Values above 3 will be highlighted in yellow or red and may indicate a problem with the display’s calibration.
The graph shows the displays actual measured grayscale color tracking using measurements taken while generating an ICC/ColorSync monitor profile of the display.
The horizontal axis shows the Delta E values for each measured point on the grayscale. The vertical axis shows the luminance for each of the measured points on the grayscale.
Now this note might be telling:
Note: Since the Delta E values are calculated based on the currently selected Target, if a Target using Intensity: Specific Level is being used, and the display was not able to actually achieve the Target’s specified Intensity value, then this will result in large Delta E values.
So, what's the specific level being asked for?
-
Now this note might be telling:
Note: Since the Delta E values are calculated based on the currently selected Target, if a Target using Intensity: Specific Level is being used, and the display was not able to actually achieve the Target’s specified Intensity value, then this will result in large Delta E values.
So, what's the specific level being asked for?
That's interesting, Andrew. NEC has been quite helpful in suggesting things to try, and the most recent was to lower my intensity to 110-120cd/m2 in order to see if that changes the dE values. They also reiterated that my readings are not too bad. I am going to continue trying their suggestions. If nothing else, the firmware update issue was solved.
Dave
-
130cd/m2, 300:1 contrast ratio (target black = 0.43):
-
I run mine at 150cd/m2 and 350:1 so our targets are closer.
-
I also tried 140cd/m2 and 350:1. Those results are almost as good; avg 0.61 max 1.00.
Thanks again for your help, Andrew.
Dave
-
I run mine at 150cd/m2 and 350:1 so our targets are closer.
Andrew, this seems to me a very bright screen for editing photographs. Might you explain your thinking on choosing this approach? Are you working in very brightly lit rooms?
-
Andrew, this seems to me a very bright screen for editing photographs. Might you explain your thinking on choosing this approach? Are you working in very brightly lit rooms?
Simple: matches my viewing booth.
http://digitaldog.net/files/Print_to_Screen_Matching.jpg
(http://digitaldog.net/files/Print_to_Screen_Matching.jpg)
-
Simple: matches my viewing booth.
http://digitaldog.net/files/Print_to_Screen_Matching.jpg
(http://digitaldog.net/files/Print_to_Screen_Matching.jpg)
Andrew, what's the purpose of the task light in addition to the viewing booth? You probably don't use both at the same time, right?
-
I also tried 140cd/m2 and 350:1. Those results are almost as good; avg 0.61 max 1.00.
Thanks again for your help, Andrew.
Dave
I wouldn't alter you calibration targets just for a better report. If you can alter them and Print viewing conditions and are still happy with the results, that's okay.
-
I wouldn't alter you calibration targets just for a better report. If you can alter them and Print viewing conditions and are still happy with the results, that's okay.
Actually, that's why I went back up to 140cd/m2. I like the monitor a little brighter than the usual 120-ish. My prints are generally lit pretty well where they end up.
Dave
-
Actually, that's why I went back up to 140cd/m2. I like the monitor a little brighter than the usual 120-ish. My prints are generally lit pretty well where they end up.
Dave
Indeed, the 120cd/m2 "recommendation" urban legend has to cease; nothing special about that setting. Might be ideal, might not. Ditto with White Point. The numbers vary.
-
Simple: matches my viewing booth.
http://digitaldog.net/files/Print_to_Screen_Matching.jpg
(http://digitaldog.net/files/Print_to_Screen_Matching.jpg)
If your viewing booth is so bright, are you also displaying prints in a very bright environment? I understand that you're getting a good match, but, from my experience, most prints are viewed in darker conditions. But, maybe my experience is not usual?
-
If your viewing booth is so bright, are you also displaying prints in a very bright environment?
Doesn't matter where the prints are viewed after seeing them under the viewing booth to result in visual match of display + booth. Our eyes adapt (http://www.colorwiki.com/wiki/The_Ideal_Room_With_A_View) and a match isn't the goal any longer. This is all outlined in my video:
Why are my prints too dark?
A video update to a written piece on subject from 2013
In this 24 minute video, I'll cover:
Are your prints really too dark?
Display calibration and WYSIWYG
Proper print viewing conditions
Trouble shooting to get a match
Avoiding kludges that don't solve the problem
High resolution: http://digitaldog.net/files/Why_are_my_prints_too_dark.mp4 (http://digitaldog.net/files/Why_are_my_prints_too_dark.mp4)
Low resolution: https://youtu.be/iS6sjZmxjY4 (https://youtu.be/iS6sjZmxjY4)
-
Does anyone know the “average” life span of the bulbs in these NEC PA271q displays.
What is the first thing to go, color accuracy or luminance?
John
-
I am fighting with my brand new PA271q and found this thread...
It looks like mine is pretty bad, so I downloaded Andrew's profile to compare to his. It's much worse.
I'm on a iMac, so I can't get the firmware to update until I go to the office tomorrow. If I can't get this to update, I guess I'll be returning the monitor.
Has anyone gotten the FW to properly update when making the thumb drive on a Mac? I need suggestions, as i followed the directions precisely.
-
Has anyone gotten the FW to properly update when making the thumb drive on a Mac? I need suggestions, as i followed the directions precisely.
According to NEC tech support it will not work if the files are loaded from a Mac. I tried every way I could imagine and could not get it to recognize the file. You need a PC. I suggest doing the whole sequence from the PC: download the file, unzip and drop on the thumb drive.
Dave
-
According to NEC tech support it will not work if the files are loaded from a Mac. I tried every way I could imagine and could not get it to recognize the file. You need a PC. I suggest doing the whole sequence from the PC: download the file, unzip and drop on the thumb drive.
Dave
Thanks Dave. I saw your post earlier, and I was afraid that was going to be the answer. Odd that NEC doesn't appear to be aware of the existence of Apple computers in this respect...
I'll get this resolved over the weekend and then I'll be back to resolve a bunch of questions I have about the woefully inadequate manuals for this monitor and the Spectraview software (I'll start a new thread for that).
-
After the firmware update, the values are better, but nowhere near as good as Andrew's...
I'm wondering if I should return this to the vendor or talk to NEC. It's brand new, and this makes me think that there is a bad batch of panels out there. It's likely to get worse as the display ages, I'd expect. It makes me wonder what this will mean for the life of the display compared to an ideal display.
-
Michael,
I’m curious what happens if you set the target at 130 cd/cm2, 300 contrast ratio, white point 5800?
Dave
-
Here it is...
-
Well, so much for that theory. Mine was also no better after updating the firmware, but something changed after several calibrations. In my case, it was after I lowered the target to 130.
Also, in preferences, calibration tab, make sure you have 52 calibration steps selected.
As much as I subscribe to “insanity is doing the same thing over and over expecting a different result,” you might go through several calibrations, particularly ones that keep the black point between 0.4 and 0.5. Start by using “default” for the contrast ratio. After that first calibration with default contrast ratio, use a manual entry to get that black target set point between 0.4-0.5. I’m sure you know this, but just in case: black target is your setting / contrast ratio, so 150cd/m2 / 300 = 0.5 black target.
“Several” meaning 3-6.
Dave
-
Looks like that doesn't help me so far. The values aren't improving, and I've reset it a few times, calibrated all of the stock options and a few of the custom ones as well.
I think I just got a crappy panel in mine.
I'll try a few more variables, but frankly, for the price of this thing, i'm a bit pissed they don't have better quality control.
I'll be sending a message to tech support to see what they suggest.
-
Looks like that doesn't help me so far. The values aren't improving, and I've reset it a few times, calibrated all of the stock options and a few of the custom ones as well.
I think I just got a crappy panel in mine.
I'll try a few more variables, but frankly, for the price of this thing, i'm a bit pissed they don't have better quality control.
I'll be sending a message to tech support to see what they suggest.
Simple: Request a new replacement.
-
Already started the wheels on a return or replacement... barring a solution that NEC can offer with the existing display.
-
I just discovered something while I was on the phone with the service people at NEC...
My monitor firmware was updated to 1.302 per the firmware update, but the firmware in the Spectraview II report summary page is showing 1.008. The monitor reports 1.302 in the OSD.
I just tried doing the FW update again, and it is still showing the 1.008 in SV.
I'm not sure, but I suspect this may be why the calibration improved some from before, but didn't get fixed entirely. I think the FW update is not completely updating the monitor and it is leaving the Green channel non-updated.
Can anyone look at their FW number in SV (in the Summary sheet after a calibration), and tell me if they are getting the same result (that the FW on the OSD does not match the SV value)? Also, are you getting a good (less than 1 delta E) range of values as Andrew showed in his monitor calibration results?
I don't know if there was a FW update before this one (1.302), so if you have updated before this version, it may be working correctly and you may be getting good results. Or maybe not...
-
I just calibrated mine again; says the same thing. Results are bad but it hasn't been on long so might not be warmed up.
Dave
-
If you do a search for "Spectraview Updates" you will get a page that lists all the versions. The latest is 1.1.39. I called Tech Support the other day complaining of suddenly getting huge Delta E's. The rep advised that I should go back and install version 1.134 or earlier. I did that and initially still got large a Delta E but was also told that I needed to reset my luminance to 83CDM. (My target was 90). I did that and my Delta E was then .55. My screen brightness does not appear to have diminished at all and seems quite bright to me. I assume because my eyes/brain adjusted but maybe something else is going on.
-
I just calibrated mine again; says the same thing. Results are bad but it hasn't been on long so might not be warmed up.
Dave
Looks like yours is doing the same thing, but your results appear much better then mine, so nothing is clear to me at this point.
NEC has escalated the ticket to Level 2. Not sure what it means, but they are looking into it.
-
So here is a question: Is this problem that a screen must manifest straight out of the box, or is it the case that the screen may be fine initially but develop this problem over time?
From my experience and what I could glean from support, Straight out of the box or you are good.
Well I’m seeing behavior that makes me wonder about that conclusion. I purchased a new PA271Q, installed the latest version (1.1.39) of Spectraview, and first calibrated it on 7/25/19 using the included NEC MDSVSENSOR3. My initial calibrations using several different target settings showed Color Tracking Delta-E values similar to Andrew’s, and I thought I had escaped with a good unit that doesn’t exhibit the problems described in this thread. A few examples of the results:
(https://pbase.com/art1/image/169610300/large.jpg)
(https://pbase.com/art1/image/169610303/large.jpg)
(https://pbase.com/art1/image/169610302/large.jpg)
(https://pbase.com/art1/image/169610301/large.jpg)
However, 9 days later on 8/3, after a restart of my Windows 10 computer, a Spectraview message popped up on my screen saying that the monitor was not calibrated. What?? Not even out of date (which it shouldn’t have been), but uncalibrated. I scratched my head over that, but recreated the target files and calibrated the monitor again. Examples of the results:
(https://pbase.com/image/169610771/large.jpg)
(https://pbase.com/art1/image/169610306/large.jpg)
(https://pbase.com/art1/image/169610305/large.jpg)
As you can see from that last set, the D-E values were much larger for all the target calibrations. Not as bad as some here have reported, but problematic nevertheless. What caused this? What happened to make SV lose its previous calibrations, and to do much worse upon recalibration? I’ve spent the last several days pondering that and testing a variety of things, trying to get back to where I was with that first set of calibrations. I’ve gone back through this thread multiple times, looking at all the reports and suggestions. Here’s what I’ve tried, not necessarily in this order:
1. Different usb ports connecting the sensor (I had used one on the monitor on 8/3, but one on the computer on 7/25). Result: no effect.
2. Factory Reset the monitor. Result: made things a little worse.
3. Updated Firmware from R1.202 to r1.302, which Czornyj reported in this thread had fixed the problem. Result: a substantial improvement, but still much worse than my initial result on 7/25, and much worse than the results Andrew has shown in this thread for his monitor.
4. Monitor Self Calibration & Validation. Result: an additional small improvement, but still much worse than my initial result on 7/25, and much worse than the results Andrew has shown in this thread.
5. Loaded Czornyj’s monitor settings that he made available earlier in this thread. Result: made things a little worse.
6. Uninstalled Spectraview 1.1.39 and installed 1.1.37, the earliest version claiming support for the PA271Q. Result: made things a little worse. (I was not able to roll back to v.1.1.34 as user Jmlphotography said above that he was told to do by NEC, because that version does not support the PA271Q, and SV thus balked when I tried. Lowering luminance of target to 83, from my target’s 120, as he said NEC also suggested, did not help.)
7. Calibration using a ColorMunki Photo sensor. Result: gave values very similar to MDSVSENSOR3, so no improvement.
8. Calibration precision test. Result: I found that if calibrations were repeated one right after the other for the same target, there seemed to be a small but steady rise in the Color Tracking D-E values; but not if some time (e.g., 1h) elapsed between them.
Thus the only things that have so far helped for me are updating the monitor firmware to version 1.1.39, and performing a monitor Self Calibration & Validation. Here are examples of my most recent SV calibrations, including one with Andrew’s target that he provided us:
(https://pbase.com/art1/image/169610310/large.jpg)
(https://pbase.com/art1/image/169610309/large.jpg)
(https://pbase.com/art1/image/169610307/large.jpg)
(https://pbase.com/image/169610843/large.jpg)
Where to go from here? Given the apparent prevalence of this problem, as illustrated in this thread, I’m not confident that a return for replacement will yield a unit lacking the issue. And even this one seemed OK at first, which is particularly puzzling. Still, I’ve been happily using this NEC line for years, and have another one at a separate workstation, so I don’t want to switch to another brand.
Are we making too much of this issue? For me, the calibration targets for printing that have reduced contrast, e.g. 200:1 or 300:1, give the lowest Grayscale Delta-E values, and that is the situation for which color fidelity is most important. Average D-E values of about 0.75 and max values of about 1.20, as I’m getting now for target 6000K 120cdm 200-1Contrast, are I think considered quite good, even though they are still significantly higher than the values of avg =0.27 and max=0.61 that I saw on my very first calibration of this target. My biggest fear though is that these somewhat elevated numbers might indicate an increased possibility of early failure, and an accelerated increase in tracking error over time.
-
Query: are the NEC Spectraview monitors worth buying?
-
Query: are the NEC Spectraview monitors worth buying?
My opinion based on my current experience, is no. I'm saying that as someone who has had quality NEC LCD displays since the beginning.
If they fix the issue, I will change my response, but frankly, I wouldn't pay $1300 for a monitor that calibrates like this fresh out of the box. Add to that the observations of Arlen above, and I wonder if they have manufactured something with an early failure problem that is going to yield loads of issues in the years to come.
-
Well I’m seeing behavior that makes me wonder about that conclusion. I purchased a new PA271Q, installed the latest version (1.1.39) of Spectraview, and first calibrated it on 7/25/19 using the included NEC MDSVSENSOR3. My initial calibrations using several different target settings showed Color Tracking Delta-E values similar to Andrew’s, and I thought I had escaped with a good unit that doesn’t exhibit the problems described in this thread. A few examples of the results:
...
Where to go from here? Given the apparent prevalence of this problem, as illustrated in this thread, I’m not confident that a return for replacement will yield a unit lacking the issue. And even this one seemed OK at first, which is particularly puzzling. Still, I’ve been happily using this NEC line for years, and have another one at a separate workstation, so I don’t want to switch to another brand.
Are we making too much of this issue? For me, the calibration targets for printing that have reduced contrast, e.g. 200:1 or 300:1, give the lowest Grayscale Delta-E values, and that is the situation for which color fidelity is most important. Average D-E values of about 0.75 and max values of about 1.20, as I’m getting now for target 6000K 120cdm 200-1Contrast, are I think considered quite good, even though they are still significantly higher than the values of avg =0.27 and max=0.61 that I saw on my very first calibration of this target. My biggest fear though is that these somewhat elevated numbers might indicate an increased possibility of early failure, and an accelerated increase in tracking error over time.
Arlen, can you look at the summary pages and determine what firmware the SV software thinks you have on the monitor BEFORE and AFTER the FW update?
Someone else had commented on their calibrations getting worse earlier in the thread, but I believe they thought it was tied to updating the SV software. Maybe it wasn't after all.
Thanks,
---Michael
-
Yes, thanks for reminding me, I forgot to mention that. All of the summary pages, both before and after the FW update, have V1.008 listed for the version. EXCEPT: for the very first calibrations that I ran, on 7/25, that looked good but later disappeared; they all say "N/A" for Firmware, Current Usage, and Last Calibration. Strange.
-
Yes, thanks for reminding me, I forgot to mention that. All of the summary pages, both before and after the FW update, have V1.008 listed for the version. EXCEPT: for the very first calibrations that I ran, on 7/25, that looked good but later disappeared; they all say "N/A" for Firmware, Current Usage, and Last Calibration. Strange.
Interesting... thanks for the info. It sounds to me like SV is not properly pulling that from the monitors.
Do you have a USB connection to your monitor in addition to the video connection? I haven't tried disconnecting that to see what will happen with SV and calibration. I'll try that tonight.
-
That's a good thought. I do have a USB connection between computer and monitor. But I'm not sure that I did on 7/25, when I first set it up and ran those first calibrations. I'll test disconnecting that too, but right now I have to shut down my computer because we're being hit by an electrical storm.
-
That's a good thought. I do have a USB connection between computer and monitor.
So do I. I will run some tests tonight too. Your comment about multiple calibrations in succession is also interesting. That’s when I got the best results.
Dave
-
Now the menu system doesn't work at all. This thing is driving me crazy.
-
Unplugged everything, switched to USB-C / Thunderbolt 3 monitor connection. MDSVSENSOR3 connected directly to computer.
On-screen menu still doesn't work, but for some reason the results are better.
-
Your results look better than any of my recent ones, Dave. If I get those results consistently, I think I will just call it good enough.
My monitor to computer connection is Displayport/Displayport. I removed the USB connection and ran the calibration on my standard print settings target. Restarted the computer, waited 30min and ran it again. No significant difference without the USB connection, unfortunately. D-E avg. ~0.70, max ~1.17.
-
After another calibration, mine is back to (ab)normal. ~1 avg dE, max ~2.8. No correlation with anything that I can figure out. Maybe how accurate the dark target ends up, but not sure.
Regardless, I've compared standard images when I get "good" results like my previous post vs. the more normal ~ 1avg/3max, and I cannot see any difference unless I really dig in shadows and convince myself I can. So, in my case I'm chasing ghosts for the most part. That doesn't mean it is ok, nor am I suggesting anyone else come to that conclusion. I've just decided to move on and not worry about it. I still suspect Marcin is correct and this is just a software/firmware problem that will eventually be fixed. In the mean time, I'll post if/when I learn anything new.
Dave
PS. A few days ago I updated new a new SV version 1.1.39 / build 181031. When I did that, there was a setting in preferences to lock the menu when calibrated that got checked as a result of the update. That's why my on-screen menu wouldn't work.
-
After another calibration, mine is back to (ab)normal. ~1 avg dE, max ~2.8. No correlation with anything that I can figure out. Maybe how accurate the dark target ends up, but not sure.
Regardless, I've compared standard images when I get "good" results like my previous post vs. the more normal ~ 1avg/3max, and I cannot see any difference unless I really dig in shadows and convince myself I can. So, in my case I'm chasing ghosts for the most part. That doesn't mean it is ok, nor am I suggesting anyone else come to that conclusion. I've just decided to move on and not worry about it. I still suspect Marcin is correct and this is just a software/firmware problem that will eventually be fixed. In the mean time, I'll post if/when I learn anything new.
Dave
PS. A few days ago I updated new a new SV version 1.1.39 / build 181031. When I did that, there was a setting in preferences to lock the menu when calibrated that got checked as a result of the update. That's why my on-screen menu wouldn't work.
Yours are pretty god and are clearly better than mine. I think I'd be satisfied if I were getting your results as well.
-
Possibly unrelated (who knows at this point?) but can anyone tell me if the XRite iDisplay Pro can be used for the internal calibration on this monitor?
I've had this for a few years because I have more than the NEC monitor here. It appears to recognize it fine, but the internal calibration fails when I try that. I suspect they locked it to the NEC only model, but I haven't found anything that confirms that the NEC model is the only one that can do internal calibration.
Thanks,
---Michael
-
I don't have that sensor, but the release notes for the latest version of Spectraview says it is compatible with that program: "X-Rite i1Display Pro: This version includes support for the X-Rite i1Display Pro color sensor." Moreover, I'm able to calibrate the PA271Q with the XRite Colormunki Photo sensor, in addition to the NEC sensor, so it's not locked to just the NEC branded model.
-
Thanks Arlen,
I got confirmation from NEC just now that they have not locked the display to the NEC-only version of the x-Rite i! Display Pro, so there is no explanation for why it may be having problems getting an internal calibration.
I'll try again tonight, but I think there may be something wrong with this monitor...
-
Michael, I may have misunderstood what you're trying to do. When you say internal calibration, do you mean Stand Alone Calibration/Self Calibration (p. 25 of the manual), connecting the sensor via the SENS/MEM top USB port? If so, then it appears that only the latest NEC sensor, MDSVSENSOR3, will work for that. The manual says "connect a supported color sensor", and then on p. 26 it says "The supported USB color sensor for SENS/MEM port is MDSVSENSOR3". It doesn't list any other supported sensors for that purpose.
-
Arlen,
Yes, the internal calibration that does not use the SV2 software. Mine won't perform that without an error. NEC has told me that they do not lock out the x-Rite device, so that shouldn't be the source of the problem.
A few days ago, I received some advice from NEC recommending that I run the monitor for a few hours before calibration, but that actually made the calibration considerably worse... I'm wondering if they are even trying some of these things before they send advice out.
I just received quite disappointing news that NEC does not believe the problems I am experiencing is a 'defect' per se, and that any replacement they may offer will likely exhibit the same problem. So it sounds like they are somewhat unwilling to make a replacement and that means I will have to return it to the vendor for a credit or exchange.
I'm so disappointed with NEC after having used a handful of their monitors over the years to good effect (since the late 90's or so). Not sure what I am going to do, as the Eizo monitors are identical in most respects, so I have little faith they are substantively better, and they certainly are a lot more expensive. They are way outside what I wanted to spend on a monitor right now.
-
A few days ago, I received some advice from NEC recommending that I run the monitor for a few hours before calibration, but that actually made the calibration considerably worse... I'm wondering if they are even trying some of these things before they send advice out.
They told me the same thing. In my case, it did make an improvement. I've calibrated twice in a row with better results. That's not much of a sample population but I will continue to do that and see how it goes. I use Amphetamine to keep things awake. I also thought at one point Andrew said he leaves his on all the time, and leaving them on is ok with these monitors. I can't find that now, so I may be mistaken.
Regardless, I do feel like NEC are making generic recommendations that may or may not help, so feedback to them is probably worthwhile.
Dave
-
Michael, I've also seen a trend for worse results after the monitor has been on for a while, vs. waiting just 30 minutes after it's turned on. However, I've run a lot of these calibration tests now and don't feel confident that anything like that is consistently reproducible for me. On average my results are now somewhat better than I reported in my last post. Can't really say why, though.
My best set of values, obtained 40 min after turning the monitor on, for my usual printing target 6000K 120cdm 200contrast were Avg=0.54, Max=0.82; and a few minutes later for Andrew's target ARsPhotoEditing values were Avg=0.56, Max=0.83. These values are not much higher than the two sets of values Andrew reported for his target, Avg=0.32, Max=0.57 and Avg=0.44, Max=0.72.
But most of the time my values for my target are in the range of Avg 0.60-0.95, Max 0.93-1.39.
I'm tentatively concluding that those values are good enough, given that dE values lower than 1.0-1.25 are said to be indistinguishable to the human eye, and I can't see any noticeable problems when I look at my images. So I think I'll stick with this monitor. But my mind is still open to other arguments.
P.S. For what it's worth, I also have a 3 year old, heavily used PA272W monitor, and its latest 2 sets of calibration values for my target are Avg=0.57, Max=0.96; and Avg=0.45, Max=0.76. Not much different than the values Andrew reported for his PA271Q.
-
Here's my values for a D50 Photo editing after about 4 hours... really not good at all.
-
I agree, those values look pretty bad.
I don't have access to my PA271Q right now, but I made a Photo Editing D50 target (just modified the stock Photo Editing target to D50, no other changes) and used it to calibrate my older PA272W. The Dark Values are similar to yours, but all the rest are much lower, as evident when they are excluded in the second image below.
(https://pbase.com/art1/image/169639874/original.jpg)
(https://pbase.com/art1/image/169639875/original.jpg)
You might try using Andrew's target file ARsPhotoEditing, so you can directly compare to the results he presented for his apparently good PA271Q monitor. That's what I've been doing. You can download his target file at the link he provided, or just modify the stock Photo Editing target to white point 5750 and contrast ratio 300:1.
It's looking like there are some PA271Q monitors like Andrew's that are good, others like mine that after some tweaking have increased dE values but are not too bad, and still others like yours and Dave's that are pretty bad.
I'm still intrigued by Czornyj's demonstration of a dramatic reduction in dE values for his initially bad monitor (post #60), though he didn't include dark values like Andrew did; and his report (post #105) that the latest firmware update fixed all 5 monitors where he had tried it.
-
My monitor is going back. I'll report back once I have a new one running.
-
Hi all!
New user just signed up. Architectural Photographer.
Anyway I just bought a BenQ SW271 and found out that their internal calibration color profiles are incorrect which causes calibration issues.
So.... I bought a NEC PA271Q which apparently has the same issue and may be the cause of these issues everyone is talking about.
Here is some brief info:
https://hub.displaycal.net/forums/topic/pa271q-target-white-point-error/
Using the wrong calibration panel correction can def cause issues, for instance higher Green curve, etc. And if NEC is using the wrong one in SpectraView (which some have confirmed is the case) they this will def cause issues. Just food for thought.
I need to test my calibration (fingers crossed). Waiting on my SpectraView software to deliver.
Anyone else having any more luck or input?
PS - What the hell is with all of these stupid verification questions, especially with a paid account.
This is going to get very old, very fast.
-
FWIW, I just sent NEC an email requesting their POV on the issues everyone has raised. I included links to this discussion as well as the one Gette posted.
I also asked them to post their response here.
-
Interesting thread. I just got this monitor a few days ago from B&H and went over all the options, read everything I could, and did the calibration with the SVII 1.1.40 latest software. I calibrate for 100nits, 2.2, D65, and native gamut. I was surprised that the delta Es were so bad. See attached. My monitor is on the latest firmware too - R1.302 (even though as others have pointed out, the SVII software incorrectly reports it as V1.008). I also have two X-Rite i1Display Pro devices that I used, with similar results. All in controlled room lighting (i.e. dark), etc. I even checked the options in SVII to "Average low luminance levels", 52 (Highest Quality) steps, and "Extend luminance stabilization time". I did many runs with both i1 devices with those options ON and OFF, with different gamuts and peak luminance values, etc. I tried Andrew's target from this thread as well. No matter what I do, this is the best delta Es I can get. Is it a panel issue? Is it a SVII software issue? I don't know. It doesn't look great however.
I also noticed a few items in the OSD menu that I wondered would impact or not things - options like "Uniformity" which is by default set to 5. I set it to OFF, but SVII puts it back to 5. Another option, "Color Stabilizer" defaults to ON. With it both ON or OFF, no difference. A lot of small items like this that I'm not sure what benefit they add if any. Or what potential problems they may cause or not. The documentation is not very descriptive. Am I frustrated? A bit. I expected more. Would an Eizo CG279X be better? Not sure. Would another NEC? Sounds like a crapshoot. So I guess it comes down to accepting it or spending more time figuring it out by exchanging, or getting an Eizo.
One other thing I noticed is that the OSD menu shows the Preset as "SpectraViewII", but the luminance and white point values are not the same as the ones on the calibration report in SVII. So yeah, it seems like NEC is not designing the best software or integrated system. I was really surprised it only does grayscale calibration and just measures 3 patches of the primaries for the chosen color gamut. Why not do many color patches for the CMS if it can program the 3D LUT? Strange. On the fence for now...