Ad
Ad
Ad
Pages: [1] 2 3 ... 7 »   Bottom of Page
Print
Author Topic: SpectraView II matrices for wide gamut monitors?  (Read 24762 times)
shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« on: June 21, 2010, 12:58:36 PM »
ReplyReply

Does anybody know whether SpectraView II uses software matrices to correct for wide gamut monitors when used in connection with regular colorimeters like the i1d2 and dtp94?

thanks in advance

shaya
Logged
shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« Reply #1 on: June 21, 2010, 01:55:49 PM »
ReplyReply

I just got confirmation via email from NEC Display Service & Support that SpectraView II does NOT make any matrices correction for wide gamut. The sensor has to support the wide gamut.

There are only three (or four) sensors that supports wide gamut,
 - the NEC custom mated i1d2 and
 - the Spyder 3.
 - Eizo's and Quato's bundle of matched screen & colorimeter using the dtp94
« Last Edit: June 21, 2010, 02:09:27 PM by shayaweiss » Logged
Steve Weldon
Sr. Member
****
Offline Offline

Posts: 1445



WWW
« Reply #2 on: June 21, 2010, 02:37:17 PM »
ReplyReply

Quote from: shayaweiss
I just got confirmation via email from NEC Display Service & Support that SpectraView II does NOT make any matrices correction for wide gamut. The sensor has to support the wide gamut.

There are only three (or four) sensors that supports wide gamut,
 - the NEC custom mated i1d2 and
 - the Spyder 3.
 - Eizo's and Quato's bundle of matched screen & colorimeter using the dtp94
NEC specifically told me the i1d2 supports wide gamut and its listed in their manual as doing so..

Not only that.. it actually works and profiles the wide gamut as you can clearly see from the attachments..

After you buy your NEC puck.. it will be interesting to see you run it side by side to your current i1d2 puck and show us the difference..

Still curious about the differences of the Euro vs. America vs. Asian spec monitors..

[attachment=22722:i1d2.JPG]  [attachment=22723:i1d2a.JPG]
Logged

----------------------------------------------
http://www.BangkokImages.com
shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« Reply #3 on: June 21, 2010, 03:50:06 PM »
ReplyReply

Quote from: Steve Weldon
NEC specifically told me the i1d2 supports wide gamut and its listed in their manual as doing so..

I think it just means they work, but those four work better. It again boils down to your needs. I would like to hear from the specialists, what the actual visual difference should be between a calibration with or without a puck corrected for wide gamut. All I can tell you is, that as long as you have only one monitor, and your eyes and/or print to screen match are not that demanding, or you are printing out of house, than you might not notice. I can't tell you all the specific situation that exist of way you might not notice. I did not see it, simply because I never saw something else, or better. But everything changed with two monitors, my eyes couldn't get used. As they had not the same tint, and not to the same degree, I suddenly realized that maybe the tint could go away completely. The only remedy was, you remember,  a colorimeter corrected for wide gamut in the software or the hardware. I am sure I could have used a spectrometer, with similar results, though some say that the colorimeters do better with displays.

Quote
Not only that.. it actually works and profiles the wide gamut as you can clearly see from the attachments..

Yes, it works, also with my monitors, all the curves and diagrams, but somehow it's different, uncorrected, unfiltered for wide gamut. Also you could run a validation, but for that you have to use a different instrument and/ or software to make sens. From experience, I can tell you, that there are variances not only between instruments, but also between different software.

Quote
After you buy your NEC puck.. it will be interesting to see you run it side by side to your current i1d2 puck and show us the difference..

I am already working on that, but it is very difficult to get one at a reasonable price and have it shipped overseas. Suggestions anyone?

Quote
Still curious about the differences of the Euro vs. America vs. Asian spec monitors..
I don't know about Asia. The MultiSync from Europe and America are identical, as stated in these pages. If I remember correctly what was said (for example here http://luminous-landscape.com/forum/index....showtopic=42040 and here http://luminous-landscape.com/forum/index....howtopic=44312), the difference is only in the Spectra View models. In America they are the same as the MultiSync but ship with the SpectraView software and the NEC i1d2 puck corrected for wide gamut. In Europe the SpectraView are handpicked among the best monitors, they come with a hood and a version for NEC of BasICColor display 4, and therefore they are also much more expensive.
Logged
Czornyj
Sr. Member
****
Online Online

Posts: 1371



WWW
« Reply #4 on: June 21, 2010, 04:49:06 PM »
ReplyReply

Quote from: shayaweiss
There are only three (or four) sensors that supports wide gamut,
 - the NEC custom mated i1d2 and
 - the Spyder 3.
 - Eizo's and Quato's bundle of matched screen & colorimeter using the dtp94

Spyder3? I wouldn't count on it  

Eizo & Quato use correction matrix for in ColorNavigator & iColor display software - it works in combination with trusty DTP94, that has decent inter instrumental agreement.

NEC uses custom mated colorimeters, and that's why they don't use correction matrix in the software.

Quote from: shayaweiss
I don't know about Asia. The MultiSync from Europe and America are identical, as stated in these pages. If I remember correctly what was said (for example here http://luminous-landscape.com/forum/index....showtopic=42040 and here http://luminous-landscape.com/forum/index....howtopic=44312), the difference is only in the Spectra View models. In America they are the same as the MultiSync but ship with the SpectraView software and the NEC i1d2 puck corrected for wide gamut. In Europe the SpectraView are handpicked among the best monitors, they come with a hood and a version for NEC of BasICColor display 4, and therefore they are also much more expensive.

All NEC panels are physically the same - the only difference is, that hand picked european Spectraview panels have firmware modification, and NEC EU blessing - so they can be hardware calibrated with Spectraview display a.k.a. basICColor. In the past there was a method to turn x90 series panel into "Spectraview mode", using simple key combination - now it's more complicated.

US Spectraview II represents more democratic approach, and calibrates all x90, PA and P series NEC displays.
« Last Edit: June 21, 2010, 05:08:33 PM by Czornyj » Logged

shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« Reply #5 on: June 21, 2010, 04:55:27 PM »
ReplyReply

Quote from: Czornyj
Spyder3? I wouldn't count on it  

Eizo & Quato use correction matrix for in ColorNavigator & iColor display software - it works in combination with trusty DTP94, that has decent inter instrumental agreement.

NEC uses custom mated colorimeters, and that's why they don't use correction matrix in the software.

Exactly.

And as you know, I recently found out through trial and error.

Only, I have no experience with the spyder3. I didn't even hear of it a lot. I mentioned it because the NEC display technician told me in his e-mail, that there where two instruments that are corrected for wide gamut monitors that he knew of, their own i1d2 and the spyder3 .  But why is it not dependable? I was thinking of it, as it is much easier to get than the NEC puck. But if you say so, I will most certainly stay away from it. Just maybe could you give us some hints or details about whats not so good about it.
« Last Edit: June 21, 2010, 05:02:53 PM by shayaweiss » Logged
shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« Reply #6 on: June 21, 2010, 04:58:37 PM »
ReplyReply

Here is the above mentioned email (with permission)

-----Original Message-----
From: NEC Display Service & Support [mailto:techsupport@necdisplay.com]
Sent: Monday, June 21, 2010 9:39 PM
To: #######################
Subject: RE: NEC sensor matrix VS SpectraView II matrix
(#8477-180783712-8459)

Shaya,

Thank you for contacting NEC Display
The calibration software does not make any matrices correction for wide gamut The sensor has to support the wide gamut.

There are only two sensors that I'm aware of that supports wide gamut, Ours and the Spyder 3

You can read more about Spectraview and download the users manual under the downloads tab at:
http://www.necdisplay.com/supportcenter/mo...ectraview2/faq/

M. Daly
NEC Display Service & Support
Telephone: 1-800-632-4662
Email: techsupport@necdisplay.com
« Last Edit: June 21, 2010, 05:25:20 PM by shayaweiss » Logged
Czornyj
Sr. Member
****
Online Online

Posts: 1371



WWW
« Reply #7 on: June 21, 2010, 05:35:03 PM »
ReplyReply

Quote from: shayaweiss
Exactly.

And as you know, I recently found out through trial and error.

Only, I have no experience with the spyder3. I didn't even hear of it a lot. I mentioned it because the NEC display technician told me in his e-mail, that there where two instruments that are corrected for wide gamut monitors that he knew of, their own i1d2 and the spyder3 .  But why is it not dependable? I was thinking of it, as it is much easier to get than the NEC puck. But if you say so, I will most certainly stay away from it. Just maybe could you give us some hints or details about whats not so good about it.

http://www.lumita.com/site_media/work/whit...xrite-wp-3a.pdf

In theory, Spyder3 with 7 color filters should match CIE XYZ function better. In practice - it seems there's still something wrong.
« Last Edit: June 21, 2010, 05:37:43 PM by Czornyj » Logged

shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« Reply #8 on: June 21, 2010, 05:59:08 PM »
ReplyReply

Quote from: Czornyj
http://www.lumita.com/site_media/work/whit...xrite-wp-3a.pdf

In theory, Spyder3 with 7 color filters should match CIE XYZ function better. In practice - it seems there's still something wrong.

Thank you for this highly informative document by Karl Lang! I understand that a custom filtered colorimeter will be the best choice. But I didn't read any hint to the 7 filters, or why it should not work.
« Last Edit: June 21, 2010, 06:02:34 PM by shayaweiss » Logged
Steve Weldon
Sr. Member
****
Offline Offline

Posts: 1445



WWW
« Reply #9 on: June 22, 2010, 12:38:03 AM »
ReplyReply

Quote from: shayaweiss
Yes, it works, also with my monitors, all the curves and diagrams, but somehow it's different, uncorrected, unfiltered for wide gamut. Also you could run a validation, but for that you have to use a different instrument and/ or software to make sens. From experience, I can tell you, that there are variances not only between instruments, but also between different software.
I'm not buying my profiles are uncorrected.  They perfectly match the physical color checker on every shade.. and I mean perfectly.

I can accept you've had problems with your Eizo.. but you haven't yet used your i1d2 with the NEC.   And when I wrote NEC they told me (and listed in their instruction manuals and on-line text) the i1d2 was compatible with their wide gamut monitors.  

Curious though.. I know people using the Munki, most are using the i1d2, some the SVII.. are they creating different profiles?  Probably.  Probably small differences.  The question is:  Is one profile more correct than the other.  How would you test it?  I test against a physical color checker.  Anyone else?
Logged

----------------------------------------------
http://www.BangkokImages.com
Steve Weldon
Sr. Member
****
Offline Offline

Posts: 1445



WWW
« Reply #10 on: June 22, 2010, 12:40:20 AM »
ReplyReply

Quote from: Czornyj
All NEC panels are physically the same - the only difference is, that hand picked european Spectraview panels have firmware modification, and NEC EU blessing - so they can be hardware calibrated with Spectraview display a.k.a. basICColor. In the past there was a method to turn x90 series panel into "Spectraview mode", using simple key combination - now it's more complicated.

US Spectraview II represents more democratic approach, and calibrates all x90, PA and P series NEC displays.
Thanks.  This helps clear things up.

What do Europeans use for calibration if they don't have a reference standard?
Logged

----------------------------------------------
http://www.BangkokImages.com
shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« Reply #11 on: June 22, 2010, 04:50:28 AM »
ReplyReply

Quote from: Steve Weldon
And when I wrote NEC they told me (and listed in their instruction manuals and on-line text) the i1d2 was compatible with their wide gamut monitors.

Steve,

I don't know what they told you, but it is not what Karl Lang writes (see the document above), it is not what I got as information from NEC (see email above), it is also not what Czornyj and others have wrote here in this forum (see quotes above), and it is not what you can read on the Argyll/dispcalGUI information (see quote http://luminous-landscape.com/forum/index....t&p=372083), and it does also not meet what I have read from other specialists across the Web.

I'm not a specialist, and you know that, but I try to understand. So what can NEC mean, when they nonetheless state in their instruction manual that the i1d2 and other unfiltered colorimeters are compatible? (BTW is there any instrument that is not in that list?) I think that's what I wrote. They mean to say that it works, just not so well.

Also, as I said, I can visually tell the difference between a calibration and profile made with a corrected colorimeter or an uncorrected one. You have to try it, to see it. Alternatively, you could try a software that corrects for wide gamut, like Quato's iColor Display 3, though the latter would not be able to do hardware calibration. As mentioned earlier, the correction can be made either in the hardware or the software.
« Last Edit: June 22, 2010, 05:30:22 AM by shayaweiss » Logged
shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« Reply #12 on: June 22, 2010, 05:01:06 AM »
ReplyReply

Quote from: Steve Weldon
I test against a physical color checker.

What is your procedure?
Logged
Czornyj
Sr. Member
****
Online Online

Posts: 1371



WWW
« Reply #13 on: June 22, 2010, 05:28:25 AM »
ReplyReply

Quote from: Steve Weldon
What do Europeans use for calibration if they don't have a reference standard?

European Spectraview display profiler has correction matrix for NEC display spectra and popular colorimeters, so it's the same approach as Eizo ColorNavigator, and Quato iColor display.
NEC Display America choosed to build custom mated colorimeter rather than make software matrix correction. The Spectraview II checks the sensor type, and when it's not NEC colorimeter or spectrophotometer it creates profiles using factory avaraged measurements taken during production. See Spectraview II Edit>Preferences>ICC Profile tag.

Testing the profile quality against CC24 gives only overall idea of how the profile works, and as a matter of fact it's impossible to get 100% visual connection - the spectra of the displays backlight and the spectra of our lighting is different, and so on...
« Last Edit: June 22, 2010, 05:29:07 AM by Czornyj » Logged

Steve Weldon
Sr. Member
****
Offline Offline

Posts: 1445



WWW
« Reply #14 on: June 22, 2010, 10:05:22 AM »
ReplyReply

Quote from: shayaweiss
Steve,

I don't know what they told you, but it is not what Karl Lang writes (see the document above), it is not what I got as information from NEC (see email above), it is also not what Czornyj and others have wrote here in this forum (see quotes above), and it is not what you can read on the Argyll/dispcalGUI information (see quote http://luminous-landscape.com/forum/index....t&p=372083), and it does also not meet what I have read from other specialists across the Web.

I'm not a specialist, and you know that, but I try to understand. So what can NEC mean, when they nonetheless state in their instruction manual that the i1d2 and other unfiltered colorimeters are compatible? (BTW is there any instrument that is not in that list?) I think that's what I wrote. They mean to say that it works, just not so well.

Also, as I said, I can visually tell the difference between a calibration and profile made with a corrected colorimeter or an uncorrected one. You have to try it, to see it. Alternatively, you could try a software that corrects for wide gamut, like Quato's iColor Display 3, though the latter would not be able to do hardware calibration. As mentioned earlier, the correction can be made either in the hardware or the software.
I'm not sure what qualified as a "Specialist.."  I'm just a guy who has been using color profiled monitors for well over a decade and have helped others set up dozens of systems.  I have 'some' experience, but nothing that qualifies me as a "specialist" and I'm naturally skeptical of anyone making that claim.

I'm just not buying NEC puts a colorimeter in their catalog and instruction manuals as "Compatible, but just not so well.."  That's silly.  I do believe it's possible all colorimeters listed meet minimum tolerances and some 'might' hold tighter (but not different) tolerances.. but now we're taking the word of "specialists" on the net.. already explained how I feel about that.

You have the NEC puck as well as the 1id2?  And you've used them on a known good NEC monitor?   I'm sorry, but from what you've listed of your experiences you can't really be sure of anything.  You've had a couple questionable Eizo monitors, many different programs, and a testing strategy that (to me) doesn't appear logical in the sense that you were trying to pin down specific things that work or not.. only if it worked.  There's a difference.

And no.. anyone running an NEC monitor and not using SVII is defeating the main advantages of owning an NEC monitor with DDC support.. I hope you don't plan on using anything but SVII.. if you do good luck.

Lets look at it this way.  I've been using color profiled monitors for many years.. on matched monitors and single workstations.  I know what the colors should look like and if they weren't correct I'd know and my entire workflow wouldn't work.  sRGB emulation is new to me.. but my colors are holding across all gamuts providing I work my entire workflow with the same gamut/profile.  Test images from Phase One and others (images worked up to show the color checker and more just for this purpose) and a physical color checker are perfect.  IF I did run another puck and got other colors.. they wouldn't be right.  Or the differences would be so small only the colorimeter would see it.. and not me.

Do I believe you're having issues?  Yes.  But I'm guessing a lot of them are a.  Self induced  or  b.  You're worrying about things you shouldn't yet be worried about concerning the NEC monitor.

Now.. when you get your NEC monitor and you have both the NEC puck and the i1d2 puck.. and you run them side by side and tell me you're getting 'significant' visible differences.. then we'll talk more.  Until then you can keep reading all the "specialist" talk on the net and wringing your hands and anticipating issues you might or might not have when you get your new monitor.

How do you know a color profile is correct?  You compare it to references.  A color checker print out in a light tent compared to the real thing gets you really close.

There are two reasons you color profile.

1.  So the prints you make, either from your own printer or a print shop, look as they should.

2.  So your files are the same as a standard that you can share across the profession.

If they do 1&2 above.. to the tolerances needed for your work.. then you're color profiled.  

It's going to be very interesting when you get your NEC monitor..
Logged

----------------------------------------------
http://www.BangkokImages.com
Steve Weldon
Sr. Member
****
Offline Offline

Posts: 1445



WWW
« Reply #15 on: June 22, 2010, 10:11:23 AM »
ReplyReply

Quote from: Czornyj
European Spectraview display profiler has correction matrix for NEC display spectra and popular colorimeters, so it's the same approach as Eizo ColorNavigator, and Quato iColor display.
NEC Display America choosed to build custom mated colorimeter rather than make software matrix correction. The Spectraview II checks the sensor type, and when it's not NEC colorimeter or spectrophotometer it creates profiles using factory avaraged measurements taken during production. See Spectraview II Edit>Preferences>ICC Profile tag.

Testing the profile quality against CC24 gives only overall idea of how the profile works, and as a matter of fact it's impossible to get 100% visual connection - the spectra of the displays backlight and the spectra of our lighting is different, and so on...
Great explanation.. This explains why a 1id2 'could' be very accurate, or slightly less accurate but close, with a NEC monitor using SVII.. and not work well at all on another brand..

It also makes me think an actual colorimeter that can measure without the need of a corrected profile would be the 'ideal' option.. and probably the most accurate consistently.  But also a corrected puck could still be very accurate assuming the person has an "average" sample..

And yes.. 100%.. never gonna happen..   Totally agree.   There comes a point where your experience and workflow matter more than numbers..
Logged

----------------------------------------------
http://www.BangkokImages.com
shayaweiss
Jr. Member
**
Offline Offline

Posts: 87


« Reply #16 on: June 22, 2010, 10:30:51 AM »
ReplyReply

I don't need decades of experience to see that a monitor has color cast or doesn't much a second monitor. I do also not need that kind of background, to see that a change of Instrument and/or software does change the outcome, and that's no anticipatory talk.

For the rest, as I told you, your critics are pointed at the wrong person, I am not a specialist and have no experience. You say different then the others (I gave you a whole list already), I don't mind, you still might be right, but chances are slim. That is what my professional experience is able to tell, as I spent decades in humanities. For the rest I am just not the right interlocutor.

I would be very pleased to follow your exchange with them. As my reason for writing here is to understand and learn. If you want to knock me down with years of experience, even though what you say goes against this or that view expressed here and there by others with decades of experience, than there is no reason to talk anymore.

Again, I would be pleased if these technical issues could be addressed in a civilized manner from now on.

thank you.

P.S.: Very interesting that you do not address your points with Czornyj, whom I was quoting among others.
« Last Edit: June 22, 2010, 10:36:40 AM by shayaweiss » Logged
AndreaPiaggesi
Newbie
*
Offline Offline

Posts: 7


« Reply #17 on: June 22, 2010, 10:38:12 AM »
ReplyReply

Quote from: Czornyj
NEC Display America choosed to build custom mated colorimeter rather than make software matrix correction. The Spectraview II checks the sensor type, and when it's not NEC colorimeter or spectrophotometer it creates profiles using factory avaraged measurements taken during production. See Spectraview II Edit>Preferences>ICC Profile tag.

It's interesting, so the Spectraview software should use the matrix corrections with the i1Display2 colorimeter.
Sorry for the question buy why the Spyder3 is not recommended?


I live in Europe and I can't afford the original NEC colorimeter (the cost of shipping and tax is too high, it's better to get the Spectraview reference version in this case) so I've two ways:
1) the Spectraview II software with the Spyder3 colorimeter;
2) the Spectraview II software with the i1Display2 colorimeter and correction matrix.

I already own the i1Display2 and I'm thinking to sell it for the Spyder3.

Thanks!!
Logged
Czornyj
Sr. Member
****
Online Online

Posts: 1371



WWW
« Reply #18 on: June 22, 2010, 11:05:14 AM »
ReplyReply

Quote from: AndreaPiaggesi
It's interesting, so the Spectraview software should use the matrix corrections with the i1Display2 colorimeter.
Sorry for the question buy why the Spyder3 is not recommended?


I live in Europe and I can't afford the original NEC colorimeter (the cost of shipping and tax is too high, it's better to get the Spectraview reference version in this case) so I've two ways:
1) the Spectraview II software with the Spyder3 colorimeter;
2) the Spectraview II software with the i1Display2 colorimeter and correction matrix.

I already own the i1Display2 and I'm thinking to sell it for the Spyder3.

Thanks!!

Spyder3 seem to have poor inter instrumental agreement, and I simply wouldn't trust that sensor. It has been discussed in this topic, post #17, 18:
http://luminous-landscape.com/forum/index....showtopic=39210
« Last Edit: June 22, 2010, 11:06:20 AM by Czornyj » Logged

digitaldog
Sr. Member
****
Offline Offline

Posts: 8619



WWW
« Reply #19 on: June 22, 2010, 12:23:41 PM »
ReplyReply

The off the shelf EyeOne-Display will work. As will other Colorimeters not filtered for this kind of unit. The target white point you ask for, and what you get will likely be off. In tests I did, the differences amounted to about CCT 500K. Its not really a huge big deal because the values you ask for are not as important as getting a White Point that produces a visual match to the print next to the display. Yes, the mated colorimeter is preferable! But its not a game changer either.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Pages: [1] 2 3 ... 7 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad