Ad
Ad
Ad
Pages: [1] 2 3 »   Bottom of Page
Print
Author Topic: Seems like you can make a really excellent sRGB mode on NEC PA  (Read 12617 times)
WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« on: July 06, 2010, 02:32:14 AM »
ReplyReply

The default factory mode has the primary locations under 3dE (as measured by NEC puck in SVII). (I got R .639,.325,25.11; G .301,.602,86.64; B .152,.063,9.25; W .307,.324,119.6; only the red y and blue luminance are really much off at all perhaps they are adapted to the cooler white point?)
The default tone response curve seems to be very close to the sRGB standard.
The default WB in all the factory presets appears to be somewhat cool though at (as measured by the NEC puck again) .307,.324 instead of .3127,.3290.
Aside from the WB the factory preset mode appears to be excellent.
Not sure how much of the error comes from the factory preset and how much from the NEC puck.

Anyway you can also use Multiprofiler combined with SVII to take measurements to program a custom mode into Multiprofiler that uses the sRGB tone curve, doing that I got the measured WB just about right on and the R,G,B primaries: R 0.3dE  G 0.7dE B1.6dE (these include not just xy but also Y luminance error as well). Of course the NEC puck might have some variance but anyway it seems you can program it to have a near perfect sRGB mode (barring your probe variance). And this is all entirely internal, so it works even with non-managed stuff.

You can also use a gamma 2.2 (it measures just about a straight line along 2.2) instead of sRGB to make a TV/disc/game mode.

I didn't measure grayscale so I'm trusting the dE are all reasonable low for it, I don't see obvious signs of troubling coloration.

Few actual sRGB monitors do better.

(I haven't found a way to get the HDTV standard tone response curve along with calibrated WB at the same time though, I trust they will offer that curve as a selectable option for full calibration with both utilities in the future. Although it's not a huge loss in that very, very, very few sets ever get calibrated that way, almost nobody has a monitor or HDTV set that way (aside from perhaps a few calibrated to that standard using Calman).)


EDIT: I removed the stuff about compensation since the number were based on too few spots measured
« Last Edit: July 08, 2010, 03:22:38 AM by LarryBaum » Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6818


WWW
« Reply #1 on: July 06, 2010, 08:23:45 AM »
ReplyReply

What exact model display are you talking about?
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
rickhatCHROMiX
Newbie
*
Offline Offline

Posts: 21


« Reply #2 on: July 06, 2010, 12:30:08 PM »
ReplyReply

There are currently two PA models that would be able to have this feature if used with SpectraView software: PA241W and PA271W.

This is made possible via MultiProfiler and the benefits of a 3D LUT memory on-board a PA model. It allows you to use a simulation profile in the monitor, INDEPENDENT of a computer (consider that for a moment). It can be an sRGB, aRGB, Rec-709, etc. color space profile, or a custom RGB monitor profile. It can even be a printer profile (now, really consider the possibilities of this!)

This exciting process potentially allows very accurate emulations on screen as compared to NEC's OSD presets. It could give new meaning to application-based 'soft-proofing' as well. It remains to be seen how well 'all' transforms will occur, but certainly more testing and feedback from users would be valuable and welcomed.

Rick Hatmaker
CHROMiX.com


Quote from: Mark D Segal
What exact model display are you talking about?
« Last Edit: July 06, 2010, 12:32:34 PM by rickhatCHROMiX » Logged
Czornyj
Sr. Member
****
Offline Offline

Posts: 1370



WWW
« Reply #3 on: July 06, 2010, 12:30:51 PM »
ReplyReply

Quote from: LarryBaum
level 0 does look a little greener and dimmer to the left and a little redder and brighter to the right (not talking glowing red or green though at all like some horror shots you see of some HP/Dell)

Mura is noticable in all wide gamut CCFL/RGB LED displays, that's why all the best have electronic uniformity compensation. Eizo CG doesn't even have the possibility to turn it off.

As for the sRGB mode - I belive that in fact it may be closer to sRGB target when calibrated via Multiprofiler, I wouldn't consider SVII puck as a source of reference measurement values.

Quote from: rickhatCHROMiX
This exciting process potentially allows very accurate emulations on screen as compared to NEC's OSD presets. It could give new meaning to application-based 'soft-proofing' as well. It remains to be seen how well 'all' transforms will occur, but certainly more testing and feedback from users would be valuable and welcomed.
Exactly - I used my Epson 7880 profiles as calibration targets in Multiprofiler, It worked really cool.
« Last Edit: July 06, 2010, 12:35:22 PM by Czornyj » Logged

WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« Reply #4 on: July 06, 2010, 01:26:20 PM »
ReplyReply

Quote from: Mark D Segal
What exact model display are you talking about?

PA241W
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6818


WWW
« Reply #5 on: July 06, 2010, 01:43:28 PM »
ReplyReply

Quote from: rickhatCHROMiX
There are currently two PA models that would be able to have this feature if used with SpectraView software: PA241W and PA271W.

This is made possible via MultiProfiler and the benefits of a 3D LUT memory on-board a PA model. It allows you to use a simulation profile in the monitor, INDEPENDENT of a computer (consider that for a moment). It can be an sRGB, aRGB, Rec-709, etc. color space profile, or a custom RGB monitor profile. It can even be a printer profile (now, really consider the possibilities of this!)

This exciting process potentially allows very accurate emulations on screen as compared to NEC's OSD presets. It could give new meaning to application-based 'soft-proofing' as well. It remains to be seen how well 'all' transforms will occur, but certainly more testing and feedback from users would be valuable and welcomed.

Rick Hatmaker
CHROMiX.com

This is an interesting post Rick. I know there are two PA models, why I wanted to know which Larry was speaking of, which he has clarified. In any event, both of them are supposed to be able to represent most of ARGB(98), which made me wonder about the emphasis on sRGB

Of more interest to me is the idea of direct softproofing in the display LUT. Could you elaborate a bit on how one would go about implementing this, and whether you have experimented with it on one of the NEC-PA models?
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
digitaldog
Sr. Member
****
Offline Offline

Posts: 8592



WWW
« Reply #6 on: July 06, 2010, 02:04:54 PM »
ReplyReply

Quote from: Mark D Segal
In any event, both of them are supposed to be able to represent most of ARGB(98), which made me wonder about the emphasis on sRGB

Outside ICC aware applications, its real useful to have a fairly accurate sRGB emulation on a wide gamut display. The new NEC’s appear to take this emulation to a new level (along with soft proofing using actual print profiles).
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
digitaldog
Sr. Member
****
Offline Offline

Posts: 8592



WWW
« Reply #7 on: July 06, 2010, 02:06:48 PM »
ReplyReply

Quote from: rickhatCHROMiX
It can even be a printer profile (now, really consider the possibilities of this!)

The only question I had when playing with this feature is why it appears so vastly different from the soft proof produced in Photoshop (either using the simulate check boxes or not). Which one is “right”?
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6818


WWW
« Reply #8 on: July 06, 2010, 02:08:50 PM »
ReplyReply

Quote from: digitaldog
The only question I had when playing with this feature is why it appears so vastly different from the soft proof produced in Photoshop (either using the simulate check boxes or not). Which one is “right”?
Have you tried making a print and comparing it with both screen emulations to see which is closer? I'm interested in this as I'm contemplating purchase of a NEC-271
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
digitaldog
Sr. Member
****
Offline Offline

Posts: 8592



WWW
« Reply #9 on: July 06, 2010, 02:14:16 PM »
ReplyReply

Quote from: Mark D Segal
Have you tried making a print and comparing it with both screen emulations to see which is closer? I'm interested in this as I'm contemplating purchase of a NEC-271

My methods for setting calibration targets is to enter said values that produce a visual match. The two proofs are so different that its an Apples to Oranges comparison and I don’t know which targets are the correct values when I see such a disconnect between the two soft proof methods.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6818


WWW
« Reply #10 on: July 06, 2010, 02:39:59 PM »
ReplyReply

Quote from: digitaldog
My methods for setting calibration targets is to enter said values that produce a visual match. The two proofs are so different that its an Apples to Oranges comparison and I don’t know which targets are the correct values when I see such a disconnect between the two soft proof methods.

Andrew, could you please unpack your first statement, as I don't understand what you are doing. That may be my problem, but there it is. When I think of setting calibration targets for profiling a display, I think of the usual things: selecting White Point, Gamma and Luminosity, and letting the puck measure the patches the software generates on the display in order to create a profile with those calibration parameters. Did you mean something else?

As well, I'm hoping Rick provides more detail on how he sets up the printer profile in the display to create a one-stop soft-proof.

Now, exactly because you find the two approaches generating such different results, and because one important end-product of this color management is to get predictable and reasonably close matching between print and display, is why I asked whether you went as far as printing to see which is more valid. For example, and not even making a print, suppose you loaded the Atkinson printer test image on the display and looked at it with both profiles. We both probably know in-spades what that test image should look like, so would this not give a clue about which display profiling soution is closer to the desired performance?
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
digitaldog
Sr. Member
****
Offline Offline

Posts: 8592



WWW
« Reply #11 on: July 06, 2010, 02:44:20 PM »
ReplyReply

Quote from: Mark D Segal
When I think of setting calibration targets for profiling a display, I think of the usual things: selecting White Point, Gamma and Luminosity, and letting the puck measure the patches the software generates on the display in order to create a profile with those calibration parameters.

And those settings are based on results that produce a visual match to my viewing booth. The numbers themselves are otherwise meaningless. With my viewing booth, 150cd/m2 is the right answer (for me). For you, it may be a totally different value. When I setup a soft proof in Photoshop, hit F and Tab to be in full screen mode, view the print and the display and get a match, that’s it. If not, I have to alter the calibration target values (or maybe if possible, adjust the brightness of the booth). The settings used to calibrate the display are altered for that result. The results I see doing this for a match using Photoshop don’t produce a match and look vastly different using the NEC soft proofing. I don’t know why they look different.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6818


WWW
« Reply #12 on: July 06, 2010, 04:32:02 PM »
ReplyReply

Quote from: digitaldog
And those settings are based on results that produce a visual match to my viewing booth. The numbers themselves are otherwise meaningless. With my viewing booth, 150cd/m2 is the right answer (for me). For you, it may be a totally different value. When I setup a soft proof in Photoshop, hit F and Tab to be in full screen mode, view the print and the display and get a match, that’s it. If not, I have to alter the calibration target values (or maybe if possible, adjust the brightness of the booth). The settings used to calibrate the display are altered for that result. The results I see doing this for a match using Photoshop don’t produce a match and look vastly different using the NEC soft proofing. I don’t know why they look different.

Yes, I'm with you for most of this but let me try for more clarity from the penultimate sentence. Am I reading you correctly that using this display you don't get a match with this technique in BOTH set-ups: (i)  softproofing in Photoshop using a normal monitor profile for the display and the usual softproof from the Photoshop View menu; and (ii) using a printer profile in the display (instead of a normal display profile) as suggested by this new technique. If that's the case there is something to ponder before buying this display, because it would indicate that it can't be colour-managed properly, so I'm keen to make sure I understand this correctly.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
rickhatCHROMiX
Newbie
*
Offline Offline

Posts: 21


« Reply #13 on: July 06, 2010, 05:33:57 PM »
ReplyReply

Mark:

Initial observations: There may be many benefactors of this type of technology usage. Film & video users for example, can load up the simulation profile (or color space profile) in the monitor LUT. Many video related applications do not directly interface with profile data or necessarily with standard video card call outs and LUTs. This could be a real benefit for viewing corrected motion images correctly at monitor level. Linux/Unix users have long complained about interaction with profile data, DDC and standard video protocols that are mostly oriented to Mac and Windows drivers. This could give them corrected viewing conditions, as long as they could find a Mac or PC to load up the simulation profile. Field usages where there is no computer dependency, etc.

We have done some initial testing of 3D LUT usage, but we are far from reaching any publishable conclusions. None-the-less, it does present some exciting possibilities. I've been meaning to have a conversation soon with Will Hollingworth at NEC further about this technology. He may be able to shed some light here as well on NEC's technology or potential capabilities.....

I'm sure NEC has made or used code that can make and facilitate the transforms described above. This technology in not that uncommon (CHROMiX made a piggy-back application to ColorThink Pro called ColorCast years ago that does something essentially similar). DeviceLink profiles and abstract profiles are other derivative technologies that merge or transform source and simulation data. Assumptions are made of course under the hood, and it doesn't always work for every condition. So it would be interesting to hear it from NEC, and of course from field testers, before any proven methods could be adopted.

Rick Hatmaker
CHROMiX



Quote from: Mark D Segal
This is an interesting post Rick. I know there are two PA models, why I wanted to know which Larry was speaking of, which he has clarified. In any event, both of them are supposed to be able to represent most of ARGB(98), which made me wonder about the emphasis on sRGB

Of more interest to me is the idea of direct softproofing in the display LUT. Could you elaborate a bit on how one would go about implementing this, and whether you have experimented with it on one of the NEC-PA models?
Logged
WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« Reply #14 on: July 07, 2010, 12:39:39 AM »
ReplyReply

Quote from: Czornyj
that's why all the best have electronic uniformity compensation. Eizo CG doesn't even have the possibility to turn it off.

yes,it is nice that they give you a choice of 5-6 different settings, you can max contrast when uniformity is not quite as key or crank it all the way if that matters more than anything (and sometimes for print profiling you still are left with a lot more CR than any print just doing that alone so it's not a problem there at all)
Logged
Rhossydd
Sr. Member
****
Online Online

Posts: 1888


WWW
« Reply #15 on: July 07, 2010, 01:22:26 AM »
ReplyReply

Could someone please clarify the situation with these NEC PA series monitors for me ?
The PA series are supplied with Multiprofiler software that uses a "Spectraview engine" according to NEC's webs site ( http://www.necdisplay.com/multiprofiler/ ).

As yet they don't seem to be offered as a "Spectraview" variant. (ie the the better old Multisyncs were offered as 'Spectraview' monitors with selected units, hoods and 'Spectraview(aka Basicolor) calibration and profiling software).

Is it planned to offer to 'Spectraview' model based on the PA series monitors or are they equal to the older series already ?

Can the supplied Multiprofiler software match the old Spectraview software with respect to controlling the hardware of the monitor, as seems to be implied in the link above ?

Anyone any idea if the previous global differences between NEC monitors continue with this series ? (In the past US models of the Spectraview weren't supplied with the software and had to be bought in addition to the monitor as I understand it). Will European versions have the Multiproifler software included ?

Edit: As it's possible to freely download the Multiprofilier software direct from NEC's site without problems, I'd guess it's a global supply.

Thanks

Paul
In the UK
« Last Edit: July 07, 2010, 01:27:03 AM by Rhossydd » Logged
rickhatCHROMiX
Newbie
*
Offline Offline

Posts: 21


« Reply #16 on: July 07, 2010, 10:10:46 AM »
ReplyReply

Paul:

Yes, the SV versions of the PA series are shipping, although at lower quantity and frequency than non-SV versions, so backorders have been occurring for short periods.

I wouldn't know about the non-USA supply issues though.

Rick Hatmaker
CHROMiX
Logged
degrub
Guest
« Reply #17 on: July 07, 2010, 11:27:26 AM »
ReplyReply

There is at least one thread here that talks about the difference between the European and the US versions of the PA series. My recollection is that there is a special European version that has the SV kit tied to the LCD in software and has a hand selected LCD panel.

Frank Worley
Logged
Rhossydd
Sr. Member
****
Online Online

Posts: 1888


WWW
« Reply #18 on: July 07, 2010, 02:30:59 PM »
ReplyReply

Quote from: rickhatCHROMiX
Paul:
Yes, the SV versions of the PA series are shipping,...
I wouldn't know about the non-USA supply issues though.
Thanks Rick. No mention of a 27" Spectraview model on the UK NEC site yet.

Can you say what advantage the Spectraview software offers over Multiprofiler ?
I can live without specially selected panels and a hood, but I'd like to know if the Sv software will be worth the extra cash.
Logged
WombatHorror
Sr. Member
****
Offline Offline

Posts: 299


« Reply #19 on: July 08, 2010, 03:25:26 AM »
ReplyReply

Quote from: Rhossydd
Thanks Rick. No mention of a 27" Spectraview model on the UK NEC site yet.

Can you say what advantage the Spectraview software offers over Multiprofiler ?
I can live without specially selected panels and a hood, but I'd like to know if the Sv software will be worth the extra cash.

I can't say since the European SV software is said to 100% different from the US SV software.
Only the multiprofiler software is the same.
Logged
Pages: [1] 2 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad