AVS Forum banner

14801 - 14820 of 29375 Posts

·
Registered
Joined
·
107 Posts
Hi folks...


Can someone tell me if the R.Masciola 4k test patterns contain a full screen bright white pattern that I can use my light meter for to measure nits? OR any other suggestions? Thanks folks...
Yes, a 100% full screen white in his patterns. I use it to check my lamp levels.
 

·
Registered
Joined
·
479 Posts
Not really.

Ideally you would also create a colorimeter correction from a reference spectrometer first.

I uploaded mine and you can use it with your i1 Display Pro to make it read JVC LCoS projectors more accurately as the spectrometer does.

https://colorimetercorrections.displaycal.net/?get&type=ccss&manufacturer_id=JVC&display=JVC NX5&instrument=i1 DisplayPro, ColorMunki Display, Spyder4&html=1


The 3DLUT process is really only good if you are using madVR, so that you have somewhere to actually load the LUT. Or you could create a 3DLUT for a Lumagen as well.
OK, got it now, I was aware of 3D LUT use within photoshop as a seperate color adjustment layer, but was wondering how to use it with a projector.

Also having a Colormunki Photo spectro I will use it to correct my I1 D3 with DisplayCal,

Another question does this colorimeter correction mandatorily been done with my projector reading the screen or can I do it using my monitor ? (would be easier the latter way, but maybe not as accurate ?)
 

·
Premium Member
Joined
·
9,826 Posts
Not really.

Ideally you would also create a colorimeter correction from a reference spectrometer first.

I uploaded mine and you can use it with your i1 Display Pro to make it read JVC LCoS projectors more accurately as the spectrometer does.

https://colorimetercorrections.displaycal.net/?get&type=ccss&manufacturer_id=JVC&display=JVC NX5&instrument=i1 DisplayPro, ColorMunki Display, Spyder4&html=1


The 3DLUT process is really only good if you are using madVR, so that you have somewhere to actually load the LUT. Or you could create a 3DLUT for a Lumagen as well.
The i1display Pro is fairly accurate on LCOS, I wouldn't use a profile made on another i1pro2 with another 1display pro. You'd have no way to know if the i1display Pro is more accurate with or without the profile. It's fine to run a 3D LUT with the i1display Pro on its own.
 

·
Registered
Joined
·
2,805 Posts
Yes, a 100% full screen white in his patterns. I use it to check my lamp levels.
Thanks, r u able to tell me specifically what pattern that is? (name and under what section) I looked last night but couldn't find it...Thanks again...
 

·
Registered
Joined
·
1,788 Posts
OK, got it now, I was aware of 3D LUT use within photoshop as a seperate color adjustment layer, but was wondering how to use it with a projector.

Also having a Colormunki Photo spectro I will use it to correct my I1 D3 with DisplayCal,

Another question does this colorimeter correction mandatorily been done with my projector reading the screen or can I do it using my monitor ? (would be easier the latter way, but maybe not as accurate ?)
The whole point of correcting is to make your colorimeter read colors from your projector in the same way that your spectro reads the colors, because the spectro "should" read it correctly regardless of the type of display it is.

Colorimeter corrections are specific to the display being measured/calibrated so you must measure the projector with the spectro and feed that correction to the colorimeter.
 

·
Registered
Joined
·
479 Posts
I uploaded mine and you can use it with your i1 Display Pro to make it read JVC LCoS projectors more accurately as the spectrometer does.

https://colorimetercorrections.displaycal.net/?get&type=ccss&manufacturer_id=JVC&display=JVC NX5&instrument=i1 DisplayPro, ColorMunki Display, Spyder4&html=1
I just now realized that your uploaded file is a correction file ........ as it has been made with an I1 Pro 2 (much better than my Colormunki) can I use it as is ? Any chances that my I1 D3 has drifted from yours ?
 

·
Registered
Joined
·
1,788 Posts
The i1display Pro is fairly accurate on LCOS, I wouldn't use a profile made on another i1pro2 with another 1display pro. You'd have no way to know if the i1display Pro is more accurate with or without the profile. It's fine to run a 3D LUT with the i1display Pro on its own.
From what I am to understand, some colorimeters have their own internal spectro offset calibrations which were set individually per colorimeter in the factory when they were created.

This includes the Spyder4, Spyder5, SpyderX, X-Rite i1 Display Pro, and X-Rite Colormunki.

This means that they can all take a .CCSS spectro correction measured for a specific display type and apply the transformation based on their own individual internal spectro table, and should be more accurate than reading that display with no correction.

From the DisplayCAL documentation page:

Special note about the X-Rite i1 Display Pro, ColorMunki Display and Spyder4/5 colorimeters
These instruments greatly reduce the amount of work needed to match them to a display because they contain the spectral sensitivities of their filters in hardware, so only a spectrometer reading of the display is needed to create the correction (in contrast to matching other colorimeters to a display, which needs two readings: One with a spectrometer and one with the colorimeter).
That means anyone with a particular screen and a spectrometer can create a special Colorimeter Calibration Spectral Set (.ccss) file of that screen for use with those colorimeters, without needing to actually have access to the colorimeter itself.
 

·
Registered
Joined
·
1,788 Posts
I just now realized that your uploaded file is a correction file ........ as it has been made with an I1 Pro 2 (much better than my Colormunki) can I use it as is ? Any chances that my I1 D3 has drifted from yours ?
Your i1d3 may have drifted, but it shouldnt have much as they are known to last well over time.

What I uploaded was a .CCSS file, here is what DisplayCAL says about .CCSS files.

Special note about the X-Rite i1 Display Pro, ColorMunki Display and Spyder4/5 colorimeters
These instruments greatly reduce the amount of work needed to match them to a display because they contain the spectral sensitivities of their filters in hardware, so only a spectrometer reading of the display is needed to create the correction (in contrast to matching other colorimeters to a display, which needs two readings: One with a spectrometer and one with the colorimeter).
That means anyone with a particular screen and a spectrometer can create a special Colorimeter Calibration Spectral Set (.ccss) file of that screen for use with those colorimeters, without needing to actually have access to the colorimeter itself.


It means that when you apply my .CCSS file to your i1D3, it will factor in the internal hardware offsets of your specific i1D3 (which were set in the factory when it was built).

However, if you have your own spectro, you should just create your own .CCMX matrix correction.
 

·
Registered
Joined
·
479 Posts
The i1display Pro is fairly accurate on LCOS, I wouldn't use a profile made on another i1pro2 with another 1display pro. You'd have no way to know if the i1display Pro is more accurate with or without the profile. It's fine to run a 3D LUT with the i1display Pro on its own.
I missed your post while I was typing mine .... so it be ! Anyway I have no mean to use a 3D LUT now, no Lumagen and no madVR, maybe later on !
 

·
aka jfinnie
Joined
·
4,221 Posts
So I did a whole re-calibration of my NX5 from the start with a fresh autocal and fresh 3DLUT.

Here are my results (everything is calibrated and measured against DCI-P3):

JVC Autocal:


3DLUT:


Gamut:


Video version:
https://youtu.be/WQs05LHycaM

3DLUT was created in DisplayCAL with just 115 patches.

It's pretty nice, I now got 90% DCI-P3 coverage and 95.8% DCI-P3 volume.
Nice. The volume number isn't useful since it is just a comparison of the size of the volume, not whether it is in P3 or not. For it to be higher than the coverage it means there is some volume which is wider than P3.
Also note that the behaviour of this LUT is suboptimal - for the saturated greens it is introducing a hue shift. A more correct LUT wouldn't look so wide on this CIE chart.
 

·
Registered
Joined
·
1,788 Posts
Nice. The volume number isn't useful since it is just a comparison of the size of the volume, not whether it is in P3 or not. For it to be higher than the coverage it means there is some volume which is wider than P3.
Ah yeah that makes sense. Learn something new, thanks!
 

·
Registered
Joined
·
479 Posts
It means that when you apply my .CCSS file to your i1D3, it will factor in the internal hardware offsets of your specific i1D3 (which were set in the factory when it was built).

However, if you have your own spectro, you should just create your own .CCMX matrix correction.
Now I am completely confused, so far I thought that I had to read the display once with the spectro then again with the colorimeter and have DisplayCal compute the corrections.
Does that mean that the internal hardware of my I1D3 plus the ccss file from the spectro is enough ?
Is that true only when using DiplayCal or can it be used with HCFR too ?
 

·
aka jfinnie
Joined
·
4,221 Posts
Ah yeah that makes sense. Learn something new, thanks!
See the comment I edited in. Though the coverage looks better with the LUT you generated, in reality this chart from post-LUT is still telling you that for the region between attainable green primary and true P3 primary you have a large colour hue shift in order to maintain saturation. (The bent green points are representing green colours which will be shifted in their colour). There is an article about this and other calibration errors at the Light Illusion website - basically a more accurate LUT approach is to maintain correct hue and reduce the maximum saturation atttainable. Edit: In Lightspace the default LUT mode would generate a "correct" LUT (peak chroma). Not sure what you need to select in DisplayCAL to achieve the same, but there are lots of options for the rendering intent so I guess one might achieve something more accurate.

The following image is taken from here:
https://www.lightillusion.com/error.html


You can see here that the MacAdam diagram of elipses of just noticeable differences appear to show that you will likely notice the hue shift more than the lack of saturation in this area of the gamut.
From here: https://www.researchgate.net/figure/MacAdam-ellipses-plotted-on-the-CIE-1931-xy-chromaticity-diagram-9-The-ellipses-show_fig1_246546361
 

Attachments

·
Registered
Joined
·
1,788 Posts
Now I am completely confused, so far I thought that I had to read the display once with the spectro then again with the colorimeter and have DisplayCal compute the corrections.
That creates a .CCMX file, called a matrix correction. It is the most accurate correction.

Does that mean that the internal hardware of my I1D3 plus the ccss file from the spectro is enough ?
Is that true only when using DiplayCal or can it be used with HCFR too ?
A .CCSS from a reading off the display with a spectro is enough for a Spyder4/5 and i1D3/Colormunki and should be better than no correction. But a .CCMX can be better if you have a spectro to make your own.

For instance, the .CCSS assumes that your i1D3 is still as perfect as it was when it left the factory. If your i1D3 has shifted at all, then it's internal spectro readings wont match its current filter performance, and thus a .CCMX where you measure both a spectro and your i1D3 will be more accurate.
 

·
Registered
Joined
·
479 Posts
That creates a .CCMX file, called a matrix correction. It is the most accurate correction.



A .CCSS from a reading off the display with a spectro is enough for a Spyder4/5 and i1D3/Colormunki and should be better than no correction. But a .CCMX can be better if you have a spectro to make your own.

For instance, the .CCSS assumes that your i1D3 is still as perfect as it was when it left the factory. If your i1D3 has shifted at all, then it's internal spectro readings wont match its current filter performance, and thus a .CCMX where you measure both a spectro and your i1D3 will be more accurate.
OK, now I am back again ....... what a sweat :smile:
 

·
Premium Member
Joined
·
9,826 Posts
From what I am to understand, some colorimeters have their own internal spectro offset calibrations which were set individually per colorimeter in the factory when they were created.

This includes the Spyder4, Spyder5, SpyderX, X-Rite i1 Display Pro, and X-Rite Colormunki.

This means that they can all take a .CCSS spectro correction measured for a specific display type and apply the transformation based on their own individual internal spectro table, and should be more accurate than reading that display with no correction.

From the DisplayCAL documentation page:

Special note about the X-Rite i1 Display Pro, ColorMunki Display and Spyder4/5 colorimeters
These instruments greatly reduce the amount of work needed to match them to a display because they contain the spectral sensitivities of their filters in hardware, so only a spectrometer reading of the display is needed to create the correction (in contrast to matching other colorimeters to a display, which needs two readings: One with a spectrometer and one with the colorimeter).
That means anyone with a particular screen and a spectrometer can create a special Colorimeter Calibration Spectral Set (.ccss) file of that screen for use with those colorimeters, without needing to actually have access to the colorimeter itself.
Sure, you can do this, I am only saying it is not recommended, simply because the i1pro2 isn't a reference spectro. There is no guarantee that your i1pro2 is in any way more accurate than the other user's i1displaypro. I tested a few i1displaypro on LCOS and I certainly wouldn't have been able to know, without a reference meter, which of these or my i1pro2 was closer to reference.

Anyway, no one will die if using a not so great profile, I'm only saying that it's not a great idea to share an i1pro2 profile for an i1display pro on LCOS. A spyder, sure, because chances are the i1pro2 will be more accurate, but an i1displaypro, not in my opinion, at least based on the 2-3 units I tested and various reviews showing how close they were both between different samples (unlike the spyders) and to the i1pro/pro2.

Personally I'd use the factory calibration specific to this meter, or a profile made also for this meter, with a reference spectro.
 

·
Registered
Joined
·
68 Posts
Hey I have the NX7 and have been considering upgrading my 110" 16:9 Carada for a 125" 2:35 Cina Neve. I havn't had the opportunity to view a Cima Neve in person but lots of people compliment it highly. If possible, I'd love to take you up on that sample.

BTW, just a general update, I'm a little over 50 hours on my NX7 and its working fantastic. NO issues. I always use the color filter and have yet to see the "yellowing" being referred to in some of the posts. Perhaps I have not watched enough content but it has not shown up once. I still have not calibrated mine or tweak many of the settings at all. Looking to do a Gamma calibration with a Spyder 5 soon, likely then get a real calibration done after the 100h mark....although I am considering the new screen and am likely to wait until we decide on that.
I also have have Carada screen I was adding backlighting to the back of the frame at night and my flashlight hit the screen and it was amazing how much light went thru it. I added blackout material to the back of the screen and it really made a difference how my NX5 looked. Just a thought before you buy a new screen.
 

·
aka jfinnie
Joined
·
4,221 Posts
Sure, you can do this, I am only saying it is not recommended, simply because the i1pro2 isn't a reference spectro. There is no guarantee that your i1pro2 is in any way more accurate than the other user's i1displaypro. I tested a few i1displaypro on LCOS and I certainly wouldn't have been able to know, without a reference meter, which of these or my i1pro2 was closer to reference.

Anyway, no one will die if using a not so great profile, I'm only saying that it's not a great idea to share an i1pro2 profile for an i1display pro on LCOS. A spyder, sure, because chances are the i1pro2 will be more accurate, but an i1displaypro, not in my opinion, at least based on the 2-3 units I tested and various reviews showing how close they were both between different samples (unlike the spyders) and to the i1pro/pro2.

Personally I'd use the factory calibration specific to this meter, or a profile made also for this meter, with a reference spectro.
I agree.

The only multiple meter comparison I've seen is the old Dry Creek one, and that showed the inter-device agreement to be quite a bit worse for the typical i1pro than i1d3 for at least LCD displays , so you really have to question what you gain from using a shared profile, given that pretty much all the shared profiles are generated with i1pro/2 units!

Maybe if folk were generating them with calibrated higher end spectros they may be considered more useful, but given that there are frequently differences observed in the gamut coverage of these projectors even that seems a stretch.
 

·
Registered
Joined
·
479 Posts
I have another silly question, and it probably will not be the last one :eek:

To do an autocal with a spyder 5, given that the spyder is facing the projector's lens, that I am not in a dedicated bat cave and that it is not easy for me to darken the room enough during day time while running the autocal and last that I take it the spyder should only measure the small light beam between its sensor and the projector's lens, could it be possibly to use a piece of tubing (let's say 3 or 4 inches internal diameter, 1 foot long) sort of hood in front of the sensor to isolate it from any external stray light (suggestion with another tripod) ?
Of course once the spyder has been positionned correctly and I could even affix black velvet inside the tube, I've read that someone uses the sensor's reflection to align it properly towards the projector lens.
 
14801 - 14820 of 29375 Posts
Top