AVS Forum banner

4821 - 4840 of 7951 Posts

·
Registered
Joined
·
483 Posts
We said we would try to confirm your reported issue when I got back in the office next week.

Nobody is ignoring you, The last email I sent said we would test it when I got back next week.

Yes it is true that the Video Forge Pro and Murideo Only receive 8 bit triplets from CalMAN, where as Virtual Forge and the Accupel receive 10bit triplets. We did extensive testing of RGB 8bit at up to 4K60p and it is bit perfect. We tested a lot of different resolution and combinations and frame rates. I plan to do the same for YCBCR on Monday.

Also measuring with a meter is not the correct way to verify a pattern generator.

We can refund you for your VFP if you would like.
Tyler,

Has it been verified that HDR BT2020 from the Murideo SixG in YCbCr 10bit is wrong? Does it mean we can only use RGB for HDR ? Pls help as I asked the same in Calman forum but there was no answer. I also have a licensing issue that needs help

Sent from my HUAWEI NXT-L29 using Tapatalk
 

·
Registered
Joined
·
5,163 Posts
sorry, u lost me. U say similar numbers but then u say u only used CM to verify... ? then how would u know the numbers are similar ?

my point was: the dE ITP numbers listed by LS are (currently) incorrect, meaning: when u open a display profile in LS and look at the dE ITP numbers for a given color patch --> as when u do validation of the current state of the display...

I'm not referring to "calibration", as LS does not use any dE formula to calculate the cLUT...

if u send me the LS display profile I can tell u the real dE ITP numbers
Hi Mike, pretty sure I understand and can work around,
See attachment.
 

Attachments

·
Registered
Joined
·
8,914 Posts
Hi Mike, pretty sure I understand and can work around,
See attachment.
SS, is that your profile measurements after 1D LUT AutoCAL using CalMAN?

You have huge clip to near black, @ your 5% Gray, about 3.0 Gamma. (and to your near reference white also). Which flashing bars of a Brightness pattern do you see after the AutoCAL?
 

·
Registered
Joined
·
8,914 Posts



The Student Paper Award recognizes the outstanding paper prepared and submitted by a Student Member.

The paper receiving the Student Paper Award will be published in the SMPTE Motion Imaging Journal. Catherine Meininger, a recent graduate of Rochester Institute of Technology (RIT) who is now a color scientist at Portrait Displays Inc., will receive an honorable mention for her paper “Determining Visibility Thresholds for Spatial and Spatiotemporal Chromatic Noise.”
 

·
Registered
Joined
·
9,208 Posts
in the words of the guy who just emailed you: bunch of bollox :):):)

the "correct value" is whatever Dolby says it is... don't like it or don't agree with it ? don't use it... simple as that. I'm not trying to tell you to use it. there are pitfalls to each and every single dE formula.
[...]
the scalars are part of dE ITP, as the formula is super simple (since the encoding space is the actual more complex transform)... "technically speaking" there is only one scalar: 720. That is what Dolby uses, hence there is no scalar, the multiplier is just part of the formula.

The 240 scalar was just offered as a relative comparison to dE2000... it is not perfect, and I would not use it.
Thankyou very much @Iron Mike! I now have my answer to the question I posted yesterday. :)
 

·
Registered
Joined
·
516 Posts
New to CalMAN

Hello all,

I just purchased CalMAN Home Enthusiast version yesterday. I would like to calibrate my JVC X990R Projector and some other LED 4K tvs in the house but was trying to learn how to use software for projector. I want to calibrate my SDR and HDR settings so I emailed CalMAN this:

Original message
Ray wrote:

I just purchased software today. I am planning on calibrating my JVC HDR projector this weekend. I want to calibrate a couple different settings. SDR and HDR rec 709 and BT2020. What workflows should I be using? I am new to CalMAN and looking for help.

Here is reply:

Mack C. replied:
Hi Ray,
As we spoke about previously, I would avoid attempting to calibrate a projector in HDR mode until we release our video on the new gamma formula.
For an SDR calibration, use the Home Advanced Workflow. That should take you thru the common steps and adjustments needed for a home theater TV or projector.
Mackenzie Collins


So at this point you can't use for HDR? I am wanting to learn how use software and where to start. I haven't use CalMAN in the past and haven't TRUELY calibrated a display before so any help would be great!
 

·
Registered
Joined
·
4,297 Posts
Discussion Starter #4,827 (Edited)
Hello all,

I just purchased CalMAN Home Enthusiast version yesterday. I would like to calibrate my JVC X990R Projector and some other LED 4K tvs in the house but was trying to learn how to use software for projector. I want to calibrate my SDR and HDR settings so I emailed CalMAN this:

Original message
Ray wrote:

I just purchased software today. I am planning on calibrating my JVC HDR projector this weekend. I want to calibrate a couple different settings. SDR and HDR rec 709 and BT2020. What workflows should I be using? I am new to CalMAN and looking for help.

Here is reply:

Mack C. replied:
Hi Ray,
As we spoke about previously, I would avoid attempting to calibrate a projector in HDR mode until we release our video on the new gamma formula.
For an SDR calibration, use the Home Advanced Workflow. That should take you thru the common steps and adjustments needed for a home theater TV or projector.
Mackenzie Collins


So at this point you can't use for HDR? I am wanting to learn how use software and where to start. I haven't use CalMAN in the past and haven't TRUELY calibrated a display before so any help would be great!
Here is the tutorial video for the BT.2390 EOTF use in CalMAN for "HDR" projectors

https://drive.google.com/file/d/1DlUM6byDlP0LhJbUZYLsQRBtp8NFr1PL/view?usp=sharing

I would also suggest asking for tips in the owners thread for your projector. Most of the guys that are calibrating their PJs post in the owners threads, not in the calibration forum.
 

·
Registered
Joined
·
5,163 Posts
SS, is that your profile measurements after 1D LUT AutoCAL using CalMAN?

You have huge clip to near black, @ your 5% Gray, about 3.0 Gamma. (and to your near reference white also). Which flashing bars of a Brightness pattern do you see after the AutoCAL?
No that's a profile done by LS. Before I posted I deleted the 1D that LS now offers. The 1D helped but raised CMS errors. Anyway the outcome before adding 1D there was a big clip near black, had to lower brightness by 2 points.
I did two profiles using LS and eecolor before determining that using LG's internal storage and processing of 3DLUT was the best way to go. eecolor outputs very low Mhz.

ss
 

·
Registered
Joined
·
8,914 Posts
No that's a profile done by LS. Before I posted I deleted the 1D that LS now offers. The 1D helped but raised CMS errors. Anyway the outcome before adding 1D there was a big clip near black, had to lower brightness by 2 points.
I did two profiles using LS and eecolor before determining that using LG's internal storage and processing of 3DLUT was the best way to go. eecolor outputs very low Mhz.

ss
The profile you have uploaded is what your measured, so the TV's un-calibrated status, but you have already clip of near black, you have to resolved this before you start cube measurements.

What do you mean that eeColor outputs very low MHz?
 

·
Registered
Joined
·
5,163 Posts
The profile you have uploaded is what your measured, so the TV's un-calibrated status, but you have already clip of near black, you have to resolved this before you start cube measurements.

What do you mean that eeColor outputs very low MHz?
Yes you are correct, but I wanted to see what LS new 1D had to offer as far as any correction. The 1D viewer in LS looked pretty good, but as you see there was a clip with black.

Yes eecolor box.

ss
 

·
Registered
Joined
·
8,914 Posts
Yes you are correct, but I wanted to see what LS new 1D had to offer as far as any correction. The 1D viewer in LS looked pretty good, but as you see there was a clip with black.

Yes eecolor box.

ss
The file you have uploaded is not post-calibration verification, its the measurement file (so native gamut with pre-calibrated by you grayscale) of 21-Point Cube (9261 measurements) + additional 92 White drift measurements.

At this stage, your data file you have uploaded its not showing to any kind of post-calibration results, you see only what is the display status as you measured it.



Since you have that large problem already, its something you have to resolve and re-measure.

Your problem to near black is probably responsible the pre-calibration you have done (the parametric one which was not required, only 100% White was required, for that reason your RGB separation graph is not looking so good, because internal TV processing is altering it) and the problem of your near white probably is from patch generator setting mismatch? (find out taking quick profile of primary only measurements.)

Before starting large 3D LUT profilings its better to verify and find the best pre-calibration settings.

BTW its better idea to discus all that LightSpace related stuff to the LightSpace thread, becasue you post there for LightSpace you get reply there. Lets keep that thread clean, talking for CalMAN only.

If you want to talk about only CalMAN, post there, if you want to talk about LightSpace post to LightSpace thread.

If you want to post for both software (or eeColor with LG) at the same post, use the LG 2018 calibration thread.

Also explain what you mean that eeColor is low MHz? It's 1080p60 input/output box (12-bit output per colorchannel) and its has zero millisecond added latency when you pass signal thru eeColor.

SDR signal don't need something else, the TV will do the upscale to its native resolution. The Internal LG 3D LUT table is 33-Point Cube (35937 Color Points) while eeColor's is 65-Point Cube (274625 Color Points), so you handling better your whole gamut when there strange anomalies to specific regions.

To comment about the new 1D LUT (you calling the Augement Data process) performance, you have to generated first the correction 3D LUT file, then to upload it to your eeColor, then with that 3D LUT as Active to eeColor to measure Quick Profile of Primary colors (to keep that verification file as 'the original' post-verification file)

Then to measure with that correction 3D LUT as active the 33-Point GrayScale, then do the process to combine the 2 profiling file, and then generate a new 3D LUT correction file....load that file to a different slot of eeColor and then measure with Quick Profile of Primary colors (to keep that verification file as the Aug. Data post verification file).

Then you compare the 2 verification files to see if that new feature of LightSpace works for your setup or not. This is the correct testing procedure. (post all results to Lightspace thread).
 

·
Registered
Joined
·
5,163 Posts
The file you have uploaded is not post-calibration verification, its the measurement file (so native gamut with pre-calibrated by you grayscale) of 21-Point Cube (9261 measurements) + additional 92 White drift measurements.

At this stage, your data file you have uploaded its not showing to any kind of post-calibration results, you see only what is the display status as you measured it.



Since you have that large problem already, its something you have to resolve and re-measure.

Your problem to near black is probably responsible the pre-calibration you have done (the parametric one which was not required, only 100% White was required, for that reason your RGB separation graph is not looking so good, because internal TV processing is altering it) and the problem of your near white probably is from patch generator setting mismatch? (find out taking quick profile of primary only measurements.)

Before starting large 3D LUT profilings its better to verify and find the best pre-calibration settings.

BTW its better idea to discus all that LightSpace related stuff to the LightSpace thread, becasue you post there for LightSpace you get reply there. Lets keep that thread clean, talking for CalMAN only.

If you want to talk about only CalMAN, post there, if you want to talk about LightSpace post to LightSpace thread.

If you want to post for both software (or eeColor with LG) at the same post, use the LG 2018 calibration thread.

Also explain what you mean that eeColor is low MHz? It's 1080p60 input/output box (12-bit output per colorchannel) and its has zero millisecond added latency when you pass signal thru eeColor.

SDR signal don't need something else, the TV will do the upscale to its native resolution. The Internal LG 3D LUT table is 33-Point Cube (35937 Color Points) while eeColor's is 65-Point Cube (274625 Color Points), so you handling better your whole gamut when there strange anomalies to specific regions.

To comment about the new 1D LUT (you calling the Augement Data process) performance, you have to generated first the correction 3D LUT file, then to upload it to your eeColor, then with that 3D LUT as Active to eeColor to measure Quick Profile of Primary colors (to keep that verification file as 'the original' post-verification file)

Then to measure with that correction 3D LUT as active the 33-Point GrayScale, then do the process to combine the 2 profiling file, and then generate a new 3D LUT correction file....load that file to a different slot of eeColor and then measure with Quick Profile of Primary colors (to keep that verification file as the Aug. Data post verification file).

Then you compare the 2 verification files to see if that new feature of LightSpace works for your setup or not. This is the correct testing procedure. (post all results to Lightspace thread).
Yes it is just the profile, you can generate any 3DLUT from it.

I really don't have much of a problem, it's just that the profile sucks.
My black level (brightness set at 50) read is about .0004, verarys slightly +/- 3 dE.

Its low measured by the Vertex.

Yes I know how Augment Data works.

What my main interest in comparing how eecolor work's compared to how using the internal processor of LG works as far as PQ and ease of how it works.

I didn't answer most of your question because as you asked, only Calman.

On a side note the updated Grayscale that Calman has in it's new beta works well.

My conclusion is that using the latest Calman beta and the LG to process and store the 1D/3DLUT, has the most benefit for me. Just to be clear, I'm not just talking about software I'm also talking about what hardware is used when not using LG to process and store the cLUT's. If using a external 3DLUT processor. Not only am I adding a external 3DLUT processor but a Vertex as a switcher and a Roku 4K for streaming.
1. Oppo 203 directly to the LG via HDMI highspeed cable. Total of one HDMI length is 6 feet.
2. Oppo 203>eecolor>Ruko>Vertex>LG. Total of four HDMI length 24 feet.

btw, can't prove it but I agree the "internal TV processing is altering it" when using a non direct software/hardware. I'm starting to think that because of how LG processes the video signal, that Calman and LG has a lag up on other calibration software and hardware used for storing, processing and what's in your video path.

ss
 
  • Like
Reactions: mrtickleuk

·
Registered
Joined
·
8,914 Posts
I really don't have much of a problem, it's just that the profile sucks.

Your measurement show the issues you have, they can't be hidden, you have to resolve them.

What my main interest in comparing how eecolor work's compared to how using the internal processor of LG works as far as PQ and ease of how it works.

My conclusion is that using the latest Calman beta and the LG to process and store the 1D/3DLUT, has the most benefit for me. Just to be clear, I'm not just talking about software I'm also talking about what hardware is used when not using LG to process and store the cLUT's. If using a external 3DLUT processor. Not only am I adding a external 3DLUT processor but a Vertex as a switcher and a Roku 4K for streaming.
1. Oppo 203 directly to the LG via HDMI highspeed cable. Total of one HDMI length is 6 feet.
2. Oppo 203>eecolor>Ruko>Vertex>LG. Total of four HDMI length 24 feet.

btw, can't prove it but I agree the "internal TV processing is altering it" when using a non direct software/hardware. I'm starting to think that because of how LG processes the video signal, that Calman and LG has a lag up on other calibration software and hardware used for storing, processing and what's in your video path.
Using internal LUT capability, you don't have backup of your 1D LUT, so generate a different 3D LUT correction using a different target gamma, you have to perform a competely new 1D/3D LUT.

So if you need 3 different gammas to have (2.4 / 2.2 / 2.35 for example....or different custom white points) you need to perform 3x 1D/3D LUT sets of measurements.

With external LUT, you are using one measurement data.... and from there you generate as many different 3D LUT you like. So you save much more time while always have you backup inside to external LUT box in case something happen to the TV.

About your connections, its simpler the solution for your setup.

Vertex has 2 inputs, you connect to input 1 the OPPO to input 2 the Ruko.

Vertex has 2 outputs, the top is HDCP 1.4 1080p, the bottom is HDCP 2.2 2160p.

You connect to top output of Vertex, the eeColor and then the TV (HDMI2)

You connect to the bottom output of Vertex directly the TV. (HDMI1)

You set the OPPO to source direct, so if you load a BD or 1080p file, it will output automatically 1080p, so the signal automatically can pass eeColor. (you will just select TVs HDMI2)

When you will watch HDR10/Dolby Vision from OPPO, it will output automatically HDR signal, so in any case you don't need to change any setting to OPPO (you will just select TVs HDMI1), just you will need to perform from Vertex the switching manually or via iOs/Android application the HD Fury has, connecting to Vertex via Bluetooth.

Think also that if your next TV can be a Samsung/Sony/Panasonic/Vizio....which will not have internal 3D LUT capabilities, so since you are familiar now with 3D LUT and your eyes will always pleasure from 3D LUT correction, you will still need an SDR 3D LUT Box for your next TV, so even if you will not use your eeColor now, you will need it to your next TV, becasue SDR signal will never change specs and it will be always the only way to match the SDR mastering specs (100% REC.709 coverage all TV's can display right now, 100 nits, gamma 2.4) and no ABL, less drift etc.
 

·
Registered
Joined
·
9,208 Posts
Think also that if your next TV can be a Samsung/Sony/Panasonic/Vizio....which will not have internal 3D LUT capabilities, so since you are familiar now with 3D LUT and your eyes will always pleasure from 3D LUT correction, you will still need an SDR 3D LUT Box for your next TV,
Hmm, that's very pessimistic. I take a different view - either he'll buy another LG (I certainly would without hesitation) or, more and more manufacturers will have to match LG and add customer-accessible built-in LUTs. So it's most likely that his next TV will have a built-in LUT again.

But even then that's only for SDR content. By the time he buys another TV there should be even more HDR content, and less interest in SDR. The speed things are moving is astonishing and I would not have predicted HDR would take hold so quickly.
 

·
Registered
Joined
·
1,081 Posts
But even then that's only for SDR content. By the time he buys another TV there should be even more HDR content, and less interest in SDR. The speed things are moving is astonishing and I would not have predicted HDR would take hold so quickly.
I think one of the main points of Ted's last post was that a lot of your titles of your SDR collection (anything up to 1080p) won't be re-released as 4K (at least not re-scanned at 4K or higher, so not a re-release that is needed) and won't be re-mastered in HDR... just think Criterion Collection alone, almost 1,000 titles...

all those titles you can watch forever with a perfectly calibrated SDR chain...

with 4K, HDR, Rec 2020 etc there will be lots of changes and hassle until we're at that state... u can see how much time people waste here trying to somewhat cal HDR, with bad results...
 

·
Registered
Joined
·
152 Posts
Think also that if your next TV can be a Samsung/Sony/Panasonic/Vizio....which will not have internal 3D LUT capabilities, so since you are familiar now with 3D LUT and your eyes will always pleasure from 3D LUT correction, you will still need an SDR 3D LUT Box for your next TV,
Hmm, that's very pessimistic. I take a different view - either he'll buy another LG (I certainly would without hesitation) or, more and more manufacturers will have to match LG and add customer-accessible built-in LUTs. So it's most likely that his next TV will have a built-in LUT again.

But even then that's only for SDR content. By the time he buys another TV there should be even more HDR content, and less interest in SDR. The speed things are moving is astonishing and I would not have predicted HDR would take hold so quickly.
I wouldn't have predicted that either and I completely agree. Also, LUT boxes are not a solution for those who use the inbuilt streaming apps.
 
  • Like
Reactions: mrtickleuk

·
Registered
Joined
·
10,531 Posts
In reality, what % of viewing is even done with HDR ?
Myself, I mostly watch satellite SDR content. About 2 % is HDR viewing. So Masicola's HDR patterns are more then enough for my purposes.
How soon will broadcasters convert to 4K? Man, most of them aren't even pushing 1080P yet!
 

·
Registered
Joined
·
629 Posts
In reality, what % of viewing is even done with HDR ?
Myself, I mostly watch satellite SDR content. About 2 % is HDR viewing. So Masicola's HDR patterns are more then enough for my purposes.
How soon will broadcasters convert to 4K? Man, most of them aren't even pushing 1080P yet!
Well, I'm watching lots of DV/HDR content. And almost 80% of the new Netflix movies are DV.
Tonight I've watched Outlaw King and for the first time I really enjoyed a DV/HDR movie, beautiful.
 
  • Like
Reactions: jrref

·
Registered
Joined
·
9,208 Posts
I think one of the main points of Ted's last post was that a lot of your titles of your SDR collection (anything up to 1080p) won't be re-released as 4K (at least not re-scanned at 4K or higher, so not a re-release that is needed) and won't be re-mastered in HDR... just think Criterion Collection alone, almost 1,000 titles...
:confused: Do you think I'm a lottery winner? I have ONE Criterion title, not a thousand, in my SDR collection ;) ! https://www.blu-ray.com/dvd/RoboCop-DVD/3692/ ;):D *

Yes, there will be lots of SDR content. The vast majority of my viewing is of course SDR television programmes, not movies. So one of the main points of my last post was that it's very highly likely his next TV will also have a built-in LUT like his LG :)

*edited to add: I know what you meant really
 

·
Registered
Joined
·
1,081 Posts
:confused: Do you think I'm a lottery winner? I have ONE Criterion title, not a thousand, in my SDR collection ;) ! https://www.blu-ray.com/dvd/RoboCop-DVD/3692/ ;):D

Yes, there will be lots of SDR content. So one of the main points of my last post was that it's very highly likely his next TV will also have a built-in LUT like his LG :)
LOOOOL hey, that's a good flick ! Verhoeven, Weller, great music by Basil Poledouris...

yeah, these TVs may have LUT storage that does not mean in any way shape or form, they will perform anything close to what the ee does (or whatever the home SDR setup is)...
 
4821 - 4840 of 7951 Posts
Top