HCFR - Open source projector and display calibration software - Page 370 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 1407Likes
Reply
 
Thread Tools
post #11071 of 11794 Old 11-16-2018, 01:44 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by inonefly View Post
I installed madVR and the readings from GDI and madVR patterns are slightly different. I think I got the madVR settings right?
Sorry, I'm not familiar with madVR. However, you still need to set the graphics card to 0-255.

Quote:
And as expected, the i1D3 and Spyder5 even more differences:
That's not unexpected.

Quote:
Do I really want to compare my spyder 5 readings to i1D3 to determine if it's accurate or not? I don't have other colorimeter but I read generally i1D3 is considered quite accurate. Not sure if I should exchange the Spyder 5.
I trust the i1D3 much more than the Spyder5.

Quote:
As for calibrating the JVC rs540, I'll try autocal this evening. I'm still going through the master tread but I think I'll need to run the autocal with normal gamma to correct it (do I run gamma only or gamma and color?), then fine tune colors with i1D3 and HCFR. Then upload the Javs curve (this is for HDR only I assume , for SDR the autocal gamma is okay)? I find the lowest nits curve he has is for 85 nits, I probably will not even get 85 nits because of the screen but will give it a try.
Manni recommends running both and then correct the colours using the spreadsheet method, as that improves the colour saturation tracking. Personally I don't normally run colour calibration with the Spyder. You can always measure first, before deciding whether to run colour autocal.
If your peak luminance is less than 85 you can try my 60 nits curve.
Dominic Chan is online now  
Sponsored Links
Advertisement
 
post #11072 of 11794 Old 11-17-2018, 03:04 PM
Member
 
inonefly's Avatar
 
Join Date: Dec 2005
Location: London
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 10
Quote:
Originally Posted by Dominic Chan View Post
Sorry, I'm not familiar with madVR. However, you still need to set the graphics card to 0-255.



That's not unexpected.





I trust the i1D3 much more than the Spyder5.





Manni recommends running both and then correct the colours using the spreadsheet method, as that improves the colour saturation tracking. Personally I don't normally run colour calibration with the Spyder. You can always measure first, before deciding whether to run colour autocal.

If your peak luminance is less than 85 you can try my 60 nits curve.

Do I have to have Calman to use the spreadsheet method?

I'll use HCFR to do a few runs and compare the i1D3 and spyder 5 see if it's a keeper.

And where can I download your 60 nits curve? I did a search on the JVC autocal tread but didn't find it.


Sent from my iPhone using Tapatalk
inonefly is offline  
post #11073 of 11794 Old 11-17-2018, 03:26 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by inonefly View Post
Do I have to have Calman to use the spreadsheet method?
The spreadsheet method calculates the errors in the messyured xy coordinates of RGBW and use those to derive the xy coordinates of the "custom" (corrected) profile. You can do it using any software that can measure those values. The spreadsheet uses the average errors for a few different stimulus levels, but that's optional.
There were a few posts in recent months that provide some additional details on how to do that in HCFR.

Quote:
I'll use HCFR to do a few runs and compare the i1D3 and spyder 5 see if it's a keeper.

And where can I download your 60 nits curve? I did a search on the JVC autocal tread but didn't find it.
See my signature.

Last edited by Dominic Chan; 11-17-2018 at 03:34 PM.
Dominic Chan is online now  
Sponsored Links
Advertisement
 
post #11074 of 11794 Old 11-17-2018, 05:18 PM
Member
 
inonefly's Avatar
 
Join Date: Dec 2005
Location: London
Posts: 67
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 10
Quote:
Originally Posted by Dominic Chan View Post
The spreadsheet method calculates the errors in the messyured xy coordinates of RGBW and use those to derive the xy coordinates of the "custom" (corrected) profile. You can do it using any software that can measure those values. The spreadsheet uses the average errors for a few different stimulus levels, but that's optional.

There were a few posts in recent months that provide some additional details on how to do that in HCFR.





See my signature.


Thanks! I'm going through the threads slowly but will look for some more details of how to do the correction with HCFR.

Also downloaded your 60 R2 file, what's the difference between 60_25 and 60_1200?


Sent from my iPhone using Tapatalk
inonefly is offline  
post #11075 of 11794 Old 11-17-2018, 05:58 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by inonefly View Post
Thanks! I'm going through the threads slowly but will look for some more details of how to do the correction with HCFR.
Here’s one post:
https://www.avsforum.com/forum/139-d...l#post56825516

Quote:
Also downloaded your 60 R2 file, what's the difference between 60_25 and 60_1200?
I changed the naming convention. The curves are for 4000 nit masters except for those with 1200 in the suffix.

Last edited by Dominic Chan; 11-17-2018 at 06:01 PM.
Dominic Chan is online now  
post #11076 of 11794 Old 11-19-2018, 01:03 AM
Member
 
Clemery76's Avatar
 
Join Date: Oct 2013
Posts: 68
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked: 15
Hi everyone...

I have a couple more calibration queries. I tend to be a bit verbose so will start with two simple questions, then give some context below:

1. Does anyone else find that the greyscale patterns (100%, 95%, etc...) don't quite line up to the IRE adjusters on their TV? (i.e. the 100 IRE slider will affect 95% white more than 100% white)

2. My TV also has sliders to adjust gamma at 10 IRE increments. Should I use these to help smooth out the gamma and get a more accurate gamma line/curve, or is it best to avoid messing with the gamma too much, relying mainly on greyscale calibration?



OK... a bit of context. After attempting calibrations on three different TV's (one backlit LED, two OLEDs), I have always found that my IRE adjustment sliders on the TV never quite match the input greyscale patterns. This is true whether I'm using Colour HCFR patterns downloaded from the internet or patterns generated automatically in HCFR and sent via my laptop over HDMI. Additionally, I find that when I play the black clipping pattern through my laptop over HDMI, I cannot see 17 or below flash, no matter how much I increase the brightness on my TV, almost as though the PC is limiting the output, even though its set to full range. The same black clipping video played through my Minix U9-H media player through the same input displays properly, so I'm finding that I have to set brightness/contrast using these patterns on my Minix, before then switching to the laptop for the auto HCFR patterns.
But no matter the values, I just can't get the patters and IRE sliders to properly match up. e.g. I'll increase brightness by 5 points to 55, and contrast 6 points to 96. Then I'll try to adjust greyscale. At this level, I'll find that the 100IRe slider affects 95% white moreso than 100% white. But it is not linear, in that at the bottom range, the 20% grey patters results in no change when adjusting the 20 IRE slider, but the 10% grey will be greatly affected. AND, this is with HCFR set to output 0-255 (and laptop set to output full range), which is not really ideal for calibrating SDR.
I have found that if I greatly reduce brightness (down to around 35), then the bottom end lines up better, but then the top end doesn't get bright enough (I'd have to display 95% white to adjust the 90 IRE slider), and I also noted that there was visibly no change in white/brightness level between 95% and 100% white windows. I tried fiddling with the laptop output settings, and found that reducing the output gamma from 1.0 to 0.8 would bring it closer into line, but still far from accurate and would have to sacrifice either the top end or bottom end, as well as the gamma line looking more like an arch than with any other mode. As it stands, my current calibration was done using 0-255 output through HCFR with a bit of guestimation thrown in, and naturally the picture looks somewhat unnatural to me. Are there any tips for getting the grey patterns to line up with the IRE levels on the TV? My laptop is a Lenovo Yoga with an Intel HD Graphics 4400 inside, set to 1920x1080, 60hz, Full Range, 32bit colour depth, YCbCr disabled.


As for gamma... in addition to being able to select the gamma from a number of presets ranging from 1.6 - 2.8 (and BT.1886), it also features sliders to adjust the gamma at 10 IRE increments (which again don't quite line up). I used these sliders on the above 0-255 calibration to help get the gamma line (2.2 in this example) straight, but this required some HUGE adjustments from 90 IRE down to 60 IRE (as in -25 to -35), but 0 adjustments from 50 IRE down. This has resulted in a reasonably flat 2.2 gamma line, but the resultant picture appears unnatural and even the greyscale ramp shows a definitive jump in brightness around the 50% mark, so I don't trust that its overly accurate.

I have attached the CHC file for the 0-255 calibration, which measures as very accurate (IMO, but I'm still a newbie so what do I know?), but visually the picture doesn't seem right. Could it possibly be my colorimeter thats just not being accurate? Its only a ColorMunki Display, with no adjustments or correction files applied.

Sorry for the exhaustive post... hopefully someone can help to shed some light on these issues.

Thanks in advance!
Attached Files
File Type: zip 2018-11-17 full tilt boogie - prof 1 warm 2.zip (33.3 KB, 4 views)
Clemery76 is offline  
post #11077 of 11794 Old 11-19-2018, 04:21 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,926
Mentioned: 186 Post(s)
Tagged: 1 Thread(s)
Quoted: 2997 Post(s)
Liked: 3695
Quote:
Originally Posted by Clemery76 View Post
Hi everyone...

I have a couple more calibration queries. I tend to be a bit verbose so will start with two simple questions, then give some context below:

1. Does anyone else find that the greyscale patterns (100%, 95%, etc...) don't quite line up to the IRE adjusters on their TV? (i.e. the 100 IRE slider will affect 95% white more than 100% white)
Hi, for that effect you see, it may be:

1) The difference of levels or RGB triplets of patterns you send with ones the TV expect from you to display. When you calibrate, you need the patterns you send to have exact same triplets (RGB or YCbCr) with the ones the software is expecting, if there mismatch, then you will have mismatch with controls affectibe range also.

When you display patterns from a notebook HDMI, you need specific settings (to notebook/TV) to be sure that the values are correct. (see 4)

2) There display which are suffering with some issues related with a un-align of available calibration controls (10-Point RGB Balance) to the working effecting range of the signal, for example when you display a 50% Gray patch and you dial 50% Gray RGB balance, internally to the display its not changing the 50% of Gray but another level...for example 42% or 56% etc...just for example.)

This makes the calibration more difficult and you will always introduce anomalies when you will finish a 10-Point RGB balance and you display a Grayscale Ramp, there will be visible issues there with strange color-shades.

These specific values (to each display) of contrast/brightness controls which are align better the calibration controls and it will be helpful to measure with 21-Point Grayscale patterns when you calibrate using the available 10-Point controls you have. Doing this will be able to spot better any anomaly and prevent issues.

Its also a better idea to take full grayscale sweep and not do real-time adjustments, to be able to spot easier your adjustments how they affect the grayscale performance....so measure fully grayscale (with menu OSD off)....then look your RGB Balance / Gamma charts....open OSD...apply the adjustments you believe they will help...then close OSD and remeasure grayscale again.

To some displays, the RGB balance controls design in such way where they will align better to exact 10-p luminance levels when you will have contrast/brightness to default positions, to other models it will align better when you will have contrast at max position, this is something you need to test. That un-alignment can interact and shift each RGB balance controls to higher or lower affecting luminance range.

Quote:
Originally Posted by Clemery76 View Post
2. My TV also has sliders to adjust gamma at 10 IRE increments. Should I use these to help smooth out the gamma and get a more accurate gamma line/curve, or is it best to avoid messing with the gamma too much, relying mainly on greyscale calibration?
Start with 2-Point RGB balance and do 100% White with RGB-Gain and 30% Gray with RGB-Offset controls, when you will do many back/forth until you have stable readings, you will move to 10-Point RGB balance, after your are ready with grayscale you will move to CMS calibration.

Quote:
Originally Posted by Clemery76 View Post
OK... a bit of context. After attempting calibrations on three different TV's (one backlit LED, two OLEDs), I have always found that my IRE adjustment sliders on the TV never quite match the input greyscale patterns. This is true whether I'm using Colour HCFR patterns downloaded from the internet or patterns generated automatically in HCFR and sent via my laptop over HDMI. Additionally, I find that when I play the black clipping pattern through my laptop over HDMI, I cannot see 17 or below flash, no matter how much I increase the brightness on my TV, almost as though the PC is limiting the output, even though its set to full range. The same black clipping video played through my Minix U9-H media player through the same input displays properly, so I'm finding that I have to set brightness/contrast using these patterns on my Minix, before then switching to the laptop for the auto HCFR patterns.
But no matter the values, I just can't get the patters and IRE sliders to properly match up. e.g. I'll increase brightness by 5 points to 55, and contrast 6 points to 96. Then I'll try to adjust greyscale. At this level, I'll find that the 100IRe slider affects 95% white moreso than 100% white. But it is not linear, in that at the bottom range, the 20% grey patters results in no change when adjusting the 20 IRE slider, but the 10% grey will be greatly affected. AND, this is with HCFR set to output 0-255 (and laptop set to output full range), which is not really ideal for calibrating SDR.
I have found that if I greatly reduce brightness (down to around 35), then the bottom end lines up better, but then the top end doesn't get bright enough (I'd have to display 95% white to adjust the 90 IRE slider), and I also noted that there was visibly no change in white/brightness level between 95% and 100% white windows. I tried fiddling with the laptop output settings, and found that reducing the output gamma from 1.0 to 0.8 would bring it closer into line, but still far from accurate and would have to sacrifice either the top end or bottom end, as well as the gamma line looking more like an arch than with any other mode. As it stands, my current calibration was done using 0-255 output through HCFR with a bit of guestimation thrown in, and naturally the picture looks somewhat unnatural to me. Are there any tips for getting the grey patterns to line up with the IRE levels on the TV? My laptop is a Lenovo Yoga with an Intel HD Graphics 4400 inside, set to 1920x1080, 60hz, Full Range, 32bit colour depth, YCbCr disabled.
If you are using AVSHD, it has a mismatch to grayscale patterns (see more details reading the ''3'' here), this can be fixed from inside HCFR with a software offset, see here what you can do to fix the issue: LG 2017 OLED Calibration Thread and Settings

Also the 4-Point Saturation of AVSHD is not compatible with HCFR (and there no fix about this), for example:

AVSHD 50% Red Saturation Pattern has RGB Triplet 190.95.95 but HCFR's Color Engine needs/calculates errors from RGB Triplet 191.96.96, it's 0.42 dE2000 error.

AVSHD 75% Magenta Saturation Pattern has RGB Triplet 203.100.203 but HCFR's Color Engine needs/calculates errors from RGB Triplet 202.99.202, it's 0.36 dE2000 error.

About the clipping below 17 you you experiencing is related with the fact that all HTPC software players are expanding by default the video signal (patterns/movies are using Video Levels aka 16-235) to data signal (PC's are using Data Levels aka 0-255), so when they are expanding, all levels 1-16 are becoming 0 and all levels 235-254 are becoming 255. (About the numbers of the flashing bars pattern.. the 1-16 looks like 16 and 235-253 looks as 235)

When you have a HTPC and you need to have the most accurate output settings, you need to create 2 presets to your display/projector, to do one calibration for video levels and one for PC levels.

For Movie/Pattern playback you will need video levels calibration since the content mastered for video levels, so you have to set your VGA output as RGB-Full, your software movie player to RGB-Video and your display/projector to receive RGB-Video.

This is the most accurate way, since you will not expand any content levels and you will left headroom for all levels above 235, do if you display a contrast pattern (for video levels) it will not clip, and it will show you all levels above 235. (when you will set correctly your contrast setting of your display)

See there why you need headroom: LG 2017 OLED Calibration Thread and Settings

If you expand to 0-255 the movie (RGB 0-255 / video playback application 0-255 / Display set to full range), you don't leave any headroom, you clip everything above 235, but colorspace conversion can have values outside video legal range, so it's a better idea to leave headroom, to prevent issues.

While above setting will be good for your movies from inside your software movie player, it will not be good for web/stuff, games which require PC Levels calibration, for that reason you will calibrate (without changing any of your PC settings) a different picture mode to your display, for PC level calibration.

BTW not all videocards can have accurate output to be used a software patch generation. To be sure that your video card is accurate, you have to compare it with your media player output... by displaying and measuring the same patterns. If you find aggrement in black/white levels, gamma, RGB balance and color gamut measurements, then you can use your notebook for patch generation. A lot of people skip that test, unfortunately. (see more details reading the ''4'' here).

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #11078 of 11794 Old 11-19-2018, 11:55 AM
Member
 
Join Date: Nov 2018
Location: Fort Lauderdale, FL
Posts: 22
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 1
Hello everyone,


I am new to calibrating displays. I chose HCFR because it was free, and so far, it seems pretty intuitive with the guide online.

I have a few questions about achieving the best accuracy for my projector and screen combo.


1. I have a UHD60 DLP projector. I am not sure if I should be using Refresh Display, or Projector under the "Display Type". I know the option for projector is there, but its referred to as "Hybrid" type.
2. How far should my meter be away from the screen ? I'm seeing conflicting information, that the meter should be near the middle, and placed 3 - 4" away from the screen. Angling the meter slightly upward unitil I see the max amount of lumens displaying on the constant check at 100 IRE using the amplitude pattern on AVS 709 disc.

Another source says to divide the height of your screen by 2 and place it that far away from the screen, also at mid level.

I have a curved silver screen. 87" H x 207" W


Any help would be super appreciated. I have taken some tests with various projector settings, and have seen a really nice improvement in color, and brightness. I just want to be sure I'm doing this correctly.



The meter I am using is a Colormunki Display.
(Cinema-Scope) is offline  
post #11079 of 11794 Old 11-19-2018, 07:20 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by (Cinema-Scope) View Post
Hello everyone,


I am new to calibrating displays. I chose HCFR because it was free, and so far, it seems pretty intuitive with the guide online.

I have a few questions about achieving the best accuracy for my projector and screen combo.


1. I have a UHD60 DLP projector. I am not sure if I should be using Refresh Display, or Projector under the "Display Type". I know the option for projector is there, but its referred to as "Hybrid" type.
I would use the projector setting.

Quote:
2. How far should my meter be away from the screen ? I'm seeing conflicting information, that the meter should be near the middle, and placed 3 - 4" away from the screen. Angling the meter slightly upward unitil I see the max amount of lumens displaying on the constant check at 100 IRE using the amplitude pattern on AVS 709 disc.
The meter should ideally be placed in line with the viewer. Placing it too close to the screen may make it hard to avoid reading its own shadow. I would place it 3-4 feet away. The distance should make much difference to the readings, as you will find out.
Dominic Chan is online now  
post #11080 of 11794 Old 11-19-2018, 07:48 PM
Member
 
Clemery76's Avatar
 
Join Date: Oct 2013
Posts: 68
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 66 Post(s)
Liked: 15
Quote:
Originally Posted by ConnecTEDDD View Post
Hi, for that effect you see, it may be:

1) The difference of levels or RGB triplets of patterns you send with ones the TV expect from you to display. When you calibrate, you need the patterns you send to have exact same triplets (RGB or YCbCr) with the ones the software is expecting, if there mismatch, then you will have mismatch with controls affectibe range also.

When you display patterns from a notebook HDMI, you need specific settings (to notebook/TV) to be sure that the values are correct. (see 4)

2) There display which are suffering with some issues related with a un-align of available calibration controls (10-Point RGB Balance) to the working effecting range of the signal, for example when you display a 50% Gray patch and you dial 50% Gray RGB balance, internally to the display its not changing the 50% of Gray but another level...for example 42% or 56% etc...just for example.)

This makes the calibration more difficult and you will always introduce anomalies when you will finish a 10-Point RGB balance and you display a Grayscale Ramp, there will be visible issues there with strange color-shades.

These specific values (to each display) of contrast/brightness controls which are align better the calibration controls and it will be helpful to measure with 21-Point Grayscale patterns when you calibrate using the available 10-Point controls you have. Doing this will be able to spot better any anomaly and prevent issues.

Its also a better idea to take full grayscale sweep and not do real-time adjustments, to be able to spot easier your adjustments how they affect the grayscale performance....so measure fully grayscale (with menu OSD off)....then look your RGB Balance / Gamma charts....open OSD...apply the adjustments you believe they will help...then close OSD and remeasure grayscale again.

To some displays, the RGB balance controls design in such way where they will align better to exact 10-p luminance levels when you will have contrast/brightness to default positions, to other models it will align better when you will have contrast at max position, this is something you need to test. That un-alignment can interact and shift each RGB balance controls to higher or lower affecting luminance range.



Start with 2-Point RGB balance and do 100% White with RGB-Gain and 30% Gray with RGB-Offset controls, when you will do many back/forth until you have stable readings, you will move to 10-Point RGB balance, after your are ready with grayscale you will move to CMS calibration.



If you are using AVSHD, it has a mismatch to grayscale patterns (see more details reading the ''3'' here), this can be fixed from inside HCFR with a software offset, see here what you can do to fix the issue: LG 2017 OLED Calibration Thread and Settings

Also the 4-Point Saturation of AVSHD is not compatible with HCFR (and there no fix about this), for example:

AVSHD 50% Red Saturation Pattern has RGB Triplet 190.95.95 but HCFR's Color Engine needs/calculates errors from RGB Triplet 191.96.96, it's 0.42 dE2000 error.

AVSHD 75% Magenta Saturation Pattern has RGB Triplet 203.100.203 but HCFR's Color Engine needs/calculates errors from RGB Triplet 202.99.202, it's 0.36 dE2000 error.

About the clipping below 17 you you experiencing is related with the fact that all HTPC software players are expanding by default the video signal (patterns/movies are using Video Levels aka 16-235) to data signal (PC's are using Data Levels aka 0-255), so when they are expanding, all levels 1-16 are becoming 0 and all levels 235-254 are becoming 255. (About the numbers of the flashing bars pattern.. the 1-16 looks like 16 and 235-253 looks as 235)

When you have a HTPC and you need to have the most accurate output settings, you need to create 2 presets to your display/projector, to do one calibration for video levels and one for PC levels.

For Movie/Pattern playback you will need video levels calibration since the content mastered for video levels, so you have to set your VGA output as RGB-Full, your software movie player to RGB-Video and your display/projector to receive RGB-Video.

This is the most accurate way, since you will not expand any content levels and you will left headroom for all levels above 235, do if you display a contrast pattern (for video levels) it will not clip, and it will show you all levels above 235. (when you will set correctly your contrast setting of your display)

See there why you need headroom: LG 2017 OLED Calibration Thread and Settings

If you expand to 0-255 the movie (RGB 0-255 / video playback application 0-255 / Display set to full range), you don't leave any headroom, you clip everything above 235, but colorspace conversion can have values outside video legal range, so it's a better idea to leave headroom, to prevent issues.

While above setting will be good for your movies from inside your software movie player, it will not be good for web/stuff, games which require PC Levels calibration, for that reason you will calibrate (without changing any of your PC settings) a different picture mode to your display, for PC level calibration.

BTW not all videocards can have accurate output to be used a software patch generation. To be sure that your video card is accurate, you have to compare it with your media player output... by displaying and measuring the same patterns. If you find aggrement in black/white levels, gamma, RGB balance and color gamut measurements, then you can use your notebook for patch generation. A lot of people skip that test, unfortunately. (see more details reading the ''4'' here).



Whoa... thanks for such an exhaustive reply. I think I followed some of that! I'll read it a few times then re-address calibrating tonight or tomorrow.
To clarify, I watch most of my (non-4K) content through my Minix Media player or my Humax PVR, not the laptop. I only used the laptop as the pattern generator because I couldn't get the patterns and IRE levels to line up through the Minix (and for the convenience and speed of using HCFR's auto patterns).
That said, I don't think I've ever run a sweep without adjusting the brightness/contrast, so I'll definitely try leaving them at defaults next time to see if its more in line with the greyscale patterns.


Thanks again for your advice. I'll let you know how I get on.
Clemery76 is offline  
post #11081 of 11794 Old 11-20-2018, 02:03 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 159
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 23
@zoyd
I created a CCSS file using HCFR. The order of rows in the CCSS created by HCFR is RGBW (1=R, 2=G, 3=B, 4=W). But the ArgyllCMS specplot tool expects the order WRGB (1=W, 2=R...) for a correct chart. It should be no problem for the correction by ArgyllCMS but specplot users could be confused by the strange chart.



More info:
https://www.avsforum.com/forum/139-d...l#post57095358


Thank you.

Warm regards,
bejoro
bejoro is offline  
post #11082 of 11794 Old 11-20-2018, 03:07 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,926
Mentioned: 186 Post(s)
Tagged: 1 Thread(s)
Quoted: 2997 Post(s)
Liked: 3695
Quote:
Originally Posted by Clemery76 View Post
Whoa... thanks for such an exhaustive reply. I think I followed some of that! I'll read it a few times then re-address calibrating tonight or tomorrow.
To clarify, I watch most of my (non-4K) content through my Minix Media player or my Humax PVR, not the laptop. I only used the laptop as the pattern generator because I couldn't get the patterns and IRE levels to line up through the Minix (and for the convenience and speed of using HCFR's auto patterns).
That said, I don't think I've ever run a sweep without adjusting the brightness/contrast, so I'll definitely try leaving them at defaults next time to see if its more in line with the greyscale patterns.

Thanks again for your advice. I'll let you know how I get on.
Its fine to use HCFR software generator with your notebook output, but only if you have verified that you have agreement with the same patterns from your actual movie source playback device you are using.

Because it doesn't matter if from your notebook the dE charts are perfect but when you will playback a movie to have added errors.

Since the consumers don't have digital level analysers to their place to see if one device output/settings is accurate (testing digital level, not measuring with meter, so you skip from the test the display stability/meter repeatability, you examine only the output of a source...for example your display a pattern from a player (disk or media fle) for 100% White, it has to be @ 8bit RGB 235.235.235 or YCbCr 235.128.128), the only way to verify if an automated pattern generation solution provide correct levels and colors, you have to use a reference to do some tests.

As reference you can use a normal blu-ray/media player, patterns that are accurate for HCFR, becasue we need the patterns displayed from your player to have the same RGB triplets with the ones the HCFR will generate also, to be able to compare for differencies, not all calibration disks are compatible with HCFR.

Most of the times, YCbCr colorspace output from a stand-alone player can provide bit-perfect/accurate output (when any image enhancing mode is disabled), so using a player and a calibration disk designed for HCFR, can be used as reference to test how an another patch generation solution is performing.

You will have to measure Grayscale and Color Gamut patterns from both patch solutions and see if there agreement to the readings.

There solutions that have correct grayscale tracking but in-correct colors, there others which has correct RGB balance but incorrect gamma, there others which have everything incorrect. Unless you test you don't know which solution/software setting is correct.

There display where are handling differently RGB signal (when you send from a notebook for example) than a YCbCr a standalone player is sending.

When movies are mastered, in color grading studios they are using monitors calibrated which are getting signal from the actual desktop system where connected, calibration software is using patterns generated from inside the color grading software (like Davinci Resolve or SGO Mystika etc.) or using calibration software generated patterns, so the full video chain is calibrated, following some standards as targets, REC709 colorspace and 100nits peak output.

For HT consumer area, you want to reproduce the kind of calibration trying to follow the same standards as the studio monitors where calibrated where the movies are mastered there, so you need to have your full video chain calibrated also, to generate patterns from your movie playback device. If you have a blu-ray player then to use a blu-ray, if you have a media player then a stream the patterns from there or if you have a HTPC then to use pattern generated from inside your software player, to have the end of the day your full video chain calibrated.

The last years there a lot of automated pattern generation solutions available.... which can be loaded to phones (mirroring) or steamers/network players (FireStick/ChromeCast/AppleTV etc.), or using notebook output.... to make the patch generation automated, so to save some time from manual pattern change, all these sound cool as solutions, but the 99% of the users ignore to compare if they are accurate, so they calibrate using them without proper testing to see if the software/hardware output settings are correct, so unless its tested, its a calibration valid in a virtual world, since its untested/unverified if these solutions provide accurate patterns. Makes no sense to use any calibration solution which have deviations from your actual movie playback source, because you will have errors during playback (color errors or wrong levels).

For that reason when you want to use such a solutions you have to first verify that its accurate and its matching the output of your movie source player.

If you find that you have agreement (black/white level, grayscale RGB balance/gamma/color gamut tracking), then you can safely use any software/hardware solutions you have tested that is matching your player, no problem, it will save you time and you will have the same accuracy like you were generating your players manually from your player.

You can see recently a test, how much digital errors were having some 'bit-perfect advertised' solutions with CalMAN: https://www.avsforum.com/forum/139-d...cy-thread.html

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #11083 of 11794 Old 11-20-2018, 03:23 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,926
Mentioned: 186 Post(s)
Tagged: 1 Thread(s)
Quoted: 2997 Post(s)
Liked: 3695
Quote:
Originally Posted by (Cinema-Scope) View Post
1. I have a UHD60 DLP projector.
2. How far should my meter be away from the screen ? I'm seeing conflicting information, that the meter should be near the middle, and placed 3 - 4" away from the screen. Angling the meter slightly upward unitil I see the max amount of lumens displaying on the constant check at 100 IRE using the amplitude pattern on AVS 709 disc.

Another source says to divide the height of your screen by 2 and place it that far away from the screen, also at mid level.
Hi, the ideal meter placement is when the meter is matching the same angle you have from your sitting position eye's height towards to the center of the screen.

You can use a mobile phone or a digital angle meter to find the angle. Imagine a virtual line (you can use a laser pointer) from your eye's @ sitting position aiming the center of the screen, the meter ideally has to placed the that line height and angle.

About the meter distance, its related with what meter you are using and how narrow optics it has, if you use Window patterns you need to find out the meter's FOV (field of view) from the distance you will place the meter.. because we don't need the meter to see larger area from the pattern area. (not see the meter shadow if you place it very close to the screen)

See that PDF about meters FOV (your meter has the same FOV as C6/i1DisplayPRO): SpectraCAL - Why Viewing Angle is Important

To take measurement from seating position you need a meter with a very narrow FOV (for example JETI which has 1.8 degree).

Here you can see how it's performed a projector calibration in post production, from siting position and eye level height:





Here you can see how it's performed a projector calibration in a commercial cinema:



The instrument has to be positioned in the middle of the auditorium in a height of 110 cm.
grendelrt likes this.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #11084 of 11794 Old 11-20-2018, 06:11 AM
Advanced Member
 
grendelrt's Avatar
 
Join Date: Jun 2004
Location: Richmond, VA
Posts: 932
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 557 Post(s)
Liked: 294
Good information, I have always been 3-4 inches away at whatever angle gives me the highest ftl at 100 ire. So it sounds like I need to get further away when calibrating. Any advice on getting the correct ftl off the screen when setting brightness using a closed down iris, would it be from that same position or trying to get closer to try screen?

Sent from my SM-G920V using Tapatalk
grendelrt is offline  
post #11085 of 11794 Old 11-20-2018, 06:28 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by grendelrt View Post
Good information, I have always been 3-4 inches away at whatever angle gives me the highest ftl at 100 ire. So it sounds like I need to get further away when calibrating. Any advice on getting the correct ftl off the screen when setting brightness using a closed down iris, would it be from that same position or trying to get closer to try screen?

Sent from my SM-G920V using Tapatalk
The FtL reading is more or less independent of the distance, as long as the pattern covers the FOV of the meter.
grendelrt likes this.

Last edited by Dominic Chan; 11-20-2018 at 06:50 AM.
Dominic Chan is online now  
post #11086 of 11794 Old 11-20-2018, 04:08 PM
Member
 
Join Date: Nov 2018
Location: Fort Lauderdale, FL
Posts: 22
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 1
Quote:
Originally Posted by Dominic Chan View Post
I would use the projector setting.



The meter should ideally be placed in line with the viewer. Placing it too close to the screen may make it hard to avoid reading its own shadow. I would place it 3-4 feet away. The distance should make much difference to the readings, as you will find out.
Thanks for the responses. Moving the meter to 4 feet away from the screen made a huge difference. I'm going to re-run a calibration now.

Another question. What's the difference between Rec 709 & Rec 709 75%/75% ?

Thanks
(Cinema-Scope) is offline  
post #11087 of 11794 Old 11-20-2018, 04:12 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by (Cinema-Scope) View Post
Thanks for the responses. Moving the meter to 4 feet away from the screen made a huge difference. I'm going to re-run a calibration now.

Another question. What's the difference between Rec 709 & Rec 709 75%/75% ?

Thanks
The latter uses colour patterns at 75% saturation and 75% stimulus, which may better represent “real life” colours. No difference for grey scale.
Dominic Chan is online now  
post #11088 of 11794 Old 11-21-2018, 11:32 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 159
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 23
Quote:
Originally Posted by ConnecTEDDD View Post
For that reason when you want to use such a solutions you have to first verify that its accurate and its matching the output of your movie source player.
I am still using a good old DVDO iScan DUO with integrated test patterns.
Would these test patterns be accurate enough to verify against my HTPC/MadVR/MadTPG?
I would have to set the iScan DUO to RGB color space, the same color space as my GPU?

Warm regards,
bejoro
bejoro is offline  
post #11089 of 11794 Old 11-21-2018, 11:43 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,926
Mentioned: 186 Post(s)
Tagged: 1 Thread(s)
Quoted: 2997 Post(s)
Liked: 3695
Quote:
Originally Posted by bejoro View Post
I am still using a good old DVDO iScan DUO with integrated test patterns.
Would these test patterns be accurate enough to verify against my HTPC/MadVR/MadTPG?
I would have to set the iScan DUO to RGB color space, the same color space as my GPU?
You can test it, but DVDO iScan DUO as a pattern generator can display patterns only with percentage values (0-100%, 1.000.000 possible color patterns) and not exact RGB-Video 8-bit code values (220x220x220 = 10.648.000 possible color patterns).

For that limitation it can't be used from CalMAN, as pattern generator for 3D LUT.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #11090 of 11794 Old 11-21-2018, 12:10 PM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 159
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 23
Quote:
Originally Posted by ConnecTEDDD View Post
You can test it, but DVDO iScan DUO as a pattern generator can display patterns only with percentage values (0-100%, 1.000.000 possible color patterns) and not exact RGB-Video 8-bit code values (220x220x220 = 10.648.000 possible color patterns).

For that limitation it can't be used from CalMAN, as pattern generator for 3D LUT.
I just wanted to compare the built-in DVDO iScan DUO patterns to the patterns generated by MadTPG to make sure that the MadVR and GPU settings are ok (bit perfect). I would make 10p+RGBYMCW readings and compare. Would that make sense?
I use HCFR/MadTPG for manual calibration/readings and DisplayCal/MadTPG for 3DLUT generation.

Warm regards,
bejoro
bejoro is offline  
post #11091 of 11794 Old 11-21-2018, 12:30 PM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,926
Mentioned: 186 Post(s)
Tagged: 1 Thread(s)
Quoted: 2997 Post(s)
Liked: 3695
Quote:
Originally Posted by bejoro View Post
I just wanted to compare the built-in DVDO iScan DUO patterns to the patterns generated by MadTPG to make sure that the MadVR and GPU settings are ok (bit perfect). I would make 10p+RGBYMCW readings and compare. Would that make sense?
I use HCFR/MadTPG for manual calibration/readings and DisplayCal/MadTPG for 3DLUT generation.
The problem is that you don't have control of what exact RGB-Triplet it will be generated, for grayscale for example, if DVDO xx% Gray percentage generates truncated or round to nearest 8-bit.

Ted's LightSpace CMS Calibration Disk Free Version for Free Calibration Software: LightSpace DPS / CalMAN ColorChecker / HCFR
S/W: LightSpace CMS, SpaceMan ICC, SpaceMatch DCM, CalMAN 5, CalMAN RGB, ChromaPure, ControlCAL
V/P: eeColor 3D LUT Box - P/G: DVDO AVLab TPG
Meters: JETI Specbos 1211, Klein K-10A, i1PRO2, i1PRO, SpectraCAL C6, i1D3, C5
ConnecTEDDD is offline  
post #11092 of 11794 Old 11-21-2018, 06:33 PM
Member
 
Join Date: Nov 2018
Location: Fort Lauderdale, FL
Posts: 22
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 1
Would Rec. 709 75 / 75 be a good measure if you are primarily using the projector for Blu-Ray & UHD Blu-Ray films ?
Or should I just stick to the regular measure 709
I haven't tried calibrating HDR yet, and I'm not sure if I'm going to. The projector puts out a good 22 ftL. But I don't think it will good enough for HDR picture on a screen my size. So instead, I have HDR turned off for now, forcing the player to run in SDR.
I may try calibrating HDR later down the line, but it just looks so dark on my screen. I like my movies to pop. HDR just looks good in some scenes, but super dark in most.

Using the Panasonic UB-900 player, I have the ability to scale HDR brightness, which does a nice job at dramatically increasing the contrast to the picture, but it causes the bloom effect when raised to high. Meaning things like clouds are super white and have a bloom effect. HDR is still a mystery to me. I'm not sure what HDR should properly look like. Should it bloom ? Should it be to the point where its about to bloom ? Should there be no bloom at all ? No idea
(Cinema-Scope) is offline  
post #11093 of 11794 Old 11-21-2018, 06:55 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by (Cinema-Scope) View Post
Would Rec. 709 75 / 75 be a good measure if you are primarily using the projector for Blu-Ray & UHD Blu-Ray films ?
Or should I just stick to the regular measure 709
It’s your choice, but in either case run the saturation sweep after your adjustments to see how the colours track through the range.
Dominic Chan is online now  
post #11094 of 11794 Old 11-22-2018, 11:25 AM
Member
 
Join Date: Oct 2018
Posts: 15
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 0
Hello, I recently used HCFR and the built-in pattern generator, and I noticed that when I chose REC709 / 75%, the color codes of the displayed patterns do not match those that the HCFR displays as the target, what's the problem? Sorry for my English.

Wysłane z mojego Redmi Note 3 przy użyciu Tapatalka
kajfux is offline  
post #11095 of 11794 Old 11-22-2018, 12:05 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by kajfux View Post
Hello, I recently used HCFR and the built-in pattern generator, and I noticed that when I chose REC709 / 75%, the color codes of the displayed patterns do not match those that the HCFR displays as the target, what's the problem? Sorry for my English.

Wysłane z mojego Redmi Note 3 przy użyciu Tapatalka
Can you provide some specific examples of the mismatch between the colour codes and displayed patterns?
Dominic Chan is online now  
post #11096 of 11794 Old 11-22-2018, 01:01 PM
Member
 
Join Date: Oct 2018
Posts: 15
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 9 Post(s)
Liked: 0
Red 75% pattern 165: 60: 60, target 186: 66: 66 Green 75% pattern 77: 176: 77, target 86: 199: 86 and so on.
Quote:
Originally Posted by Dominic Chan View Post
Can you provide some specific examples of the mismatch between the colour codes and displayed patterns?
Wysłane z mojego Redmi Note 3 przy użyciu Tapatalka
kajfux is offline  
post #11097 of 11794 Old 11-22-2018, 02:20 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,617
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 4634 Post(s)
Liked: 1717
Quote:
Originally Posted by kajfux View Post
Red 75% pattern 165: 60: 60, target 186: 66: 66 Green 75% pattern 77: 176: 77, target 86: 199: 86 and so on.

Wysłane z mojego Redmi Note 3 przy użyciu Tapatalka
Looks like a bug in the RGB Levels display when you use GDI 16-235. However, it shouldn't actually affect the measurements or the delta-E.
Dominic Chan is online now  
post #11098 of 11794 Old 11-25-2018, 06:21 PM
Member
 
Join Date: Jan 2017
Posts: 84
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 6
Has anyone tried this with a Sony Bravia X800D? My dad wasn't happy with the way the tv worked in their house so he bought a new one and gifted me this tv. I tried using HCFR on my old 1080p LG tv and it worked great because I could tweak individual colors. This Sony doesn't seem to have the fine controls that the lg did.
Tatts4Life is offline  
post #11099 of 11794 Old 11-25-2018, 06:45 PM
Advanced Member
 
Anderegg's Avatar
 
Join Date: Jul 2012
Posts: 702
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 473 Post(s)
Liked: 270
Quote:
Originally Posted by Tatts4Life View Post
Has anyone tried this with a Sony Bravia X800D? My dad wasn't happy with the way the tv worked in their house so he bought a new one and gifted me this tv. I tried using HCFR on my old 1080p LG tv and it worked great because I could tweak individual colors. This Sony doesn't seem to have the fine controls that the lg did.
The HCFR white balance color adjustments are under Expert 1 or 2, under advanced color temperature. See the link below to see what the X800D 10 point screens look like.

https://www.rtings.com/tv/reviews/sony/x800d/settings

Paul
Anderegg is online now  
post #11100 of 11794 Old 11-26-2018, 12:05 AM
Member
 
bejoro's Avatar
 
Join Date: Aug 2002
Posts: 159
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 53 Post(s)
Liked: 23
Simultaneous Measures - Free Measure F9

HCFR 3.5.1.4, Win10pro 1803

EODIS3 / i1pro2 (two documents)

MadTPG (via LAN)

HCFR activates test pattern on MadTPG, nothing happens for about 10 seconds. If you use the the mouse, HCFR hangs for about 10 seconds (bright window) and then returns to normal operation.

Simultaneous measures of i.e.Secondary Colors works perfectly.

Warm regards,
bejoro
bejoro is offline  
Sponsored Links
Advertisement
 
Reply Display Calibration

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off