HCFR - Open source projector and display calibration software - Page 168 - AVS Forum
Forum Jump: 
 96Likes
Reply
Thread Tools
post #5011 of 5037 Old 12-14-2014, 09:50 PM
Advanced Member
 
WPWoodJr's Avatar
 
Join Date: May 2007
Location: Villanova, PA
Posts: 779
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 10
Quote:
Originally Posted by zoyd View Post
no, samsung calls 0-255 normal and 16-235 low
I'm confused. When I set my NVidia card to output RGB, then the F8500 HDMI level must be set to Low to show blackest blacks. When I set the card to output YCbCr, the HDMI level is grayed out at Normal, and blacks look the same as RGB with Low setting.

Doesn't this contradict what you said above?

Samsung PN60F8500, TiVo Roamio, ASRock Vision 3D HTPC, Chromecast
WPWoodJr is offline  
Sponsored Links
Advertisement
 
post #5012 of 5037 Old 12-15-2014, 08:46 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
no that means your video card is sending black at 16 and white at 235 over hdmi.
zoyd is offline  
post #5013 of 5037 Old 12-15-2014, 09:05 AM
Newbie
 
Join Date: Dec 2014
Posts: 3
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 0
I personally use the particular EODIS3 with all the HCFR (3.A single.6) computer software to calibrate my EPSON TW-9200 projector (5030UB). I??am through Germany and in a German born AV forum you will find a few reports this combination (EODIS3 as well as HCFR) leads to the color red will be measured to be able to lower which this kind of brings about a reddish photo.
aboveave is offline  
post #5014 of 5037 Old 12-15-2014, 12:33 PM
Advanced Member
 
WPWoodJr's Avatar
 
Join Date: May 2007
Location: Villanova, PA
Posts: 779
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 10
Quote:
Originally Posted by zoyd View Post
no that means your video card is sending black at 16 and white at 235 over hdmi.
Let me see if I get what the implications are.

The PC is an RGB (0-225) device. When using RGB over HDMI, it must be mapping 0-255 to 16-235 before sending, losing color resolution. This would not affect Blue Ray video (from the PC) which is already only using 16-235, but the Blue Ray player must map 16-235 to 0-255? The loss of color resolution would possibly be visible with photographs or other static content as banding or something.

When using YCbCr over HDMI, the PC is again mapping 0-255 to 16-235, with same implications as above.

Is that correct? BTW I couldn't detect a significant difference in HCFR between the two settings (RGB/Low vs YCbCr/Normal).

Samsung PN60F8500, TiVo Roamio, ASRock Vision 3D HTPC, Chromecast
WPWoodJr is offline  
post #5015 of 5037 Old 12-15-2014, 03:28 PM
Advanced Member
 
gwgill's Avatar
 
Join Date: Jan 2013
Location: Melbourne, Australia
Posts: 683
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 102 Post(s)
Liked: 88
Quote:
Originally Posted by aboveave View Post
I personally use the particular EODIS3 with all the HCFR (3.A single.6) computer software to calibrate my EPSON TW-9200 projector (5030UB). I??am through Germany and in a German born AV forum you will find a few reports this combination (EODIS3 as well as HCFR) leads to the color red will be measured to be able to lower which this kind of brings about a reddish photo.
That will happen if you are not using an appropriate calibration for your display type. I could locate no credible reports of ArgyllCMS/HCFR id13 measurements being less accurate than with any other software. If anyone has a reproducible example of such a thing, then they should bring it forward.
gwgill is online now  
post #5016 of 5037 Old 12-16-2014, 09:13 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
Quote:
Originally Posted by WPWoodJr View Post
Let me see if I get what the implications are.

The PC is an RGB (0-225) device. When using RGB over HDMI, it must be mapping 0-255 to 16-235 before sending, losing color resolution. This would not affect Blue Ray video (from the PC) which is already only using 16-235, but the Blue Ray player must map 16-235 to 0-255? The loss of color resolution would possibly be visible with photographs or other static content as banding or something.
Some video cards scale everything over HDMI to video levels 16-235 so the rendering software (MPC etc.) will have to expand blu-ray to 0-255 first to end up with correct output this way. A better situation is to force the video card to leave the levels alone when outputting RGB over HDMI, that way if you choose to render at video levels or full RGB there will be no scaling in the video card and no possibility of banding. This is the preferred HTPC set-up.

Quote:
When using YCbCr over HDMI, the PC is again mapping 0-255 to 16-235, with same implications as above.

Is that correct? BTW I couldn't detect a significant difference in HCFR between the two settings (RGB/Low vs YCbCr/Normal).
If you set the card to output YCC it will also scale everything to video levels with the possibility of banding.

My preferred calibration set-up is HCFR 16-235 + video card passthrough (0-255) + display set to RGB limited (low).
WPWoodJr likes this.
zoyd is offline  
post #5017 of 5037 Old 12-17-2014, 09:57 AM
Member
 
trp3383's Avatar
 
Join Date: Nov 2008
Posts: 127
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 24 Post(s)
Liked: 11
Hey, can't get my saturation's any better on my 64f8500.
Using oppo-103d set to 4:4:4
GDC disk with 10% APL and hcfr set to 75/75
Cell 17
Contrast 95
Brightness 49
Sharp 6
color 59

The last pic shows saturation sweep using 100/100 and Primary's and secondary's using 75/75.

Anything I can do to improve it? Thanks
Attached Thumbnails
Click image for larger version

Name:	GSGAMMA.png
Views:	28
Size:	256.3 KB
ID:	427498   Click image for larger version

Name:	GSRGB.png
Views:	27
Size:	207.4 KB
ID:	427506   Click image for larger version

Name:	BlueSat.png
Views:	27
Size:	378.2 KB
ID:	427530   Click image for larger version

Name:	GreenSat.png
Views:	28
Size:	378.2 KB
ID:	427538   Click image for larger version

Name:	RedSat.png
Views:	25
Size:	378.1 KB
ID:	427546  

Click image for larger version

Name:	Color75Sat100.png
Views:	29
Size:	373.0 KB
ID:	427570  

Last edited by trp3383; 12-17-2014 at 11:31 AM.
trp3383 is offline  
post #5018 of 5037 Old 12-17-2014, 10:29 AM
AVS Special Member
 
LastButNotLeast's Avatar
 
Join Date: Feb 2007
Location: 08077
Posts: 4,888
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 316 Post(s)
Liked: 294
It looks great. If you're worried about the numbers, you need to run everything with preferences back to 100%/100%.
Just expect your numbers at 100% to be worse than your numbers at 75% (which is fine).

Downloadable FREE demo discs:
http://www.avsforum.com/t/1475769/de...ently-authored 

Did you really need to quote that entire post in your reply?
LastButNotLeast is offline  
post #5019 of 5037 Old 12-17-2014, 12:22 PM
AVS Special Member
 
LastButNotLeast's Avatar
 
Join Date: Feb 2007
Location: 08077
Posts: 4,888
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 316 Post(s)
Liked: 294
Until midnight today:
http://www.bhphotovideo.com/c/produc...splay_Pro.html
i1Pro $129 AR.

Downloadable FREE demo discs:
http://www.avsforum.com/t/1475769/de...ently-authored 

Did you really need to quote that entire post in your reply?
LastButNotLeast is offline  
post #5020 of 5037 Old 12-17-2014, 12:51 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
Quote:
Originally Posted by trp3383 View Post
Anything I can do to improve it? Thanks
Looks darn good to me. How are the two skin tones on the color checker patterns and overall average? If they are a little hot you might back off the color control a couple of clicks and/or play with the flesh tone setting to tweak those (mine ended up at +3).
zoyd is offline  
post #5021 of 5037 Old 12-17-2014, 03:27 PM
Advanced Member
 
xvfx's Avatar
 
Join Date: Feb 2013
Location: UK
Posts: 744
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 123 Post(s)
Liked: 87
Does such a set exist where 75/75 lines up with 100/100 Luminance values? I've always noticed quite a bit of variation but 75/75 looks just a smidge too saturated. More so with Big Trouble in Little China at the start. 100/100 rectifies glowing skin tones around Kurt Russell's neck and the dog's hair looks natural ginger.
xvfx is offline  
post #5022 of 5037 Old Yesterday, 02:16 AM
Newbie
 
CharlieS.'s Avatar
 
Join Date: Dec 2014
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 0
A question.
Is it normal that the measurements without changes to the TV up to 3% can vary?
I use a i1 Display Pro which is the profiled with an i1 Pro2.
If I calibrate 10-P White Balance may vary at each measurement Red and Blue
at any white-points between 100% and 103%. The measurements are not stable.
I have a 55VT60 and use the pattern disc from Ted.
What can I do?


Sorry for my english.
CharlieS. is online now  
post #5023 of 5037 Old Yesterday, 07:18 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
Quote:
Originally Posted by CharlieS. View Post
A question.
Is it normal that the measurements without changes to the TV up to 3% can vary?
Some variation is normal at all levels so it depends on how much you are talking about. It's not unusual for low levels to vary more than high levels.

Quote:
I use a i1 Display Pro which is the profiled with an i1 Pro2.
If I calibrate 10-P White Balance may vary at each measurement Red and Blue
at any white-points between 100% and 103%. The measurements are not stable.
I have a 55VT60 and use the pattern disc from Ted.
What can I do?
How unstable? Make sure you are running the probe in refresh mode and that you have synced to an 80% pattern when setting up the probe.
CharlieS. likes this.
zoyd is offline  
post #5024 of 5037 Old Yesterday, 07:24 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
Quote:
Originally Posted by xvfx View Post
Does such a set exist where 75/75 lines up with 100/100 Luminance values? I've always noticed quite a bit of variation but 75/75 looks just a smidge too saturated. More so with Big Trouble in Little China at the start. 100/100 rectifies glowing skin tones around Kurt Russell's neck and the dog's hair looks natural ginger.
For plasmas it's typical to get some variation in the saturation tracking using the built-in calibration controls. You can usually get this variation down to the 1-2 dE2k level.
zoyd is offline  
post #5025 of 5037 Old Yesterday, 11:14 AM
Advanced Member
 
vega509's Avatar
 
Join Date: Apr 2003
Location: The Northeast
Posts: 872
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 65 Post(s)
Liked: 39
ITU-R BT.1886 Gamma formulae

does anyone have a link to the ITU-R BT.1886 Gamma formulae for target and measured. I'm building a spreadsheet to let me see the effect of varying 0% readings on the curve. Calman 5.4 has a problem when the 0% reading is 0 or close to it.








HCFR will return an almost flat gamma curve at 2.4 with a 0 reading.
Attached Thumbnails
Click image for larger version

Name:	0.0001 reading.jpg
Views:	6
Size:	25.7 KB
ID:	429297   Click image for larger version

Name:	0.030 reading.jpg
Views:	4
Size:	32.1 KB
ID:	429305   Click image for larger version

Name:	.0001 reading.jpg
Views:	4
Size:	230.4 KB
ID:	429313   Click image for larger version

Name:	.03 reading.jpg
Views:	4
Size:	229.3 KB
ID:	429321  

Last edited by vega509; Yesterday at 11:56 AM.
vega509 is offline  
post #5026 of 5037 Old Yesterday, 11:48 AM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
zoyd is offline  
post #5027 of 5037 Old Yesterday, 12:10 PM
Advanced Member
 
vega509's Avatar
 
Join Date: Apr 2003
Location: The Northeast
Posts: 872
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 65 Post(s)
Liked: 39
thanks zoyd.

is there a reason why 0% and 100% gamma is not calculated?
vega509 is offline  
post #5028 of 5037 Old Yesterday, 12:39 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
At 0% log(0) is undefined and at 100% point gamma is infinite because log(1) = 0 in the denominator.
zoyd is offline  
post #5029 of 5037 Old Yesterday, 01:07 PM
Newbie
 
CharlieS.'s Avatar
 
Join Date: Dec 2014
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 0
Quote:
Originally Posted by zoyd View Post
Some variation is normal at all levels so it depends on how much you are talking about. It's not unusual for low levels to vary more than high levels.
How unstable? Make sure you are running the probe in refresh mode and that you have synced to an 80% pattern when setting up the probe.
For example, I'm in the first measurement at 70% White has a value of 97% Red / 99% Green / 102% blue, at the next measurement (without changing anything on TV)
I have 102% Red / Green 98% / 98% blue. Each measurement is different, not only in the dark areas.


Yes, i have synced with 80% pattern. But what ist the refresh mode?
CharlieS. is online now  
post #5030 of 5037 Old Yesterday, 01:25 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
If it asks you to sync then you are in refresh mode (display type = refresh, CRT, or plasma). 1-2% fluctuations are typical in each channel. Why don't you run a 10 pt. scan using a fixed 20% pattern and then a fixed 80% pattern (so 10 measurements at each level) and post the results. Do this with the probe integration time set to 0.5 sec.
CharlieS. likes this.
zoyd is offline  
post #5031 of 5037 Old Yesterday, 01:33 PM
Newbie
 
CharlieS.'s Avatar
 
Join Date: Dec 2014
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 0
Quote:
Originally Posted by zoyd View Post
If it asks you to sync then you are in refresh mode (display type = refresh, CRT, or plasma). 1-2% fluctuations are typical in each channel. Why don't you run a 10 pt. scan using a fixed 20% pattern and then a fixed 80% pattern (so 10 measurements at each level) and post the results. Do this with the probe integration time set to 0.5 sec.




Should I measure a 80% pattern 10 times and then a 20% pattern 10 times?
Or should I calibrate 20% and 80%, and then measure 10 times the entire grayscale?

Sorry, my english is very bad.
CharlieS. is online now  
post #5032 of 5037 Old Yesterday, 01:42 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
Measure 80% 10 times and then 20% 10 times. This will tell us what the random variations are in your set-up for the two levels.
CharlieS. likes this.
zoyd is offline  
post #5033 of 5037 Unread Yesterday, 02:39 PM
Newbie
 
CharlieS.'s Avatar
 
Join Date: Dec 2014
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 0
I dont know how i can show the rgb level for all measurements.
So here are the entire grayscale. Hope it is ok.
It's not as bad as yesterday, but it's a bit frustrating if you do not know what you should now adjust the TV,
if you do not know what values are correct.
Attached Thumbnails
Click image for larger version

Name:	RGB.jpg
Views:	21
Size:	132.1 KB
ID:	429537  
CharlieS. is online now  
post #5034 of 5037 Unread Yesterday, 07:06 PM - Thread Starter
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 5,035
Mentioned: 40 Post(s)
Tagged: 0 Thread(s)
Quoted: 450 Post(s)
Liked: 430
They might all be "correct" in that you always have variations within measurement sets based on the stability of the probe and display. I can't tell much from the plots, can you save the two measurement runs (10 at 20% and 10 at 80%), zip the .chc files and post them?
zoyd is offline  
post #5035 of 5037 Unread Today, 12:02 AM
Newbie
 
CharlieS.'s Avatar
 
Join Date: Dec 2014
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 0
Quote:
Originally Posted by zoyd View Post
They might all be "correct" in that you always have variations within measurement sets based on the stability of the probe and display. I can't tell much from the plots, can you save the two measurement runs (10 at 20% and 10 at 80%), zip the .chc files and post them?
No, I have to make new measurements I have not saved the file.
One more question. If I calibrate the colors I use REC 709 75% / 75%,
if I use REC 709 75% / 75% for gray balance, I can see a slight increase of the curves as if I use 709 REC. This can be seen when
a measurement of the gray scale with REC 709 makes and then to REC 709 75% / 75% changes.
The measured values remain the same, only the graphics changed.
Attached Thumbnails
Click image for larger version

Name:	REC709.jpg
Views:	5
Size:	335.8 KB
ID:	430505  
CharlieS. is online now  
post #5036 of 5037 Unread Today, 02:17 AM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 272
Mentioned: 50 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
rec709 75% uses the exact same whitepoint/grayscale (D65 if you didn't change it) - so the difference you are interested in is artificial.

You've picked up your error margin elsewhere (Measured point on the panel, panel brightness, meter fluctuations, ..).

Last edited by harlekin; Today at 02:22 AM.
harlekin is online now  
post #5037 of 5037 Unread Today, 02:31 AM
Senior Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 272
Mentioned: 50 Post(s)
Tagged: 0 Thread(s)
Quoted: 85 Post(s)
Liked: 20
Also - more in general - there is a whole subset of TVs, where Calibration with a meter is actually quite useless, because of what is described as "decent screen uniformity".

hdtvtest picked up on that finally - and now looks at scren uniformity as well:
http://www.hdtvtest.co.uk/news/55ec9...1412163962.htm

If you are looking at this review - there simply is no correct whitepoint for that display. And its not even uncommon for the majority of displays on the market.

http://www.rtings.com/info/gray-unif...een-effect-dse

Again - this point of failure is kept mostly silent by this community - because plasmas are not effected and the professionals in here just want to sell a service.

The error margin measured by me on two top of the line Sony LED Backlight tvs was dE 7 - depending on where on the screen you would measure - there even were significant variations in the center region).
harlekin is online now  
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