or Connect
AVS › AVS Forum › Display Devices › Display Calibration › HCFR - Open source projector and display calibration software
New Posts  All Forums:Forum Nav:

HCFR - Open source projector and display calibration software - Page 103

post #3061 of 3435
Quote:
Originally Posted by smokarz View Post


I'm assuming it was correct to use round down levels for 709 disc, Mascior's as well?

Ah, right.
Quote:
Originally Posted by zoyd View Post

correct, AVCHD,DVD=checked, GCD,internal patterns=unchecked

For CMS, I'm a little confused with Mascior's disc.

http://cdn.avsforum.com/4/4d/4dc441a8_FFPatterns.png

Far left is for the colour setting itself? and saturation sweep for the main points?
Edited by xvfx - 1/22/14 at 10:57am
post #3062 of 3435
Thread Starter 
Mascior's disk uses normal rounding so box should be unchecked. This was discussed in his thread.


The left side patterns in that picture are Full field 75% amplitude and 100% amplitude primaries and secondaries (@100% saturation). The right side are 75% Amplitude and 100% Amplitude sweeps (25%, 50%, 75%, 100% saturation)
post #3063 of 3435
Yes, I am very confused myself.

The more I read, the more I get confused.

I wish there's a sticky or FAQ somewhere in this forum that would clearly states:

- What saturation/luminance level to use for which type of displays (Plasma. LED, Projector, etc) to calibrate CMS
- Which type size patterns to use for which display type (windows, fields, APL, etc.)
- What options need to be checked/unchecked in HCFR for the different display type measures (Plasma, LED, etc)
- What options to checked/unchecked in HCFR depending on the patterns you use and the disc that you use

As I am scouring for information in the various threads here, I am trying to document my findings so that I have something organized that I could get back to. And if it's comprehensive enough, I will post it here so that others won't have to spend so much time trying to find these info.
post #3064 of 3435
Zoyd,

I am in the process of breaking in my Plasma this week and will try to calibrate it this weekend. I've been reading up on the GCD thread which you also have a large part in developing the patterns.

I plan to run HCFR with GCD disc. Could you provide a few tips on setting up for Plasma calibration with HCFR/GCD patterns?

1) Any specific changes I need to make in HCFR preferences/menu?

2) Which patterns to run to calibrate greyscale and colors (primaries and secondaries)

3) Which patterns to run to do a sweep of the colors/saturation after I am done calibrating it?

Thanks much!
post #3065 of 3435
For the level of saturation it's depends : some display have great linear color tracking (calibrate at 100% saturation and the others level of saturation is ok (25% sat/50% sat/75% sat) the only solution is try and choose the best overall dE

the luminance is quite different : in general it's recommended to calibrate at 75% luminance (most of the color luminance average is more near from the 50%/75% luminance level than 100 % luminance. but like the saturation, try and choose the best overall dE

for the size : plasma : Never full fields (to avoid abl...) but you can use Window or apl (small size like 10/18% max) but the best option : try and choose the best solution to your eyes
for led, lcd or projector fullfields is ok (attention to dynamic contrast or iris...) but you cab use Window (attention to the fov with projector, to not measure outside the window pattern...)

For the hcfr option, zoyd made hcfr with preferences settings, choose only your probe, if you want to use a matrix or spectral sample, choose a colorspace (hdtv rec 709 or hdtv rec 709 75% - ie 75 % saturation). gamma at default is bt1886 but you can try display gamma (2,2/2,3/2,4...)

There are too many options to describe all the possibilities...
post #3066 of 3435
Thanks for the reply realzven. I am taking notes as I ask questions.

So, it's reccomended to calibrate at 75% luminance?

Let's assumed I calibrate my Samsung Plasma TV at 75%luminance/100% saturation, do I use the exact same patterns to run my sweeps after I finished my calibration?
post #3067 of 3435
Hmm, the saturation sweeps on Mascior's disc really isn't HCFR friendly, are they? Jumping about all over the disc and the 75%/100% white in the sweeps seems to be giving me horrendous dE. Even though my 21 step greyscale is 0.37.

Is it normal for the room lighting off to have no bearing on the RGB levels? and the fact the peak white seems to stay the same (100cd/m2) in a pitch black room as a test... I would have thought dimming the lights, the peak white would have become brighter.
post #3068 of 3435
Quote:
Originally Posted by smokarz View Post

Thanks for the reply realzven. I am taking notes as I ask questions.

So, it's reccomended to calibrate at 75% luminance?

Let's assumed I calibrate my Samsung Plasma TV at 75%luminance/100% saturation, do I use the exact same patterns to run my sweeps after I finished my calibration?
Yes at 75 % luminance for the gamut calibration but when you want to do a saturation sweep or color checker use 100% sat / 100% lum patterns
post #3069 of 3435
How do you perform that? HCFR wants 0% saturation first on each sweep. Doesn't exist on Mascior's disc.
post #3070 of 3435
Use gcd
post #3071 of 3435
Will check it out. Though his greyscale patterns give great results.
post #3072 of 3435
Thread Starter 
It's on my list to skip the 0% measurement when using the MCD disk, you can measure 25% twice - it won't affect anything other than that slot.
post #3073 of 3435
zoyd, is there a way you can implement ambient lighting into the calibration readings? I was so surprised by changing my uplighter lighting how it had no bearing on the RGB levels. Even switching the uplighter off, the peak white never became brighter. Thats what made me think... how would I know I'm even looking at 100cd/m2 with very low lighting or the light at maximum.

Or is this where Calman is required?
post #3074 of 3435
You use your meter in contact mode ? If not light in the room will affect readings...
post #3075 of 3435
I do, as it's up on the wall and too high for a tripod mount for direct line. Maybe 5ft 7 for the centre of the screen.

Need to see if I can mount it on the unit below. Wonder if I've really been looking at something higher than 100cdm2.
Edited by xvfx - 1/24/14 at 9:54am
post #3076 of 3435
You have a projector ?
post #3077 of 3435
No. LED TV with tilted mount. Though, with the X-Rite software for PC displays... how can it know the ambient light with contact but HCFR cannot?
Edited by xvfx - 1/24/14 at 10:22am
post #3078 of 3435
Use your colorimeter in contact mode with hcfr, continuous reading and a 100% white pattern then measure and look at Y and raise contrast to see if Y is measure above 100 cd/m2
post #3079 of 3435
It has been 100cdm2 but that then doesn't detect ambient lighting. As I said, playing with lights made no difference to peak white in contact. Thats why I just assumed HCFR would have tapped into the ambient diffusor just like X-Rites software.
Edited by xvfx - 1/24/14 at 1:20pm
post #3080 of 3435
Quote:
Originally Posted by xvfx View Post

... how can it know the ambient light with contact but HCFR cannot?

If I understand you correctly, you are saying that i1profiler is measuring ambient light simultaneously with taking contact measurements? I find this hard to believe since diffusor needs to be in different position (my colormunki first takes ambient measure, then proceeds to calibration/profiling....

It seems both x-rite and dispcalgui insist or recommend taking ambient light and color temp for rec.709 and adjust refeterces accordingly,so perhaps something that could be incorporated into HCFR, only not sure how important for picture accuracy it would be....
post #3081 of 3435
It has an option in the GUI to adjust to room lighting, thats why I was assuming...
Quote:
Originally Posted by mytbyte View Post

It seems both x-rite and dispcalgui insist or recommend taking ambient light and color temp for rec.709 and adjust refeterces accordingly,so perhaps something that could be incorporated into HCFR, only not sure how important for picture accuracy it would be....

I thought that was the whole point about calibration... adjusted to the room conditions. Change of light/new bulb/lamps and your calibration is kaput?...

Thats why I was shocked from temporarily adjusting the uplighter from it's fixed position, I was amazed it had no bearing on peak white/RGB levels. Or am I thinking all screwed up?
post #3082 of 3435
Quote:
Originally Posted by xvfx View Post

It has an option in the GUI to adjust to room lighting, thats why I was assuming...



Thats why I was shocked from temporarily adjusting the uplighter from it's fixed position, I was amazed it had no bearing on peak white/RGB levels. Or am I thinking all screwed up?

X-rite soft and dispcalgui modify graphics card lut on the fly to compensate for ambient lighting, something you can't do automatically with a TV...you need to keep separate fixed calibrations for different ambient lighting...

My understanding is that if you use contact mode, probe sits fully on the screen, blocking the optics from outside light...exception would be TVs with double glass and/or gap between glass and screen where outside light can creep in between probe and screen, also reflections may be an issue
post #3083 of 3435
Quote:
Originally Posted by zoyd View Post

no need for the spreadsheet, HCFR does all the calculations when you choose the 75% Rec709 color space.

Unless I'm crazy/doing something wrong, why are the targets different for Rec709 (75%) in HCFR than they are on the GCD spreadsheet? i.e. Red is .549, .330 in the CIE target chart in HCFR, but the spreadsheet says .564, .330?
post #3084 of 3435
Quote:
Originally Posted by CheYC View Post

Unless I'm crazy/doing something wrong, why are the targets different for Rec709 (75%) in HCFR than they are on the GCD spreadsheet? i.e. Red is .549, .330 in the CIE target chart in HCFR, but the spreadsheet says .564, .330?

Spreadsheet says Red 75% 0.558,0.330 and HCFR says 0.552,0.330 so someone's off somewhere....

Good catch CheYC...
post #3085 of 3435
Thread Starter 
Targets will change depending on the value of the power law exponent being used. Higher gamma, more saturated target, so check that you have entered the same value in the spreadsheet that HCFR is using.
post #3086 of 3435
Quote:
Originally Posted by zoyd View Post

Targets will change depending on the value of the power law exponent being used. Higher gamma, more saturated target, so check that you have entered the same value in the spreadsheet that HCFR is using.

Ahh, didn't even notice, thanks. So in HCFR, is I used Display Gamma (power law) will it use the actual measured average gamma (which was 2.3 in my case) or will it use the 2.22 that I have inputed in the preferences for the targets? If that's the case, would it techincally be more accurate to input 2.3 in the spreadsheet and use those values?
post #3087 of 3435
Upon noticing people with Calman and their greyscale results, some are using CIE2000, some CIE1994. Is anyone using this with HCFR? Since default is CIE76.
post #3088 of 3435
Thread Starter 
Quote:
Originally Posted by CheYC View Post

Ahh, didn't even notice, thanks. So in HCFR, is I used Display Gamma (power law) will it use the actual measured average gamma (which was 2.3 in my case) or will it use the 2.22 that I have inputed in the preferences for the targets? If that's the case, would it techincally be more accurate to input 2.3 in the spreadsheet and use those values?

If "use measured gamma" is checked, the program uses the average value from your grayscale measurement. If unchecked it uses a value of 2.22.
post #3089 of 3435
What distance should the probe be away from the screen? As I'm getting strange results. Well, not so much strange but the RGB levels are more or less identical except playing with the room lighting... Has no effect? The other problem is the probe is only 24cm away from the screen due to the height of the TV and unit on the floor. Room lights off with backlight 9 out of 20 and it's still 100cd/m2.

It also wants a gamma of 1.8 to 2.24. Not the case if it's against the screen. 2.1 to 2.31. I also don't understand how the probe is reading 0% black 0.14cd/m2 clearly for 5 minutes but against the screen it's 0.3.
post #3090 of 3435
Quote:
Originally Posted by realzven View Post

For the level of saturation it's depends : some display have great linear color tracking (calibrate at 100% saturation and the others level of saturation is ok (25% sat/50% sat/75% sat) the only solution is try and choose the best overall dE

the luminance is quite different : in general it's recommended to calibrate at 75% luminance (most of the color luminance average is more near from the 50%/75% luminance level than 100 % luminance. but like the saturation, try and choose the best overall dE

Whats the reason AVS709 75% HCFR patterns aren't as good as Mascior's and GCD's 75%? Between the tripod and contact, cyan only required one click. That and between Mascior's disc and GCD there was very little difference. AVS709 gave me badly undersaturated results. The pixels seems so populated now.

In the past for The Dark Knight Rises through the tunnel scene there was a pitted/grungy look through dark parts on the dashboard and mirror.

I've never seen muted colours so bold plus the smoothness of the gradients. Especially Underworld Rise of the Lycans. The skin tones and the eerie coloured blood in the glasses surrounded by all that oily black leather.

Shadow detail truly is fantastic. No blotchy speckles in Man of Steel gradients flying above the earth with BT.1886. Some faint banding in that glow but it's so tiny.

Primaries and secondaries 1.37dE
Red sat 2.13dE, Green sat 2.42, Blue sat 2.25dE, Yellow sat 0.88dE, Cyan sat 0.88dE Magenta 1.16dE, Color Checker 1.57dE/Dark skin 0.7dE and Light skin 1.4 dE.

CIE2000 gave me a warmer white compared to CIE76uv. Really love it. It's subtle though but it makes a big difference.

Sky HD broadcasting matches up well with the Blu-Ray's. Some ever so slightly lighter blacks in certain scenes but it's so small you wouldn't notice it, unless using a colour picker. Maybe the compression in some films....

m30u.png
d471.jpg
sogk.jpg
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › HCFR - Open source projector and display calibration software