CalMan 5 Release Notes and Discussion - Page 79 - AVS Forum

AVS Forum > Display Devices > Display Calibration > CalMan 5 Release Notes and Discussion

Display Calibration

AVfile's Avatar AVfile
01:23 PM Liked: 63
post #2341 of 2595
05-18-2014 | Posts: 2,050
Joined: Dec 2002
I don't recall 5.3 being released. If so they need to update their web site for the new features and hardware support page.
ACappo's Avatar ACappo
02:39 PM Liked: 44
post #2342 of 2595
05-18-2014 | Posts: 141
Joined: Jun 2013
Can BT1886 reveal banding?

my issue is the following My son is watching X-Men 2 on sky (satellite) an in a dark scene i notices some banding on the background, initially I was thinking Image compression is not uncommon on TV, however I decided (much to my sons frustration smile.gif ) to pause the scene and turn the cLUT off , this revealed that there still was banding however not as much, could this be down to the fact that I use BT1886 on my cLUT and that my TV in general have a flat Power Law 2.2. i.e. without the cLUT the dark back ground was darker, and brighter with the cLUT applied (more revealing).
ConnecTEDDD's Avatar ConnecTEDDD
02:52 PM Liked: 515
post #2343 of 2595
05-18-2014 | Posts: 2,441
Joined: Sep 2010
Quote:
Originally Posted by ACappo View Post

Quote:
Originally Posted by ConnecTEDDD View Post

If you see the LightSpace's Customers List, Shooters Inc. and Park Road Post Productions are using LightSpace and propably used monitors calibrated for Blu-Ray (REC.709) Release with Power Gamma Law 2.2-2.3 Power Law Gamma.

BT1886 is a fairly new feature in LS right? So yes based on that BT1886 is unlikely.

Ted you know of the cube solutions available for the pro marked, how many other have BT1886 as a gamma option?

The defacto applications that post-production studios are using all these years for large 3D LUT Profilings are:

BaseLight TrueLight (September 2003)

Rising Sun Research CineSpace (around 2002) changed to CineTal CineSpace (August 2008) - > later became THX CineSpace CineCube (April 2012) were it was locked to REC.709 only (with reduced price).

Light Illusion LightSpace CMS (January 2009) (The Chief Color Scientist of LightSpace was responsible for the colour science in TrueLight)

CalMAN Studio where it was entered lately to pro industry (July 2012).

During my latest access of CineSpace's Private Secured FTP, they didn't had support for BT1886, just at August 2013 they added the Target Profiling File for REC.709 2.4 Power Gamma, before August it were available only REC709 2.2

THX CineSpace/CineCube has 2 worklows: ITU-R_709_g2.2 + ITU-R_709_g2.4, no BT1886...

It was not possible to enter custom gamma value with CineSpace.

Also THX CineSpace/CineCube project has now discontinued... frown.gif Their site are offline also... frown.gif

BaseLight TruLight has BT1886 as an option but I don't see any option for Black Offset, I can't test a evaluation to tell you this 100% because I don't have a Mac or a Truelight Projector / Monitor Probe...

TruLight works only with their probe, no other probes are supported. frown.gif But meter correction matrix is available.

Since the Post Production Industry don't widely using BT1886, you can calibrate to any gamma target you want, just note that REC.709 2.2 Power Gamma is the most common gamma used for the movies that have been mastered since now...

If you take 3 different displays, lets say, one plasma, one LED-backlit LCD and one CCFL-LCD backlighting and you calibrate them using the same gamma target value, if you display the same material after that the same time from that 3 panels, it will be different at shadow detai/mid-tones etc. etc.. to your eyes....

An idea of having 2-3 different memories to your LUT Box of to your internal display memories and swap them to find which one that it gives you the best performace of your display to each specific movie using your eyes is the best solution right now.

We can't do anything about it, and we don't know if the monitors that are calibrated at studios are calibrated so tight to the specs..... or how frequently they are calibrated... a mystery.....
Iron Mike's Avatar Iron Mike
04:24 PM Liked: 362
post #2344 of 2595
05-18-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by ACappo View Post

Can BT1886 reveal banding?

my issue is the following My son is watching X-Men 2 on sky (satellite) an in a dark scene i notices some banding on the background, initially I was thinking Image compression is not uncommon on TV, however I decided (much to my sons frustration smile.gif ) to pause the scene and turn the cLUT off , this revealed that there still was banding however not as much, could this be down to the fact that I use BT1886 on my cLUT and that my TV in general have a flat Power Law 2.2. i.e. without the cLUT the dark back ground was darker, and brighter with the cLUT applied (more revealing).

can you elaborate on your latest calibration ?

which display/PJ did you calibrate, CM options and cube size, validation workflow.... did the 1000+pt val patch set run through okay the second time ?

Can you post pics here of the CM report of the large validation patch set ?

Thanks.

- M
cadett's Avatar cadett
07:31 PM Liked: 85
post #2345 of 2595
05-18-2014 | Posts: 823
Joined: Aug 2009
Quote:
Originally Posted by ConnecTEDDD View Post

Hello guys, the last official release was CalMAN 5 Home Theater 5.3.0.1450
I pulled that info directly from the public site. Apparently they have another version somewhere else.
HDTVChallenged's Avatar HDTVChallenged
12:25 AM Liked: 154
post #2346 of 2595
05-19-2014 | Posts: 8,650
Joined: Jan 2002
Quote:
Originally Posted by ACappo View Post

Can BT1886 reveal banding?.

Yes ... absolutely, yes. Perhaps the worst example of banding I've ever seen is on the remake of "Fright Night" (delivered via D*, HBO/Sho/Starz ... don't recall which at the moment)

It's not BT1886's fault, the banding is clearly in the source ... it's just much more noticeable when you flatten out the "gamma" at the dark end.
ACappo's Avatar ACappo
02:56 PM Liked: 44
post #2347 of 2595
05-19-2014 | Posts: 141
Joined: Jun 2013
Mike I did not try again with the large patch set, but I have just set it of again with a gamma of 2.22.

This bring up a question.

All my tv modes have undersaturated green with the exception of dynamic, has anyone tried to calibrate that? I have reduced contrast and backlight and set brightness so it look ok, however my pre calibration grayscale if very off, can the cube sort that out? I guess time will tell smile.gif

HDTV

I will try with 2.22 hopefully that will help, I must admit I have yet to see a movie where I preferred bt1886
Iron Mike's Avatar Iron Mike
03:01 PM Liked: 362
post #2348 of 2595
05-19-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by ACappo View Post

Mike I did not try again with the large patch set, but I have just set it of again with a gamma of 2.22.

This bring up a question.

All my tv modes have undersaturated green with the exception of dynamic, has anyone tried to calibrate that? I have reduced contrast and backlight and set brightness so it look ok, however my pre calibration grayscale if very off, can the cube sort that out? I guess time will tell smile.gif

HDTV

I will try with 2.22 hopefully that will help, I must admit I have yet to see a movie where I preferred bt1886

the eeColor 65^3 LUT can/will definitely improve your GS, it has 65 GS points, but you should try to setup GS as good as possible before profiling, do at least 30/100 and try to get 100 as accurate as possible...

post back the large validation patch set report when u had a chance to run it... that will give us great feedback how the LUT performs !

- M
ACappo's Avatar ACappo
03:15 PM Liked: 44
post #2349 of 2595
05-19-2014 | Posts: 141
Joined: Jun 2013
Problem is in dynamic mode does not have any gray scale control.

I will post the validation files when done, but might be a few days, depending on the dynamic test
Iron Mike's Avatar Iron Mike
03:45 PM Liked: 362
post #2350 of 2595
05-19-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by ACappo View Post

Problem is in dynamic mode does not have any gray scale control.

I will post the validation files when done, but might be a few days, depending on the dynamic test

I meant, setup the display BEFORE you start the profiling process... that's the correct approach...

ergo, load the CM 30/100 chart (there is one somewhere under Advanced Calibration IIRC) - adjust 30 and 100 on your screen via the internal display CMS, and when all good then start your LUT profiling process in CM... smile.gif
ACappo's Avatar ACappo
03:56 PM Liked: 44
post #2351 of 2595
05-19-2014 | Posts: 141
Joined: Jun 2013
I'm clearly not good at explaining myself smile.gif
To avoid undersaturated green am I using the dynamic profile on my tv, however when selecting dynamic all advanced options in the tv menu is greyed out, no 2pg grayscale settings and no CMS.
Iron Mike's Avatar Iron Mike
04:01 PM Liked: 362
post #2352 of 2595
05-19-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by ACappo View Post

I'm clearly not good at explaining myself smile.gif
To avoid undersaturated green am I using the dynamic profile on my tv, however when selecting dynamic all advanced options in the tv menu is greyed out, no 2pg grayscale settings and no CMS.

what is a "dynamic profile" and what is the name and model of this mysterious tv ? smile.gif
JimP
04:11 PM Liked: 211
post #2353 of 2595
05-19-2014 | Posts: 15,547
Joined: Sep 2002
Quote:
Originally Posted by Iron Mike View Post

what is a "dynamic profile" and what is the name and model of this mysterious tv ? smile.gif

Better yet, put your display and calibration equipment/software in your signature so we can better help you.
Holiday121's Avatar Holiday121
07:25 PM Liked: 39
post #2354 of 2595
05-19-2014 | Posts: 1,737
Joined: Feb 2009
Is this ok for placemebt it's about 3-4 inches away but it's a HP dalite screen so it's kinda in direct path of my projector.

My projector is shelf mounted above my head for best results with that screen.


sotti's Avatar sotti
09:59 PM Liked: 176
post #2355 of 2595
05-19-2014 | Posts: 6,644
Joined: Aug 2004
Quote:
Originally Posted by Holiday121 View Post

Is this ok for placemebt it's about 3-4 inches away but it's a HP dalite screen so it's kinda in direct path of my projector.

My projector is shelf mounted above my head for best results with that screen.


Looks pretty good. You'd want the meter at about the same angle of incident as your eyeballs for viewing. I'd guess that would mean a little more horizontal if you can work it that way.
Holiday121's Avatar Holiday121
10:39 PM Liked: 39
post #2356 of 2595
05-19-2014 | Posts: 1,737
Joined: Feb 2009
Thanks was just making sure I was doing it right
ACappo's Avatar ACappo
01:06 AM Liked: 44
post #2357 of 2595
05-21-2014 | Posts: 141
Joined: Jun 2013
Quote:
Originally Posted by Iron Mike View Post

what is a "dynamic profile" and what is the name and model of this mysterious tv ? smile.gif

im in a bit of a rush but will update tonight meanwhile please find attached data of the validation set
1085.xlsx 1010k .xlsx file
Attached: 1085.xlsx (1,009.6 KB) 
Iron Mike's Avatar Iron Mike
03:30 AM Liked: 362
post #2358 of 2595
05-21-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by ACappo View Post

im in a bit of a rush but will update tonight meanwhile please find attached data of the validation set
1085.xlsx 1010k .xlsx file

THANK YOU smile.gif

you are one of the first to provide actual validation data of a CM LUT running a large custom patch set in CM...

alright, 1087pt custom patch set

dE 2000
avg: 1.075503039
max: 6.8854
points with dE > 1.5: 233
points with dE > 2.0: 124
points with dE > 2.5: 75
points with dE > 3.0: 54

dE 2000 luminance compensated
avg: 1.132358748
max: 8.1761
points with dE > 1.5: 259
points with dE > 2.0: 150
points with dE > 2.5: 99
points with dE > 3.0: 77


- M
Holiday121's Avatar Holiday121
03:37 AM Liked: 39
post #2359 of 2595
05-21-2014 | Posts: 1,737
Joined: Feb 2009
I am so close to pulling the trigger on Calman 5.

Eh
ConnecTEDDD's Avatar ConnecTEDDD
06:54 AM Liked: 515
post #2360 of 2595
05-21-2014 | Posts: 2,441
Joined: Sep 2010
Quote:
Originally Posted by Iron Mike View Post

Quote:
Originally Posted by ACappo View Post

im in a bit of a rush but will update tonight meanwhile please find attached data of the validation set
1085.xlsx 1010k .xlsx file

THANK YOU smile.gif

you are one of the first to provide actual validation data of a CM LUT running a large custom patch set in CM...

alright, 1087pt custom patch set

dE 2000
avg: 1.075503039
max: 6.8854
points with dE > 1.5: 233
points with dE > 2.0: 124
points with dE > 2.5: 75
points with dE > 3.0: 54

dE 2000 luminance compensated
avg: 1.132358748
max: 8.1761
points with dE > 1.5: 259
points with dE > 2.0: 150
points with dE > 2.5: 99
points with dE > 3.0: 77


- M

Hello ACappo, how many color points did you calibrated with CalMAN to have this report?

Can you re-measure these 1085 Patches without correction LUT Active?
zoyd's Avatar zoyd
07:41 AM Liked: 434
post #2361 of 2595
05-21-2014 | Posts: 5,041
Joined: Sep 2006
Quote:
Originally Posted by ConnecTEDDD View Post

Hello ACappo, how many color points did you calibrated with CalMAN to have this report?

Can you re-measure these 1085 Patches without correction LUT Active?

I was going to ask the same thing, there is a trend for high dE points at lower saturations so it would be good to know what the pre-LUT values are for these points. I've found that a saturation density of 9 (17/9 is what I typically use) provides good coverage in combination with DLC, and the DLC threshold set to 1.5

Out of curiosity I ran the same test set ACappo used through my RC1 calibrated eecolor memory slot to compare to what I see using my typical random set and the statistics were very close to each other.

1000 Random

Avg= 0.69
Max= 2.60
# > 1.0 = 139
# > 2.0 = 4
# > 3.0 = 0

1086 set

Avg=0.64
Max=2.76
# > 1.0 = 136
# > 2.0 = 5
# > 3.0 = 0
1forsnow's Avatar 1forsnow
01:51 PM Liked: 41
post #2362 of 2595
05-21-2014 | Posts: 189
Joined: Mar 2010
I can semi add to this. The screenshot below is from the latest version of calman. I have a kuro so the blue is undersaturated, so the spike in errors if from the color points that are undersaturated. That can clearly be seen in the charts. As a confirmation i suppose I can create a custom colorspace to move the blue point up to remove the saturation error in the charts. However that wont do anything for me but give me better looking charts.


ACappo's Avatar ACappo
02:19 PM Liked: 44
post #2363 of 2595
05-21-2014 | Posts: 141
Joined: Jun 2013
Okay let's try:)

My test display is an older samsung LN46A560. It has issues with undersaturated red and green and at low luminance it desaturate colors (mainly red)

To avoid the undersaturated red/green did use dynamic mode (i have the options of standard/movie/dynamic) I know this is my far the worst mode for calibration, since it turn blue up in the grey scale without the option to fix it (advanced menu is disabled) so pre calibration the picture is as bad as possible. I will try again in standard this weekend.

I used CalMAN's detailed option when starting the cube calibration, I unfortunately I didn't Check the total number of readings

I will recalibrate this weekend, doing a full pre calibration patch set.
Iron Mike's Avatar Iron Mike
02:22 PM Liked: 362
post #2364 of 2595
05-21-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by 1forsnow View Post

I can semi add to this. The screenshot below is from the latest version of calman. I have a kuro so the blue is undersaturated, so the spike in errors if from the color points that are undersaturated. That can clearly be seen in the charts. As a confirmation i suppose I can create a custom colorspace to move the blue point up to remove the saturation error in the charts. However that wont do anything for me but give me better looking charts.


what are the stats for this ?

Charts like this can look "better" because of resolution of the point and the target area, that's why they don't mean much...

what are the numbers ?
Iron Mike's Avatar Iron Mike
02:24 PM Liked: 362
post #2365 of 2595
05-21-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by ACappo View Post

Okay let's try:)

My test display is an older samsung LN46A560. It has issues with undersaturated red and green and at low luminance it desaturate colors (mainly red)

To avoid the undersaturated red/green did use dynamic mode (i have the options of standard/movie/dynamic) I know this is my far the worst mode for calibration, since it turn blue up in the grey scale without the option to fix it (advanced menu is disabled) so pre calibration the picture is as bad as possible. I will try again in standard this weekend.

I used CalMAN's detailed option when starting the cube calibration, I unfortunately I didn't Check the total number of readings

I will recalibrate this weekend, doing a full pre calibration patch set.

not sure what exactly "dynamic mode" does on this display, but u def want to disable any "dynamic" behavior... display needs to be set up as good as it can be in pre-profile and then remain stable...

"dynamic mode" does not sound good to me.... eek.gif
ACappo's Avatar ACappo
02:48 PM Liked: 44
post #2366 of 2595
05-21-2014 | Posts: 141
Joined: Jun 2013
Its what is also known as torch mode
1forsnow's Avatar 1forsnow
03:07 PM Liked: 41
post #2367 of 2595
05-21-2014 | Posts: 189
Joined: Mar 2010
Quote:
Originally Posted by Iron Mike View Post

what are the stats for this ?

Charts like this can look "better" because of resolution of the point and the target area, that's why they don't mean much...

what are the numbers ?

I am not sure what you mean mike? The average for 1000 points is .5272 with the highest error at 2.1335. The delta graph at the bottom shows this. Is this not good?
Iron Mike's Avatar Iron Mike
03:24 PM Liked: 362
post #2368 of 2595
05-21-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by 1forsnow View Post

I am not sure what you mean mike? The average for 1000 points is .5272 with the highest error at 2.1335. The delta graph at the bottom shows this. Is this not good?

yeah, for whatever reason the image does not enlarge when I click on it... it's hard to tell what the numbers are on the Y axis... so it's dE 1, 2, 3 ?

Edit: got it, had to restart Firefox to get it to work ! smile.gif
Iron Mike's Avatar Iron Mike
03:41 PM Liked: 362
post #2369 of 2595
05-21-2014 | Posts: 799
Joined: Oct 2006
Quote:
Originally Posted by 1forsnow View Post

I am not sure what you mean mike? The average for 1000 points is .5272 with the highest error at 2.1335. The delta graph at the bottom shows this. Is this not good?

so, can you elaborate on your validation set up ?

Is this a custom patch set or a 10^3 grid ? did you use an external patch generator, HDMI Out or Ted's disk via Blu-ray ?

- M
1forsnow's Avatar 1forsnow
05:06 PM Liked: 41
post #2370 of 2595
05-21-2014 | Posts: 189
Joined: Mar 2010
Quote:
Originally Posted by Iron Mike View Post

so, can you elaborate on your validation set up ?

Is this a custom patch set or a 10^3 grid ? did you use an external patch generator, HDMI Out or Ted's disk via Blu-ray ?

- M

I used this:



i have to admit, i am a rookie with this patch generator, so i wanted to start with the basics. I used hdmi out, but I have verified its accuracy against a dvdo tpg which calman now has triplet support for.
Once i get more familiar with the patch generator, i will expand my patch options.

Reply Display Calibration

Subscribe to this Thread

Powered by vBadvanced CMPS v3.2.3