Another BT.1886 vs. Power gamma question - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 7Likes
Reply
 
Thread Tools
post #1 of 40 Old 03-16-2019, 08:44 PM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Another BT.1886 vs. Power gamma question

I have been calibrating with Calman (and at least for a bit longer but to be seen) just for myself a long time and used to watching calibrated displays.

I am familiar with BT.1886 and how at a zero input or measured black point it approximates power gamma 2.4 and how it comes out of black slower(edit: Faster with a non-zero black point...so brighter quicker per results below) than power gamma.

My question is just to understand my results, not asking which is better as that is subjective and I think I have read every thread on that over the years.

I have been watching my Sony XBR75X940D with a nice 2 point relative calibration for a while but finally took the time to do a 10 point calibration.

With FALD off and BT.1886 measured black point with Delta E under 1 my result looked decidedly thin to me. I have had good results with BT.1886 and power gamma in the past on other displays.

I then tried power 2.3 (as a compromise between 2.2 and 2.4 as most viewing is in a dim room but wanted to avoid black crush). Got delta E under 1 and of course a much richer look.

So hey, done, view and enjoy! But here is my question.

Most reviews show that display measures around .05 cd/m2 with FALD on. Ironically if I use that as my input black point for BT.1886 the curve is not too far off from my 2.3 power gamma. In the future should I just use the displays true black point with FALD on as my BT.1886 target for calibrating with FALD off? I know my BT.1886 with measured black point assumed my set did not get that black and set the curve accordingly. And it had me push luminance in calibration to match.

A related question. With power gamma 2.3 target 10 IRE was around 2.0 and of course by 30IRE was at 2.3. My average was 2.22 and black level chart is dead in. This would not crush blacks would it?

Thanks for your thoughts,
BJBBJB

Last edited by BJBBJB; 03-17-2019 at 07:51 AM.
BJBBJB is offline  
Sponsored Links
Advertisement
 
post #2 of 40 Old 03-17-2019, 01:19 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,728
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 722 Post(s)
Liked: 1045
Quote:
Originally Posted by BJBBJB View Post
...I am familiar with BT.1886 and how at a zero input or measured black point it approximates power gamma 2.4 and how it comes out of black slower than power gamma...
That is not correct.
With a zero black a BT1886 gamma will be a pure 2.4 power law.
With a raised black display BT1886 will come out of black faster, lifting the shadows.
See: https://www.lightillusion.com/error.html#bt1886_gamma

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #3 of 40 Old 03-17-2019, 07:47 AM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Quote:
Originally Posted by Light Illusion View Post
That is not correct.
With a zero black a BT1886 gamma will be a pure 2.4 power law.
With a raised black display BT1886 will come out of black faster, lifting the shadows.
See: https://www.lightillusion.com/error.html#bt1886_gamma

Steve
Steve,
Sorry about that, got it backwards. Coming out of black quicker does mean lifting shadows which is what I meant to say. Brighter quicker.
I will add comment so readers not mislead.

Any thoughts on my actual questions as it relates to that raised black level with dimming on vs. off and the results and crushed blacks with the gamma I have?
BJBBJB
Bruce2019 likes this.

Last edited by BJBBJB; 03-17-2019 at 07:55 AM.
BJBBJB is offline  
Sponsored Links
Advertisement
 
post #4 of 40 Old 03-18-2019, 10:29 AM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
My personal opinion on gamma is that it all depends on what is happening in the low end.. Zero to 20%... I am not saying, ignore the rest, but more or less...


The human eye does not work linear, it works in a way that 50% of the brightness (aka of a tv screen) is 18% of the lumince of the screen... So in those lowest 20% there is everything going on what makes a picture appear to dark, washed out, crushed, overexposed.. That is not happening at 30% and on top of that.. On top of that.. flat is totally fine.. Chose 2.2, 2.3.2.4 what ever you like and/ or wherever your screen is placed..


So no matter how accurate you follow a gamma curve.. Just a slight variation in that dark area has a huge impact..


If you like the flat gamma more then the BT1886 Curve that gives you a clear indication that this curve is over brightening the dark parts too much.. No brainer? Right? But brightening the dark parts is the whole point of BT1886, but if it is just a bit too much, you get this "flat, less pop" result.. So this is just a delicate balance thing...


So just take your 2.3 flat gamma and give it a slight boost at 20%, and a bit larger boost at 10% that usually will do the job.. And then is the big question, what will the panel do at 5%.. will it smoothly follow the curve or behave otherwise..


Hopefully one day they will built in a low end gamma control, that is working like a low end high pass filter.. where you can easily bend that low end curve.. (in a more analog way)
Attached Thumbnails
Click image for larger version

Name:	recommend_gammas.jpg
Views:	44
Size:	33.1 KB
ID:	2540846  

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner

Last edited by Bruce2019; 03-18-2019 at 10:48 AM.
Bruce2019 is offline  
post #5 of 40 Old 03-18-2019, 10:44 AM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
And one comment about black level..


Flat power gamma is ok if you have zero black AND infinite contrast panel.. This is a very theoretical thing..


Personally that black level only makes a difference in a complete dark room..


With just slight ambient light.. you will not see a difference in black level.. Proof: 0,1 nits/cd black level vs. 0,013.. You dont see it with a dim light bulb in a room..It does not matter..


Here is my gamma that I have on my old S10 Plasma tv.. and I calibrated the gamma next to a sony trinitron crt.. And this was an eye opener for me, why gamma was never an issue with those CRTs.. There was not even a gamma control, it was just perfect..


Now I have not messured the gamma on the CRT, but of the plasma after eye balling it in a long session vs the CRT.. No way these CRTs have a flat gamma.. They have 2.3 or 2.4 gamma, but not in the low end.. Shadow detail was phenomenal on the Trinitron (in cinema mode there was even the brightness setting locked), but just as good on the plasma, after adjusting it the same way...
Attached Thumbnails
Click image for larger version

Name:	IMG_20190313_210421.jpg
Views:	75
Size:	1.01 MB
ID:	2540838   Click image for larger version

Name:	IMG_20190316_202308.jpg
Views:	53
Size:	304.2 KB
ID:	2540840   Click image for larger version

Name:	IMG_20190316_204213.jpg
Views:	49
Size:	416.0 KB
ID:	2540842   Click image for larger version

Name:	IMG_20190316_204551.jpg
Views:	51
Size:	790.5 KB
ID:	2540844   Click image for larger version

Name:	recommend_gammas.jpg
Views:	39
Size:	33.1 KB
ID:	2540848  


Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner
Bruce2019 is offline  
post #6 of 40 Old 03-18-2019, 10:51 AM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
Quote:
Originally Posted by BJBBJB View Post
Steve,
Sorry about that, got it backwards. Coming out of black quicker does mean lifting shadows which is what I meant to say. Brighter quicker.
I will add comment so readers not mislead.

Any thoughts on my actual questions as it relates to that raised black level with dimming on vs. off and the results and crushed blacks with the gamma I have?
BJBBJB

A flat (dark) gamma will crush blacks more then a BT1886 curve..


Gamma 2.0 in the dark and going up to 2.3 at 30% will NOT crush blacks.. It might be too much.. So you could try going from 2.1 to 2.3.. The brighter your room, the more brightness boost you can adjust in that low end..

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner
Bruce2019 is offline  
post #7 of 40 Old 03-18-2019, 10:59 AM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
Think of the low end gamma, as some kind of base boost controle..


You increase it so that you can see the dark detail (it sounds better)


But if you boost it too much.. the result will suffer.. You have to find the sweet spot..

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner
Bruce2019 is offline  
post #8 of 40 Old 03-18-2019, 11:06 AM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Thanks to all! A huge help.

Something Bruce 2019 said reminded me to mention that I am not using a totally flat power gamma curve with black input at zero which is a flat line. I am using the measured black" setting for my power gamma curve set at 2.3 with FALD off. So same as BT.1886. Not sure if that was assumed.

I totally agree with the outsize impact of 10IRE to 30IRE. I will have to look at this more closely but not sure how much ability I have to keep good color balance and play with the luminanace there. But definitely plan to look at all the good advice here regarding that area.

Thanks!
BJBBJB
BJBBJB is offline  
post #9 of 40 Old 03-18-2019, 11:33 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,728
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 722 Post(s)
Liked: 1045
Any Gamma value should relative to the display's black point, or it will clip.
And I mean clip, not crush.
So what ever is your display's actual black level should become the 'zero' for the gamma setting.

Actually, that is one issue with PQ based HDR, as it is an absolute standard, so a black roll-off has to be used to prevent clipping, but that will then inevitably introduce some 'crush' - worse as the black point rises.
(And some HDR TVs don't implement a black roll-off, so do clip...)

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #10 of 40 Old 03-18-2019, 11:49 AM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
Quote:
Originally Posted by Light Illusion View Post
Any Gamma value should relative to the display's black point, or it will clip.
And I mean clip, not crush.
So what ever is your display's actual black level should become the 'zero' for the gamma setting.

Actually, that is one issue with PQ based HDR, as it is an absolute standard, so a black roll-off has to be used to prevent clipping, but that will then inevitably introduce some 'crush' - worse as the black point rises.
(And some HDR TVs don't implement a black roll-off, so do clip...)

Steve

What will clip.. Please explain..

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner
Bruce2019 is offline  
post #11 of 40 Old 03-18-2019, 11:52 AM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
Quote:
Originally Posted by BJBBJB View Post
Thanks to all! A huge help.

Something Bruce 2019 said reminded me to mention that I am not using a totally flat power gamma curve with black input at zero which is a flat line. I am using the measured black" setting for my power gamma curve set at 2.3 with FALD off. So same as BT.1886. Not sure if that was assumed.

I totally agree with the outsize impact of 10IRE to 30IRE. I will have to look at this more closely but not sure how much ability I have to keep good color balance and play with the luminanace there. But definitely plan to look at all the good advice here regarding that area.

Thanks!
BJBBJB

Just to be clear..


Dont touch 30%.. chose a flat gamma from 30% to 100% 2.3 is a great gamma.. CRTs had 2.25.. it works for day and night..


Zero to 20% that is your focus.. remember.. More then 50% of the eyes brightness information is there..

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner
Bruce2019 is offline  
post #12 of 40 Old 03-18-2019, 12:19 PM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,728
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 722 Post(s)
Liked: 1045
Black detail will clip, as the gamma is targeting below the black level the display can actually reach.

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #13 of 40 Old 03-18-2019, 12:50 PM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Quote:
Originally Posted by Light Illusion View Post
Any Gamma value should relative to the display's black point, or it will clip.
And I mean clip, not crush.
So what ever is your display's actual black level should become the 'zero' for the gamma setting.

Actually, that is one issue with PQ based HDR, as it is an absolute standard, so a black roll-off has to be used to prevent clipping, but that will then inevitably introduce some 'crush' - worse as the black point rises.
(And some HDR TVs don't implement a black roll-off, so do clip...)

Steve
Steve,
I always enjoy your posts and knowledge as I have read them over the years. Thanks for your input.
.
So if my display's actual black level should become the "zero" for gamma setting......and I calibrate with FALD dimming off and clearly when reading 0 IRE that is not what my 0 IRE looks like when watching actual content with FALD dimming on (I can tell that with my eyes, I don't need a meter for that)....

Does this mean that for both Power 2.3 or BT.1886 I should be inputting (vs. measuring) my black level to what my set is really displaying for black with FALD on (most reviews put it around .05-.06 cd/m2)? This would be lower than as measured with FALD off which is what I was trying to understand.

Thanks again,
BJBBJB
BJBBJB is offline  
post #14 of 40 Old 03-18-2019, 12:52 PM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Quote:
Originally Posted by Bruce2019 View Post
Just to be clear..


Dont touch 30%.. chose a flat gamma from 30% to 100% 2.3 is a great gamma.. CRTs had 2.25.. it works for day and night..


Zero to 20% that is your focus.. remember.. More then 50% of the eyes brightness information is there..
Bruce 2019,
Got it. Thanks!

BJBBJB
Bruce2019 likes this.
BJBBJB is offline  
post #15 of 40 Old 03-18-2019, 03:38 PM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
Quote:
Originally Posted by Light Illusion View Post
Black detail will clip, as the gamma is targeting below the black level the display can actually reach.

Steve

That is black crush.. loss of darkest details.. (in theorie)


But that does not happen.. a) difference in black level is so small, there is no clipping or loss in anything.. 2) the Black level is the black level.. the tvs start their gamma at this point.. It is the mathematical starting point of picture information.. The picture will show what it gets Black is black..and above black is above black.. A tv with a bad black level does not have any different behavior above black that crushes details..

Gamma is a way to describe the increasing of brightness.. If you feed 2 different tvs with different black levels with the same gamma.. what will happen? One will appear a bit brighter then the other.. There is no black clipping/black crush because of gamma..


Proof: 0,1 nits vs. 0,013 nits black level, gamma behaves 100% the same..
Attached Thumbnails
Click image for larger version

Name:	IMG_20190313_210908.jpg
Views:	38
Size:	957.1 KB
ID:	2541052   Click image for larger version

Name:	IMG_20190313_210559.jpg
Views:	39
Size:	703.2 KB
ID:	2541054  

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner

Last edited by Bruce2019; 03-18-2019 at 04:12 PM.
Bruce2019 is offline  
post #16 of 40 Old 03-18-2019, 03:49 PM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
The Zero Black level and unlimited contrast is a theoretical mathematic therm, only then you should use a flat gamma 2.4.. No real scenario.. Ignore it..



Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner

Last edited by Bruce2019; 03-18-2019 at 04:14 PM.
Bruce2019 is offline  
post #17 of 40 Old 03-18-2019, 05:05 PM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Quote:
Originally Posted by Bruce2019 View Post
A flat (dark) gamma will crush blacks more then a BT1886 curve..

Gamma 2.0 in the dark and going up to 2.3 at 30% will NOT crush blacks.. It might be too much.. So you could try going from 2.1 to 2.3.. The brighter your room, the more brightness boost you can adjust in that low end..
After pulling up a gamma chart of my data that showed more lines, I show 10 IRE at 2, 20 IRE at 2.2 and 30 IRE somewhere around 2.25. I really don't have any more adjustment range to make 0 to 20 IRE any darker i.e.higher gamma number. I would have to start over with different TV gamma, bias settings to get that if it is possible.

When I looked at the BT.1886 curve with my as measured black level, it had me taking 10 IRE to 1.75 20 IRE to 1.9 and 30 IRE to 2. No wonder it looked thin! But that is because that black level is so high compared to the reality with dimming on.

Stupid question. The zero measured IRE does not impact the actual gamma curve of 10, 20, 30 IRE, just the target correct?

Thanks again,
BJBBJB
BJBBJB is offline  
post #18 of 40 Old 03-18-2019, 05:15 PM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
Quote:
Originally Posted by BJBBJB View Post
After pulling up a gamma chart of my data that showed more lines, I show 10 IRE at 2, 20 IRE at 2.2 and 30 IRE somewhere around 2.25. I really don't have any more adjustment range to make 0 to 20 IRE any darker i.e.higher gamma number. I would have to start over with different TV gamma, bias settings to get that if it is possible.

When I looked at the BT.1886 curve with my as measured black level, it had me taking 10 IRE to 1.75 20 IRE to 1.9 and 30 IRE to 2. No wonder it looked thin! But that is because that black level is so high compared to the reality with dimming on.

Stupid question. The zero measured IRE does not impact the actual gamma curve of 10, 20, 30 IRE, just the target correct?

Thanks again,
BJBBJB

Usually the chaning settings just change the curve where you target.. not the actual reading..


Can you just chose flat gamma 2.2 , make a reading and post a picture of the result.. That will clear things up..

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner
Bruce2019 is offline  
post #19 of 40 Old 03-19-2019, 12:27 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,728
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 722 Post(s)
Liked: 1045
Sorry, clipping is clipping, crushing is, well, crushing.
You need to understand the difference, and why the application of a gamma curve as absolute vs. relative will cause clipping.

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #20 of 40 Old 03-19-2019, 08:05 AM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Quote:
Originally Posted by Bruce2019 View Post
Usually the chaning settings just change the curve where you target.. not the actual reading..


Can you just chose flat gamma 2.2 , make a reading and post a picture of the result.. That will clear things up..

I will strive to do that when back in front of display and laptop. Will be a few days.
Thanks,
BJBBJB
BJBBJB is offline  
post #21 of 40 Old 03-19-2019, 03:37 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,286
Mentioned: 118 Post(s)
Tagged: 0 Thread(s)
Quoted: 4364 Post(s)
Liked: 1604
Quote:
Originally Posted by Bruce2019 View Post
That is black crush.. loss of darkest details.. (in theorie)


But that does not happen.. a) difference in black level is so small, there is no clipping or loss in anything.. 2) the Black level is the black level.. the tvs start their gamma at this point.. It is the mathematical starting point of picture information.. The picture will show what it gets Black is black..and above black is above black.. A tv with a bad black level does not have any different behavior above black that crushes details..

Gamma is a way to describe the increasing of brightness.. If you feed 2 different tvs with different black levels with the same gamma.. what will happen? One will appear a bit brighter then the other.. There is no black clipping/black crush because of gamma..


Proof: 0,1 nits vs. 0,013 nits black level, gamma behaves 100% the same..
A pure gamma curve start at 0, not at the elevated black level of the TV. If you calibrate white to 100 nits, the signal at 50% input will be at 22 nits, regardless of the black level of the TV. It follows that a TV with 0.1 nits black level cannot be following a pure gamma curve below 5%, as it will require the signal to go below the black level.
Bruce2019 likes this.

Last edited by Dominic Chan; 03-19-2019 at 07:15 PM.
Dominic Chan is online now  
post #22 of 40 Old 03-19-2019, 07:09 PM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
I have to admit that I was wrong.. I used HCFR heavily today and examined the BT1886 functions and options and I realise that I was using it wrong.. and that the experts here where right.. BT1886 takes the black level performance of the screen into account..


The Calibration software will plot your own BT1886 gamma curve that fits perfect to your display (and also to your gamma that you chose, you can chose for example a gamma 2.0 curve for a bright room or 2.4 for a dark room, or 2.3)


A short explanation and I might have an idea why you did not get the right gamma curve with your input black level..


At least with HCFR, the software needs following information: The Black level, the effective input offset at 50% (that is your actual gamma at 50%,) and the max light output.. In my case it is somehow fixed to 120 nits...


The software will then plot you a gamma curve for your display with your specific black level, and your gamma at 50% stimulus.. So you should just make one gamma reading with your prefered gamma and then you have all the information.. Black level and at what gamma your tv ends up in the 50% area and above..


In my case there was also a trick thing with the black level.. I had to type 0.1 rather then 0,1 to get the number accepted..


https://forum.doom9.org/showthread.php?t=173532


With my plasma and a really bad black level I could not use my chosen darkest gamma, so I went over to the stock 2.2 setting.. With a darker gamma I would have needed to increase the darkest parts much more, but this set has no detailed gamma controls..


Very interesting.. and also very helpfull, because the BT1186 shows you the way, you can try different gammas and you see where you end up, what you need to change and maybe if you just cant get close with this stock gamma..


www.spectracal.com/Documents/White%20Papers/BT.1886.pdf
Attached Thumbnails
Click image for larger version

Name:	IMG_20190320_024255.jpg
Views:	33
Size:	390.1 KB
ID:	2541646  

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner

Last edited by Bruce2019; 03-19-2019 at 07:13 PM.
Bruce2019 is offline  
post #23 of 40 Old 03-20-2019, 01:23 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,728
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 722 Post(s)
Liked: 1045
With regard to BT1886 I would red this, as posted previously: https://www.lightillusion.com/error.html#bt1886_gamma
The shadow lift really is not good on most displays.

Also, display gamma should be relative, no absolute, as clipping would be an issue on all TVs, other than those with near zero blacks.
And clipping is not good!
BT1886 will not correct the clipping if the gamma is applied absolute, and will still lift the shadows.

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #24 of 40 Old 03-20-2019, 03:33 AM
AVS Forum Special Member
 
ConnecTEDDD's Avatar
 
Join Date: Sep 2010
Location: Athens, Greece
Posts: 7,744
Mentioned: 178 Post(s)
Tagged: 1 Thread(s)
Quoted: 2910 Post(s)
Liked: 3545
The black level of the display is mapped to Zero and the white level mapped to One, then the gamma curve applied between the two.

Therefor there should never be clipping, regardless of the gamma used.

If the contrast ration of the display is also very low, as the black level is high, that is a separate issue.

But its all related with about how the content is being mastered, if its mastered with 2.4 gamma and based to your BT1886 calibration you will have 2.0 @ 10% Gray for example, then you will be able to see details you should not have to see, so image will look very wash-out, flat and with less depth.

Currently, studios are using Gamma 2.4 ...about 2016 and later.

Before about 2016 it was 2.35

Before about 2014 it was 2.2

If you watch a movie which been released at 2012 @ Blu-Ray, with 2.4 gamma now, it may look a bit dark.

Ideally you need a secondary picture mode calibrated with different gamma, to give you flexibility when you are watching various content (older releases).

When you have zero black (OLED) with BT.1886 then it has the exact same luma targets as Power Law Gamma 2.4
Bruce2019 and Jin-X like 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 #25 of 40 Old 03-20-2019, 05:50 AM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
Quote:
Originally Posted by ConnecTEDDD View Post
The black level of the display is mapped to Zero and the white level mapped to One, then the gamma curve applied between the two.

Therefor there should never be clipping, regardless of the gamma used.

If the contrast ration of the display is also very low, as the black level is high, that is a separate issue.

But its all related with about how the content is being mastered, if its mastered with 2.4 gamma and based to your BT1886 calibration you will have 2.0 @ 10% Gray for example, then you will be able to see details you should not have to see, so image will look very wash-out, flat and with less depth.

Currently, studios are using Gamma 2.4 ...about 2016 and later.

Before about 2016 it was 2.35

Before about 2014 it was 2.2

If you watch a movie which been released at 2012 @ Blu-Ray, with 2.4 gamma now, it may look a bit dark.

Ideally you need a secondary picture mode calibrated with different gamma, to give you flexibility when you are watching various content (older releases).

When you have zero black (OLED) with BT.1886 then it has the exact same luma targets as Power Law Gamma 2.4



1) That explains a lot.. I see also differences in older video games.. they are "darker" then modern ones.. Modern computer games almost always have a built in "gamma" adjustment setting when you install the game.. These guys that develop those games for sure work on tons of different displays while programming and they are aware of the problem.. So usually the game starts with a black level / dark detail test pattern where you can adjust the video output to the screen.. This is pretty smart thinking there..


2) I placed my Sony trinitron CRT next to my plasma.. Just to get an idea how the CRT looks.. And there is no way this is a 2.2 gamma.. it is much darker.. I had to put my Plasma in the lowest gamma setting, which was messured around 2.3 to 2.4.. but the sony had incredible good shadow detail, so I had to work with the Plasma on increasing brightness in the dark area.. and voila.. A nice BT1886 curve.. These CRTs had it.. No way they have a flat gamma..


3) So I ask myself when and why did this flat gamma obsession even started..when CRTs did not and dont behave that way..

The International Telecommunication Union seem to spottet that problem and introduced the CRT like behavior BT1886 gamma in 2011 (!) But in every review where they calibrated displays, every tv shootout even years later, and even today, they all calibrated to flat gammas.. Was nobody listening?
Attached Files
File Type: pdf BT.1886.pdf (149.5 KB, 5 views)

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner
Bruce2019 is offline  
post #26 of 40 Old 03-20-2019, 06:24 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,286
Mentioned: 118 Post(s)
Tagged: 0 Thread(s)
Quoted: 4364 Post(s)
Liked: 1604
Quote:
Originally Posted by ConnecTEDDD View Post
The black level of the display is mapped to Zero and the white level mapped to One, then the gamma curve applied between the two.
With power law gamma a display's output is characterized by

Vout=A*Vin^gamma. Strictly speaking, for "pure gamma" there is no mapping of black level to a display's zero, even though it's common to apply black compensation (which modifies the EOTF).

Unfortunately the term "gamma" is often used to mean EOTF in a broader sense, which can cause confusion.

Last edited by Dominic Chan; 03-20-2019 at 06:47 AM.
Dominic Chan is online now  
post #27 of 40 Old 03-20-2019, 06:54 AM
AVS Forum Special Member
 
Light Illusion's Avatar
 
Join Date: Aug 2010
Posts: 1,728
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 722 Post(s)
Liked: 1045
All LightSpace gamma calculations are relative, except for PQ HDR, as that is an absolute standard.
So a power law gamma is a pure power law, applied from the actual black to the actual white of the display.

To do anything else would always cause black clipping on any display with a black that is higher than zero.

Steve

Steve Shaw
LIGHT ILLUSION

Light Illusion is offline  
post #28 of 40 Old 03-23-2019, 08:16 PM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Quote:
Originally Posted by Bruce2019 View Post
Usually the chaning settings just change the curve where you target.. not the actual reading..


Can you just chose flat gamma 2.2 , make a reading and post a picture of the result.. That will clear things up..
I re-ran a 21 point grayscale/gamma reading. The first attachment is vs. the 2.2 power gamma target as requested. The second is against the 2.3 power gamma target that I calibrated to. Black level is measured black level with local dimming off. The overall average delta E was under 1 and zero through 20 was 1.5 or under vs. the Power gamma 2.3 target. Best blacks I have seen with this display.

Hope that helps!
Thanks,
BJBBJB
Attached Thumbnails
Click image for larger version

Name:	2-point-2-power-gamma-target.jpg
Views:	30
Size:	16.1 KB
ID:	2543432   Click image for larger version

Name:	2point3PowerGamma-Target-as-calibrated.jpg
Views:	29
Size:	15.9 KB
ID:	2543434  
BJBBJB is offline  
post #29 of 40 Old 03-24-2019, 08:27 AM
Advanced Member
 
Bruce2019's Avatar
 
Join Date: May 2013
Location: Vienna
Posts: 968
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 512 Post(s)
Liked: 288
The human eye sees 18% brightness as 50% brightness.. so in that low 20% there is almost everything going on, that makes the picture pop, or flat looking.. This is where the gamma makes the difference.. Not saying the brighter parts are not important, but much less...


And in that low end, you are close, but still too bright.. Just a slight over brightening makes a huge difference..

Panasonic 42inch ST60 owner
Panasonic 42inch S10 ES (Europe/Silver) owner
Bruce2019 is offline  
post #30 of 40 Old 03-24-2019, 08:58 AM - Thread Starter
Senior Member
 
BJBBJB's Avatar
 
Join Date: May 2002
Posts: 393
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 121 Post(s)
Liked: 36
Quote:
Originally Posted by Bruce2019 View Post
The human eye sees 18% brightness as 50% brightness.. so in that low 20% there is almost everything going on, that makes the picture pop, or flat looking.. This is where the gamma makes the difference.. Not saying the brighter parts are not important, but much less...


And in that low end, you are close, but still too bright.. Just a slight over brightening makes a huge difference..
Bruce2019,

Thanks for the feedback. I see that I am a bit bright at 10IRE and 20IRE so clearly not crushing, if anything a bit too bright is what you are saying? Assuming a power 2.3 target.

My 10 and 20 IRE cuts are at minimum for some colors so guess I have to live with that miss there. Having tried various overall cut/gain/TV gamma settings, this is the best balance of grayscale and luminance accuracy at the low end getting me to this result of around 1.5 or under from 10IRE to 20IRE.

This does remind me of one nagging question. We are talking about black levels etc but the chart we are looking at (vs. what I am seeing when I am watching the display) has 0 IRE significantly different than when I am watching the display as 0 IRE with dimming off is much darker (viewing) than with on (calibrating). The chart is starting at .7 cd/m2 ish while my set with dimming off is around .05 or .06 cd/m2.

I understand dimming needs to be off so gamma is stable, but wonder what it means to the dark end of the gamma results? Or is it irrelevant due to the fact that by definition, it impacts zero IRE?

Looking forward to the fun of using my HDFury Linker for my first HDR10 calibration next!

Thanks,
BJBBJB
BJBBJB 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