***Official B/C/E/G6P OLED Calibration Thread - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 527Likes
Reply
 
Thread Tools
post #1 of 1720 Old 11-15-2016, 02:31 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
***Official B/C/E/G6P OLED Calibration Thread

I came here expecting to find an official calibration thread for the 2016 LG OLEDs but after scanning through three pages of thread titles, couldn't't find anything.

If there is an official 2016 OLED calibration thread, I'd appreciate a pointer.

If the 2016 OLEDs are do good out-of-the-box that nothing but basic calibration of brightness and contrast is required, I'd appreciate a heads-up to avoid the time and trouble of diving in further.

And since I've now seen a coulee HDR calibration threads with posters owning 2016 OLEDs, feel free to use this thread as a location to collect your experience and wisdom for the many new 2016 OLED owners who are likely going to be seeking advice on calibration as we get into the Black Friday a Holiday Shopping Period.
SandstormGT likes this.
fafrd is offline  
Sponsored Links
Advertisement
 
post #2 of 1720 Old 11-15-2016, 02:33 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Reserved for whatever reason in case this thread does end up being the official 2016 OLED Calibration Thread.

If there is already an official thread that is being maintained, I will let this one die a quick and quiet death ;-)
fafrd is offline  
post #3 of 1720 Old 11-18-2016, 02:39 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Bumping my own thread in the hopes that any members that have done DIY calibration of these OLEDs can provide helpful info for new owner's seeking expertise & advice.
fafrd is offline  
 
post #4 of 1720 Old 11-18-2016, 06:04 PM
Member
 
Join Date: Feb 2016
Location: Australia
Posts: 171
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 89 Post(s)
Liked: 42
Planning to, but still waiting for the suggested 100 hours before doing it.

LG OLED55C6T 4K HDR TV, Yamaha RX-A2050 Receiver, Q Acoustics 3010, 3050, 3070 and 3090 speakers, Yamaha YST-SW030 rear sub, Sony BDP-S6500 Blu ray, LG UP970 UHD, Fetch TV Mighty, IQ2HD, Apple TV Series 4, Pioneer CLD2730K, eeColor LUT+Lightspace HTL+i1Display Pro and DreamScreen.
vollans is offline  
post #5 of 1720 Old 11-19-2016, 10:31 AM
AVS Forum Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,578
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 539 Post(s)
Liked: 1296
Some general observations:

1. This model year (the B-G6) seem more prone to changes in white balance depending on what was displayed for the few minutes prior to measurement. For instance, say you have the white balance set at 100% and you put up a black screen for a few minutes. If you then go and measure white balance at 100% again, chances are red is going to be a bit higher than it was before. Or, if you set it and then put up a white field or a high APL image for a few minutes and then go back and check, chances are green is going to be a bit high. It's not by a lot; usually it's only about half to one dE, but it can be somewhat confusing if you're not aware of it. Because of this, I like to put up a moving zone plate during any long time gaps to keep things consistent.

2. There is a difference in calibration of HDR mode. The B's code value white balance adjustments start out very fine, whereas the C/E/G's start out very coarse at the low end (code value 127 is so coarse as to be completely unusable) and become finer and more workable as brightness and code value increases. The B's fine adjustments work to it's advantage in the first several code value steps, but they then have too little strength to do any good at the high end. I have found that attempting to calibrate the higher code values on a B may cause problems in tone mapping, so I just concentrate on getting the green emphasis out of the low and mid tones on it.

3. Before calibration in the ISF modes, there is a general tendency for the white balance in all the 2016 models to emphasize green in the low and mid tones and then transition to a lack of green in the brighter levels.

4. The CMS, though improved over some earlier LG OLED models, still does not function very well. Attempting to calibrate it can cause visual problems like blotchy colors and contouring. The problem is, the adjustment only effects a narrow slice of the color spectrum, and the transition from effected color shades to non effected are far too abrupt. Let's say you turn red saturation all the way down. If you have an image that is supposed to smoothly transition from an orangish shade to a pure red, there is absolutely no change until it reaches pure red, and then all of a sudden it becomes black & white because the red saturation is lowered. Ideally, instead of this sharp, abrupt behavior, the control should start having a gradual effect on orange, violet, and pink shades and then smoothly reach it's maximum effect at pure red. I've also seen this behavior on LG LED LCDs.

5. Though windows should be used rather than full fields, there is no significant difference between any measurement window size up to 25%. There is also no significant difference between APL and conventional windows.

6. Like previous LG OLEDs, there is a static brightness limiter. If you leave a test pattern up too long it starts to dim and it's measured characteristics change. Because of this, I prefer to set the 100% and 5% white balance with the 2 point control in real time, and then do the 20 point adjustment pass by pass (not in real time). I can explain this more later.

7. Turning OLED light to 100 and reducing contrast to achieve the desired light output will minimize ABL, but lowering contrast significantly below 85 reduces the bit depth and is not recommended in most situations.

8. After calibrating grayscale and gamma, measuring the SG fleshtones color checker patterns will generally show that skin tones lean towards red. Because of the CMS control's poor behavior, the only way to help this is to move the main tint control a bit toward green. Usually the SG fleshtones will be oversaturated as well, which can be improved by lowering the main color control, though this isn't always the case. I typically end up with the main color and tint at something like 48 and G2.

9. My personal opinion is that the OLEDs do not look natural calibrated to BT1886 (which in this case should be a straight 2.4) gamma. That tends to look too contrasty and lack shadow detail. On the other hand, a straight 2.2 power law gamma can lack depth. CalMAN gives the option of calibrating to a sliding power law gamma, and using this option with a starting point of 2.3-2.35 and specifying a fake black level of .001 to .002 fL depending on the environment and light output produces a very natural and realistic image on these sets.
RIppolito, fafrd, JimP and 17 others like this.

touring ISF/THX calibrator with Jeti 1211 and Klein K-10
Latest reviews:JVC RS500 and 600, JVC RS400, LG 65C7P, JVC RS420
Copied settings measured, older LG OLED calibration tips, Review index

Last edited by Chad B; 11-19-2016 at 11:56 AM.
Chad B is offline  
post #6 of 1720 Old 11-19-2016, 06:59 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Thanks Chad - this is exactly the sort of input helpful to DIY enthusiasts like myself that I was hoping for. Your post has inspired me to break out my meter and get the greyscale/gamma calibrated on my brand-new 65C6P

CMS sounds pretty flaky still and put-of-the-box color looks pretty good to my eyes, so I'm going to focus on greyscale/gamma for now.

What I did on my EF9500 sounds pretty similar to what you are suggesting, but I would appreciate you comments:

1/ I used the BT.1886 spreadsheet to insert an artificial black level of 0.001 (similar to what my old Vizio P70 would deliver) and calculated an ideal BT.1886 with that black level.

2/ I put the OLED in 2.4 (or BT.1886, tried.both and can't remember which ended up working out better).

3/ I then used 21-pt controls to force in that gamma curve

4/ I then used a combination of Brightness control and 5% whitepoint controls to assure that video level 16 (and also pretty much 17) were true black.

I was pretty happy with the shadow-detail resulting from this approach on the EF9500 - any comments as to whether this should be effective and/or other ways to achieve a better result (Ibam using HCFR)???

Really appreciate your contributing what you have learned about getting the most out of these 2016 OLEDs.

Calibration of HDR10 seems pretty straightforward once you have the right patterns.

Do you have any advise for what to do about Dolby Vision (if anything)?
fafrd is offline  
post #7 of 1720 Old 11-22-2016, 04:49 PM
AVS Forum Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,578
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 539 Post(s)
Liked: 1296
The 2016 models are a little less flaky near black than the EF9500, so you probably won't have to play around with the 5% and brightness controls as much.

I usually have to take the green cut down in the 2 point on the new models, which raises low end gamma. Because of that, I have found the 2.2 selection in the TV gives me the characteristics I am looking for near black. The 20 point can then be used to raise the gamma somewhat across the rest of the range.

For instance, if I reduce green cut to -20 or so (may be a lot less on the B series), the brightness with a gamma selection of 2.2 is usually correct at 50-52, but the gamma may be 2.25 to 2.3 at 5%. If I were to do the same thing with a gamma selection of 2.4 in the TV, gamma might be 2.5 at 5%, which is way too high. I then leave 5% alone in the 20 point and adjust from 10-95% (I do the 2 point controls at 5% and 100%, so they are already done).
RIppolito and mo949 like this.

touring ISF/THX calibrator with Jeti 1211 and Klein K-10
Latest reviews:JVC RS500 and 600, JVC RS400, LG 65C7P, JVC RS420
Copied settings measured, older LG OLED calibration tips, Review index
Chad B is offline  
post #8 of 1720 Old 11-22-2016, 05:12 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by Chad B View Post
The 2016 models are a little less flaky near black than the EF9500, so you probably won't have to play around with the 5% and brightness controls as much.

I usually have to take the green cut down in the 2 point on the new models, which raises low end gamma. Because of that, I have found the 2.2 selection in the TV gives me the characteristics I am looking for near black. The 20 point can then be used to raise the gamma somewhat across the rest of the range.

For instance, if I reduce green cut to -20 or so (may be a lot less on the B series), the brightness with a gamma selection of 2.2 is usually correct at 50-52, but the gamma may be 2.25 to 2.3 at 5%. If I were to do the same thing with a gamma selection of 2.4 in the TV, gamma might be 2.5 at 5%, which is way too high. I then leave 5% alone in the 20 point and adjust from 10-95% (I do the 2 point controls at 5% and 100%, so they are already done).
Very helpful Chad - thanks.

I got out my meter last night, loaded up the latest update of HCFR, and started the process of getting back in the 'calibration zone' (been over 9 months ).

This advice will help.

From some quick screwing around, it seems like 2-pt low is actually 15% rather than 20% and 2-pt high is actually 85% rather than 80% - have you seen the same?

Have ever used the BT.1886 setting and it there a reason you think 2.2 works better?

HCFR now allows you to skip the 0% measurement and enter the level to use for black. For night-time viewing, would you suggest a black level or 0.0001 or 0.0002 cd/m2?

I have usually calibrated to 120cd/m2 for dark-room viewing, but after having watched quite a few episodes of Marco Polo in HDR, I actually found the resulting images at Rec.709 @ 120 cd/m2 peak a bit too dark for my taste and may go a bit bright - perhaps 150 cd/m2 or so. Do you have any recommended 100% brightness for those viewing in the dark who want a slightly brighter image?

And finally, in terms of adjusting 100% output, what is your advice as far as Contrast versus OLED light? Should OLED light be set to 100 and then adjust 100% using Contrast, Contrast set to 100 and peak brightness set using OLED light, or some other combination of the two controls.

Thanks in advance.
fafrd is offline  
post #9 of 1720 Old 11-22-2016, 06:19 PM
AVS Forum Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,578
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 539 Post(s)
Liked: 1296
Quote:
Originally Posted by fafrd View Post
Very helpful Chad - thanks.

I got out my meter last night, loaded up the latest update of HCFR, and started the process of getting back in the 'calibration zone' (been over 9 months ).

This advice will help.

From some quick screwing around, it seems like 2-pt low is actually 15% rather than 20% and 2-pt high is actually 85% rather than 80% - have you seen the same?

Have ever used the BT.1886 setting and it there a reason you think 2.2 works better?

HCFR now allows you to skip the 0% measurement and enter the level to use for black. For night-time viewing, would you suggest a black level or 0.0001 or 0.0002 cd/m2?

I have usually calibrated to 120cd/m2 for dark-room viewing, but after having watched quite a few episodes of Marco Polo in HDR, I actually found the resulting images at Rec.709 @ 120 cd/m2 peak a bit too dark for my taste and may go a bit bright - perhaps 150 cd/m2 or so. Do you have any recommended 100% brightness for those viewing in the dark who want a slightly brighter image?

And finally, in terms of adjusting 100% output, what is your advice as far as Contrast versus OLED light? Should OLED light be set to 100 and then adjust 100% using Contrast, Contrast set to 100 and peak brightness set using OLED light, or some other combination of the two controls.

Thanks in advance.
Not sure what you're asking about the 2 point controls being 15% and 85%. I calibrate them at 5% for the lows and 100% for the highs, because the 20 point adjustment goes better if you leave the 5% and 100% adjustments alone. This will leave a big green hump in the middle, but it can be calibrated out with the 20 point.

I have experimented with the BT1886 setting and may have used it once or twice. However, like 2.4, it usually comes out of black too slowly (high gamma at 5%) in my opinion.
2.2 works better because, after adjusting the 2 pt WB and brightness, it gives a gamma of around 2.25-2.3 at 5%, which is closest to the behavior I desire coming out of black. I sometimes do a very mild adjustment of the "adjusting luminance" control only at 5%, but I'd rather not if the TV cooperates.

My preferred method for night time viewing is a 2.35 sliding power with a .001 fL (.0034 cd/m2) manual black level. With a BT1886 formula, I'd increase that a little, to perhaps .005 or .006 cd/m2, or go with a 2.3 power law. You could do one of each and experiment.

I fully agree with the desire to go brighter; I've had similar feelings with these displays. As long as you don't get eyestrain, that is. Keeping in mind my Jeti 1211 may read luminance a bit differently than your meter, I'd suggest a 50 fL (about 170 cd/m2) peak light output in a dark room for those who want a little more punch.

I keep contrast at 85 on the new models and then reduce OLED light to the target peak white output. Reducing contrast much below 85 degrades the bit depth, and 85 gives full WTW headroom for whatever it's worth (not much IMO but might as well if there's no downside).
RIppolito, fafrd, chunon and 2 others like this.

touring ISF/THX calibrator with Jeti 1211 and Klein K-10
Latest reviews:JVC RS500 and 600, JVC RS400, LG 65C7P, JVC RS420
Copied settings measured, older LG OLED calibration tips, Review index
Chad B is offline  
post #10 of 1720 Old 11-23-2016, 09:58 AM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by Chad B View Post
Not sure what you're asking about the 2 point controls being 15% and 85%. I calibrate them at 5% for the lows and 100% for the highs, because the 20 point adjustment goes better if you leave the 5% and 100% adjustments alone. This will leave a big green hump in the middle, but it can be calibrated out with the 20 point.
Great advice - thanks. Will use 5% for low and 100% for high from now on...

Quote:
I have experimented with the BT1886 setting and may have used it once or twice. However, like 2.4, it usually comes out of black too slowly (high gamma at 5%) in my opinion.
2.2 works better because, after adjusting the 2 pt WB and brightness, it gives a gamma of around 2.25-2.3 at 5%, which is closest to the behavior I desire coming out of black. I sometimes do a very mild adjustment of the "adjusting luminance" control only at 5%, but I'd rather not if the TV cooperates.
I've noticed the 'Adjusting Luminance' control (which did not exist on the EF9500) but did not understand what it is for. What is it's function and if the TV is 'not cooperating', what is the mild adjustment at 5% than might be needed?

Quote:
My preferred method for night time viewing is a 2.35 sliding power with a .001 fL (.0034 cd/m2) manual black level. With a BT1886 formula, I'd increase that a little, to perhaps .005 or .006 cd/m2, or go with a 2.3 power law. You could do one of each and experiment.
Helpful, thanks. Both of those levels are brighter than what I have used in the past and I will try those two levels and see what I think.

Quote:
I fully agree with the desire to go brighter; I've had similar feelings with these displays. As long as you don't get eyestrain, that is. Keeping in mind my Jeti 1211 may read luminance a bit differently than your meter, I'd suggest a 50 fL (about 170 cd/m2) peak light output in a dark room for those who want a little more punch.
I'll give 170 a try. Have used 120 for dark and 250 for bright which was too much for a dark room. Among other things, a brighter 100% helps mask the impact of LG's ever-improving (generation-by-generation) near-black nonuniformity...

Quote:
I keep contrast at 85 on the new models and then reduce OLED light to the target peak white output. Reducing contrast much below 85 degrades the bit depth, and 85 gives full WTW headroom for whatever it's worth (not much IMO but might as well if there's no downside).
Great - thanks. In screwing around the other night, I reduced contrast to about 30 or so and noticed a great deal of posteration on blacks - you've solved the mystery of what caused it (reduced bit depth = increased blockiness/posteration, especially near black).

If WTW does not matter to me, is there any advantage to having Contrast at 100 rather than 85? I think I read somewhere that ABSL can be disabled by setting Contrast to 100...

This was very helpful advice, Chad - thanks. Think I'm finally ready to plunge in for my first greyscale/gamma calibration of the 65C6P.

I don't have many hours on it yet but figure I can tune-up refresh every 50 hours or so as things settle-in. How many hours do you usually recommend on these OLEDs before performing a calibration that will be left alone for a long time?

And one more question on a different subject - are you currently performing calibration of HDR10 on these OLEDs? What about Dolby Vision - are there any adjustments/calibrations you perform for that?
Chad B likes this.
fafrd is offline  
post #11 of 1720 Old 11-23-2016, 10:39 AM
AVS Forum Special Member
 
chunon's Avatar
 
Join Date: Dec 2007
Location: Wilmington Ohio
Posts: 9,065
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 2444 Post(s)
Liked: 2194
The luminance control manipulates gamma, same effect as adjusting red green and blue the same amount.

Sent from my SM-G920V using Tapatalk
Chad B likes this.

65C6 OLED
Time Warner Cable/Panasonic UB900/ Roku Premiere +/Sony S780 Soundbar/Harmony 650
Calman Enthusiast/Chromapure Standard with Auto Cal
chunon is offline  
post #12 of 1720 Old 11-23-2016, 11:02 AM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by chunon View Post
The luminance control manipulates gamma, same effect as adjusting red green and blue the same amount.

Sent from my SM-G920V using Tapatalk
Thanks. Is there any advantage to adjusting luminance by 1 click versus adjusting R, G, and B all by 1 click?

The problem with all of these complimentary controls is that it's not clear what is 'best' and most natural for the TV (fewest artifacts/nonlinarities).

I believe you said you had a 3D LUT, correct?

So is your OLED calibrated using native controls or through your LUT?

And if video through your AVR is calibrated through your LUT, what are you doing as far as calibration for any OTA or native apps (Netflix, Amazon)?
fafrd is offline  
post #13 of 1720 Old 11-23-2016, 11:05 AM
AVS Forum Special Member
 
chunon's Avatar
 
Join Date: Dec 2007
Location: Wilmington Ohio
Posts: 9,065
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 2444 Post(s)
Liked: 2194
Quote:
Originally Posted by fafrd View Post
Thanks. Is there any advantage to adjusting luminance by 1 click versus adjusting R, G, and B all by 1 click?

The problem with all of these complimentary controls is that it's not clear what is 'best' and most natural for the TV (fewest artifacts/nonlinarities).

I believe you said you had a 3D LUT, correct?

So is your OLED calibrated using native controls or through your LUT?

And if video through your AVR is calibrated through your LUT, what are you doing as far as calibration for any OTA or native apps (Netflix, Amazon)?
The net effect is the same Chad can confirm. I used to have a lumagen don't have that anymore. Chad will be calibrating my set soon.

Sent from my SM-G920V using Tapatalk
Chad B likes this.

65C6 OLED
Time Warner Cable/Panasonic UB900/ Roku Premiere +/Sony S780 Soundbar/Harmony 650
Calman Enthusiast/Chromapure Standard with Auto Cal
chunon is offline  
post #14 of 1720 Old 11-23-2016, 02:17 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by Chad B View Post

My preferred method for night time viewing is a 2.35 sliding power with a .001 fL (.0034 cd/m2) manual black level. With a BT1886 formula, I'd increase that a little, to perhaps .005 or .006 cd/m2, or go with a 2.3 power law. You could do one of each and experiment.
Followed your suggestions but difficult to get gamma lining up at the high end.

I suspect the problem may be partly because of the way HCFR is computing luminance targets based on average gamma. Because the 5% and 10% points are below 2.35' the high end of the range ends up being higher than 2.35 (so the average is very close to 2.35).

I need to use a spreadsheet to compute luminance targets. What is '2.35 with a sliding power' and how do you compute your luminance targets?

If I use 2-pt to set 5% and 100%, I just need to know what targets to adjust in between using 21-pt.

Any advise appreciated.
Attached Thumbnails
Click image for larger version

Name:	image.jpg
Views:	1468
Size:	228.8 KB
ID:	1786529  
fafrd is offline  
post #15 of 1720 Old 11-23-2016, 02:37 PM
AVS Forum Special Member
 
chunon's Avatar
 
Join Date: Dec 2007
Location: Wilmington Ohio
Posts: 9,065
Mentioned: 24 Post(s)
Tagged: 0 Thread(s)
Quoted: 2444 Post(s)
Liked: 2194
Are you hitting your "Y" target in the high end ?

Sent from my SM-G920V using Tapatalk

65C6 OLED
Time Warner Cable/Panasonic UB900/ Roku Premiere +/Sony S780 Soundbar/Harmony 650
Calman Enthusiast/Chromapure Standard with Auto Cal
chunon is offline  
post #16 of 1720 Old 11-23-2016, 05:23 PM
AVS Forum Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,578
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 539 Post(s)
Liked: 1296
Here are the targets at each 5% increment, starting with black and ending with 100%:

Target Y cd/m2: 0.0034 0.228 0.9463 2.2818 4.3222 7.1386 10.7916 15.3345 20.815 27.2771 34.761 42.4831 52.0204 62.6829 74.5022 87.5085 101.7305 117.196 133.9314 151.9622 171.313

Target Y fL: 0.001 0.0665 0.2762 0.666 1.2615 2.0835 3.1497 4.4756 6.0751 7.9612 10.1455 12.3993 15.1829 18.2948 21.7445 25.5405 29.6914 34.2052 39.0897 44.3522 50

Target Gamma:Abs: 1.1722(this is black, ignore this) 2.2138 2.2622 2.2818 2.2928 2.3 2.3051 2.3089 2.3119 2.3143 2.3163 2.3179 2.3194 2.3206 2.3218 2.3228 2.3237 2.3245 2.3252 2.3259 2.3265
djkms, T( )( )L and Carlo46 like this.

touring ISF/THX calibrator with Jeti 1211 and Klein K-10
Latest reviews:JVC RS500 and 600, JVC RS400, LG 65C7P, JVC RS420
Copied settings measured, older LG OLED calibration tips, Review index
Chad B is offline  
post #17 of 1720 Old 11-23-2016, 05:35 PM
AVS Forum Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,578
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 539 Post(s)
Liked: 1296
Here is something I posted a while ago that may help at the high end:

Calibrate the 20 point pass-by-pass, NOT in real time. Take a 20 point reading and then, observing an RGB absolute graph (like the large graph in the white balance page of my workflow), put in corrections at each 5% increment as indicated are needed in the graph. Multiply each 1% that the reading is off by 3 and put that correction in the multipoint adjustment (3 is a nominal value; the multiplier is actually closer to 2 from 10-35%, 3 from 40-80%, and 4 from 85-100%, and will be VERY different for software other than CalMAN). Example: if the absolute RGB graph says at 55% blue is 5% low, green is 1% high, and red is 2% low, at interval 55 of the 20 point adjustment increase blue by 15, reduce green by 3, and increase red by 6. Do that for all levels from 10% to 100%.

Attached is an illustration of an RGB absolute graph.
Attached Thumbnails
Click image for larger version

Name:	rgb absolute.jpg
Views:	1514
Size:	29.6 KB
ID:	1786825  

touring ISF/THX calibrator with Jeti 1211 and Klein K-10
Latest reviews:JVC RS500 and 600, JVC RS400, LG 65C7P, JVC RS420
Copied settings measured, older LG OLED calibration tips, Review index
Chad B is offline  
post #18 of 1720 Old 11-23-2016, 06:53 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by Chad B View Post
Here are the targets at each 5% increment, starting with black and ending with 100%:

Target Y cd/m2: 0.0034 0.228 0.9463 2.2818 4.3222 7.1386 10.7916 15.3345 20.815 27.2771 34.761 42.4831 52.0204 62.6829 74.5022 87.5085 101.7305 117.196 133.9314 151.9622 171.313

Target Y fL: 0.001 0.0665 0.2762 0.666 1.2615 2.0835 3.1497 4.4756 6.0751 7.9612 10.1455 12.3993 15.1829 18.2948 21.7445 25.5405 29.6914 34.2052 39.0897 44.3522 50

Target Gamma:Abs: 1.1722(this is black, ignore this) 2.2138 2.2622 2.2818 2.2928 2.3 2.3051 2.3089 2.3119 2.3143 2.3163 2.3179 2.3194 2.3206 2.3218 2.3228 2.3237 2.3245 2.3252 2.3259 2.3265
I was going to ask for your targets - thanks a lot.

I went back to the BT1886Calc spreadsheet and used the 100% level of 170 cd/m2, black level of 0.0034 cd/m2 and power law gamma of 2.35 to end up with the following targets:

5% 0.2023
10% 0.8492
15% 2.0711
20% 3.9616
25% 6.5978
30% 10.0461
35% 14.3656
40% 19.6100
45% 25.8286
50% 33.0673
55% 41.3688
60% 50.7738
65% 61.3206
70% 73.0456
75% 85.9839
80% 100.1687
85% 115.6322
90% 132.4055
95% 150.5185
100% 170.0000

Adjusted OLED Light to hit 170, then cycled through 2-pt a few times to get 5% and 100% on target, then did 21-pt in revers to hit these targets from 95% to 10%, the did a forward pass from 5% to 100%.

Not finalized yet but the results came out far closer to what I was hoping for - time to test on some content tonight .

I'll store your luminance targets and try that as a second calibration - how are those luminance targets generated? Automatically by Calman?
Attached Thumbnails
Click image for larger version

Name:	image.jpg
Views:	1480
Size:	148.1 KB
ID:	1786913  

Last edited by fafrd; 11-24-2016 at 12:26 AM.
fafrd is offline  
post #19 of 1720 Old 11-23-2016, 07:03 PM
AVS Forum Special Member
 
Chad B's Avatar
 
Join Date: Oct 2002
Location: Piqua, OH
Posts: 2,578
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 539 Post(s)
Liked: 1296
Looks like an error in your 10% target. I would double check that. Looks like both the table and measurements have a problem.

However, other than that pretty close, just slightly higher gamma so slightly less shadow detail.

I generated that automatically in CalMAN by specifying a the white and black targets and then generating a custom data table.

touring ISF/THX calibrator with Jeti 1211 and Klein K-10
Latest reviews:JVC RS500 and 600, JVC RS400, LG 65C7P, JVC RS420
Copied settings measured, older LG OLED calibration tips, Review index
Chad B is offline  
post #20 of 1720 Old 11-24-2016, 12:32 AM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by Chad B View Post
Looks like an error in your 10% target. I would double check that. Looks like both the table and measurements have a problem.
Fixed the typo in the original post. New measurements below.

Quote:
However, other than that pretty close, just slightly higher gamma so slightly less shadow detail.

I generated that automatically in CalMAN by specifying a the white and black targets and then generating a custom data table.
I made another pass before calling a night. Also checked some content and happy with the increased shadow detail. If I can hit close to your 100% brightness, I may try dialing-in your Calman targets, but the BT.1886 spreadsheet makes it easy to calculate targets - do you know what function Calman uses or if there exists a spreadsheet to calculate the same targets?
Attached Thumbnails
Click image for larger version

Name:	image.jpg
Views:	1388
Size:	218.9 KB
ID:	1787249  
fafrd is offline  
post #21 of 1720 Old 11-26-2016, 12:29 AM
Member
 
bilditup1's Avatar
 
Join Date: Nov 2006
Location: brooklyn
Posts: 137
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 58 Post(s)
Liked: 18
A few questions for Chad B

About to embark on a calibration of my own B6P and appreciate all this advice. I have a few questions though to make sure I understood you clearly (am a bit of a beginner here, as well soon become crystal):

Quote:
Originally Posted by Chad B View Post
1. ...Because of this, I like to put up a moving zone plate during any long time gaps to keep things consistent.
So I had to Google what this was just now (have only calibrated three sets, all of which were FALD Vizios). Will
suffice? Or else - which calibration disc are you using? So far I've been using the AVS HD 709 disc/files (in conjunction with ChromaPure) and it doesn't look like a pattern of this kind is up there. Or can I just avoid this situation by shutting off the TV's screen saver (which of course I'll do anyway) and making sure the screen is always fed with something other than black? Do you know more precisely what the threshold is before this becomes an issue - in other words, how often would I want to switch to the moving zone plate video?

Quote:
Originally Posted by Chad B View Post
2. There is a difference in calibration of HDR mode. The B's code value white balance adjustments start out very fine, whereas the C/E/G's start out very coarse at the low end (code value 127 is so coarse as to be completely unusable) and become finer and more workable as brightness and code value increases. The B's fine adjustments work to it's advantage in the first several code value steps, but they then have too little strength to do any good at the high end. I have found that attempting to calibrate the higher code values on a B may cause problems in tone mapping, so I just concentrate on getting the green emphasis out of the low and mid tones on it.
Noted for when I actually have HDR content and calibration software that can calibrate for HDR. Thanks.

Quote:
Originally Posted by Chad B View Post
3. Before calibration in the ISF modes, there is a general tendency for the white balance in all the 2016 models to emphasize green in the low and mid tones and then transition to a lack of green in the brighter levels.
Is it your professional opinion that the starting point for calibrations should be one of the ISF modes, as opposed to, say, Cinema mode? Can you explain why? Should the starting point for calibrations be Bright Room for day and Dark Room for night?

Quote:
Originally Posted by Chad B View Post
4. The CMS, though improved over some earlier LG OLED models, still does not function very well. Attempting to calibrate it can cause visual problems like blotchy colors and contouring. The problem is, the adjustment only effects a narrow slice of the color spectrum, and the transition from effected color shades to non effected are far too abrupt. Let's say you turn red saturation all the way down. If you have an image that is supposed to smoothly transition from an orangish shade to a pure red, there is absolutely no change until it reaches pure red, and then all of a sudden it becomes black & white because the red saturation is lowered. Ideally, instead of this sharp, abrupt behavior, the control should start having a gradual effect on orange, violet, and pink shades and then smoothly reach it's maximum effect at pure red. I've also seen this behavior on LG LED LCDs.
This sounds pretty awful. To be sure I understood, this means that the CMS should be entirely avoided and any attempt to use it will result in these kinds of abrupt changes in hue, correct? And we should just live with (...) the colors once greyscale and gamma calibration are complete?

Quote:
Originally Posted by Chad B View Post
5. Though windows should be used rather than full fields, there is no significant difference between any measurement window size up to 25%. There is also no significant difference between APL and conventional windows.
Noted. Are the window patterns from AVS709's ChromaPure section sufficiently small (25% or smaller)? Here's one, for example.

Quote:
Originally Posted by Chad B View Post
6. Like previous LG OLEDs, there is a static brightness limiter. If you leave a test pattern up too long it starts to dim and it's measured characteristics change. Because of this, I prefer to set the 100% and 5% white balance with the 2 point control in real time, and then do the 20 point adjustment pass by pass (not in real time). I can explain this more later.
Quote:
Originally Posted by Chad B View Post
Here is something I posted a while ago that may help at the high end:

Calibrate the 20 point pass-by-pass, NOT in real time. Take a 20 point reading and then, observing an RGB absolute graph (like the large graph in the white balance page of my workflow), put in corrections at each 5% increment as indicated are needed in the graph...
By 'high end', do you mean '80%+ Gray'? Also, to be clear, by 'not in real time', do you just mean: take initial measurements, calibrate using 2-point IRE (I guess this means switching to 20 point later will not eliminate the values inputted for 2-point?) and then apply this heuristic to try to get a good result, instead of adjusting IRE values while monitoring them with your calibration software/meter?
Finally, you mention touching 20-point IRE only in the context of greyscale, or at least that's what it sounds like. But the workflow is still 'normal', right? IOW, get greyscale right, either using this heuristic or in 'real time' (though I'm still not sure yet if I understood the distinction you were making there...) and then further modify those values in targeting gamma. Correct?

Quote:
Originally Posted by Chad B View Post
7. Turning OLED light to 100 and reducing contrast to achieve the desired light output will minimize ABL, but lowering contrast significantly below 85 reduces the bit depth and is not recommended in most situations.
Noted. Can you explain, though, the difference between the 'Backlight' control, as opposed to the 'OLED LIGHT' control? (The two are listed as distinct in the manual at least.) Is there a good balance you would recommend between the OLED light and contrast values, beyond making sure that contrast stays relatively high (close to 85)? I suppose this would somehow depend on the light output I was going for (let's stick with 170 cd/m2 I think).

Quote:
Originally Posted by Chad B View Post
8. After calibrating grayscale and gamma, measuring the SG fleshtones color checker patterns will generally show that skin tones lean towards red. Because of the CMS control's poor behavior, the only way to help this is to move the main tint control a bit toward green. Usually the SG fleshtones will be oversaturated as well, which can be improved by lowering the main color control, though this isn't always the case. I typically end up with the main color and tint at something like 48 and G2.
Re: 'SG fleshtones color checker patterns' - can you recommend a disk with these patterns? AVS709 doesn't appear to have them. It looks like this one does but I'm not sure if it's worth investing in. So I guess as in (1) above I'm asking for a recommendation on this again, heh.
In any case, what should the workflow here be? Measure, change main color and tint, and measure again? Before embarking on calibration of greyscale and gamma, then, do you recommend that Color be left at 50 and Tint at 0, instead of using flashing color bars and the TV's blue filter to get the 'boxes' to match up beforehand (that's what I've been doing until now).

Quote:
Originally Posted by Chad B View Post
9. My personal opinion is that the OLEDs do not look natural calibrated to BT1886 (which in this case should be a straight 2.4) gamma. That tends to look too contrasty and lack shadow detail. On the other hand, a straight 2.2 power law gamma can lack depth. CalMAN gives the option of calibrating to a sliding power law gamma, and using this option with a starting point of 2.3-2.35 and specifying a fake black level of .001 to .002 fL depending on the environment and light output produces a very natural and realistic image on these sets.
It does not appear that ChromaPure allows for custom gamma targets like this. Presumably though I can just calibrate to the target Y's you posted here (I'm OK with 170 for nighttime/dim viewing) and the only thing I'll lose is the benefit of a nice-looking graph, right? (I am realizing more and more, btw, the limitations of this software. Le sigh.)

I know these many of these questions are pretty basic, so thanks in advance for your time, effort, and patience.
bilditup1 is offline  
post #22 of 1720 Old 11-26-2016, 11:37 AM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by bilditup1 View Post
About to embark on a calibration of my own B6P and appreciate all this advice. I have a few questions though to make sure I understood you clearly (am a bit of a beginner here, as well soon become crystal):



So I had to Google what this was just now (have only calibrated three sets, all of which were FALD Vizios). Will something like this suffice? Or else - which calibration disc are you using? So far I've been using the AVS HD 709 disc/files (in conjunction with ChromaPure) and it doesn't look like a pattern of this kind is up there. Or can I just avoid this situation by shutting off the TV's screen saver (which of course I'll do anyway) and making sure the screen is always fed with something other than black? Do you know more precisely what the threshold is before this becomes an issue - in other words, how often would I want to switch to the moving zone plate video?



Noted for when I actually have HDR content and calibration software that can calibrate for HDR. Thanks.



Is it your professional opinion that the starting point for calibrations should be one of the ISF modes, as opposed to, say, Cinema mode? Can you explain why? Should the starting point for calibrations be Bright Room for day and Dark Room for night?



This sounds pretty awful. To be sure I understood, this means that the CMS should be entirely avoided and any attempt to use it will result in these kinds of abrupt changes in hue, correct? And we should just live with (...) the colors once greyscale and gamma calibration are complete?



Noted. Are the window patterns from AVS709's ChromaPure section sufficiently small (25% or smaller)? Here's one, for example.




By 'high end', do you mean '80%+ Gray'? Also, to be clear, by 'not in real time', do you just mean: take initial measurements, calibrate using 2-point IRE (I guess this means switching to 20 point later will not eliminate the values inputted for 2-point?) and then apply this heuristic to try to get a good result, instead of adjusting IRE values while monitoring them with your calibration software/meter?
Finally, you mention touching 20-point IRE only in the context of greyscale, or at least that's what it sounds like. But the workflow is still 'normal', right? IOW, get greyscale right, either using this heuristic or in 'real time' (though I'm still not sure yet if I understood the distinction you were making there...) and then further modify those values in targeting gamma. Correct?



Noted. Can you explain, though, the difference between the 'Backlight' control, as opposed to the 'OLED LIGHT' control? (The two are listed as distinct in the manual at least.) Is there a good balance you would recommend between the OLED light and contrast values, beyond making sure that contrast stays relatively high (close to 85)? I suppose this would somehow depend on the light output I was going for (let's stick with 170 cd/m2 I think).



Re: 'SG fleshtones color checker patterns' - can you recommend a disk with these patterns? AVS709 doesn't appear to have them. It looks like this one does but I'm not sure if it's worth investing in. So I guess as in (1) above I'm asking for a recommendation on this again, heh.
In any case, what should the workflow here be? Measure, change main color and tint, and measure again? Before embarking on calibration of greyscale and gamma, then, do you recommend that Color be left at 50 and Tint at 0, instead of using flashing color bars and the TV's blue filter to get the 'boxes' to match up beforehand (that's what I've been doing until now).



It does not appear that ChromaPure allows for custom gamma targets like this. Presumably though I can just calibrate to the target Y's you posted here (I'm OK with 170 for nighttime/dim viewing) and the only thing I'll lose is the benefit of a nice-looking graph, right? (I am realizing more and more, btw, the limitations of this software. Le sigh.)

I know these many of these questions are pretty basic, so thanks in advance for your time, effort, and patience.
That's a lot of questions and I'll let respond with professional advice when he finds the time.

Here is some DIY advice.

Start with mastering greyscale/gamma. This is both because color on these OLEDs is very good OOTB and also because getting near-black and gamma right has the biggest impact on getting the most out of these perfect-black OLEDs.

I can't comment on Chromapure, but HCFR is a very good freeware package to learn with. Going through the basics of setting OLED light for 170 cd/m2 output, Brightness for video level 18 just visible and 16 perfect black, then adjusting 2-pt for 5% and 100% by cycling back and forth a few times sets a good baseline.

From there, you can use 21-pt to dial in whitepoint/greyscale. There's a few ways to do it, but I now find it easiest to sweep through making adjustments to red and blue based on the prior readings until all delta xy measurements are equally small. I almost never adjust green, but sometimes adjusting green by a click or two to get delta xy down to the next level should be considered.

You are ignoring dE and gamma at this stage - just getting white balance right.

With that done, I then used the new Adjusting Luminance control to dial-in gamma. This new control was pretty effective at allowing me to get luminance targets dialed-in without screwing up white balance again.

Chad or someone else may pipe up with a reason these new controls should not be used, but my calibration-in-progress is developing nicely and I now have Dark-Room ISF dialed in to BT.1886 as specified by the luminance targets defined a few posts back and Bright Room ISF dialed in to Chad's luminance targets.

There is no specific reason to start with the ISF modes. The Cinema made offers pretty much the same full set of controls (at least for white balance) and I used Cinema to dial-in a day mode at 250 cd/m2 using PLG 2.2.

I plan to limit myself to calibration of white-point until it becomes more clear how to calibrate Dolby Vision...

Good luck.
fafrd is offline  
post #23 of 1720 Old 11-26-2016, 12:14 PM
Member
 
bilditup1's Avatar
 
Join Date: Nov 2006
Location: brooklyn
Posts: 137
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 58 Post(s)
Liked: 18
Quote:
Originally Posted by fafrd View Post
Start with mastering greyscale/gamma. This is both because color on these OLEDs is very good OOTB and also because getting near-black and gamma right has the biggest impact on getting the most out of these perfect-black OLEDs.
Makes sense, I'll keep this in mind.

Quote:
Originally Posted by fafrd View Post
I can't comment on Chromapure, but HCFR is a very good freeware package to learn with.
Well just for calibrating grayscale (not gamma) I'm sure ChromaPure would be sufficient - there is no confounding variable here of calibrating to a custom target. Can you comment on which pattern disc and window size you're using though? And how you're avoiding the problem of the ABL feature changing your readings?

Quote:
Originally Posted by fafrd View Post
Going through the basics of setting OLED light for 170 cd/m2 output, Brightness for video level 18 just visible and 16 perfect black,
Yeah, the general guides I've seen in the past say 16 barely visible rather than perfect black, but that seems wrong to me. Thanks for confirming my suspicions.

Quote:
Originally Posted by fafrd View Post
then adjusting 2-pt for 5% and 100% by cycling back and forth a few times sets a good baseline.

From there, you can use 21-pt to dial in whitepoint/greyscale. There's a few ways to do it, but I now find it easiest to sweep through making adjustments to red and blue based on the prior readings until all delta xy measurements are equally small. I almost never adjust green, but sometimes adjusting green by a click or two to get delta xy down to the next level should be considered.
Sounds good/basically like what I was going to do, except for the tip about doing 5% and 100% first. Thanks for that.

Quote:
Originally Posted by fafrd View Post
With that done, I then used the new Adjusting Luminance control to dial-in gamma. This new control was pretty effective at allowing me to get luminance targets dialed-in without screwing up white balance again.
I haven't seen this control just yet and will have to look out for it, cool.

Quote:
Originally Posted by fafrd View Post
Chad or someone else may pipe up with a reason these new controls should not be used, but my calibration-in-progress is developing nicely and I now have Dark-Room ISF dialed in to BT.1886 as specified by the luminance targets defined a few posts back and Bright Room ISF dialed in to Chad's luminance targets.
as specified by the luminance targets defined a few posts back" - sorry? Which ones? "Chad's luminance targets" I thought referred to the ones for 170 cd/m2 which I thought was intended to replace the 120 cd/m2 mode. Are you saying you're using 120 cd/m2 with ISF Dark and 170 cd/m2+Chad's custom targets with ISF Bright?

Quote:
Originally Posted by fafrd View Post
There is no specific reason to start with the ISF modes. The Cinema made offers pretty much the same full set of controls (at least for white balance) and I used Cinema to dial-in a day mode at 250 cd/m2 using PLG 2.2.
OK great, cool beans.

Quote:
Originally Posted by fafrd View Post
Good luck.
Thanks!
bilditup1 is offline  
post #24 of 1720 Old 11-26-2016, 01:15 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by bilditup1 View Post
Well just for calibrating grayscale (not gamma) I'm sure ChromaPure would be sufficient - there is no confounding variable here of calibrating to a custom target. Can you comment on which pattern disc and window size you're using though? And how you're avoiding the problem of the ABL feature changing your readings?
I use Ted's Lightspace disk as my reference, but also have the GCD and AVSHD709 test patterns. I believe Ted's patterns are 10% window.

It's far faster using the pattern generator built into HCFR. I use the GDI patterns and 1% windows when doing that. Generally calibrate using HCFR patterns and then check using Ted's Lightspace.

As far as ABL, you will occassionally see a real-time reading drop so you know ABL or AVSL kicked-in. When this happens, I just advance by one pattern and then shift back and reading returns to usual value.

But more and more, I am avoiding real-time measurements and adjustments while reading. I find it faster to make a full sweep reading using HCFR patterns, cycle through the entire 21-pt settings making adjustments as per the prior reading, and then make another sweep.


Quote:
Yeah, the general guides I've seen in the past say 16 barely visible rather than perfect black, but that seems wrong to me. Thanks for confirming my suspicions.



Sounds good/basically like what I was going to do, except for the tip about doing 5% and 100% first. Thanks for that.
If you read through this entire thread, you will see an early post from Chad where he suggests that. I have only started using that technique with this calibration but I find it to be much easier and more effective on these OLEDs.

Quote:
I haven't seen this control just yet and will have to look out for it, cool.
It's brand new and I have not seen anyone recommend using it. I tried it yesterday and found it very effective for dialing-in gamma. Chad or someone else may eventually chime in with a reason it should be avoided, but my logic is even if that should be the case, backing off the single new control in exchange for appropriate adjustments to R, G and B should always be possible....

So my advice is: explore, enjoy .

Quote:
as specified by the luminance targets defined a few posts back" - sorry? Which ones? "Chad's luminance targets" I thought referred to the ones for 170 cd/m2 which I thought was intended to replace the 120 cd/m2 mode. Are you saying you're using 120 cd/m2 with ISF Dark and 170 cd/m2+Chad's custom targets with ISF Bright?
After viewing Dolby Vision in the dark, I now find 120 cd/m2 peak too dark/dim on these OLEDs, even when watching in a pitch-black room (though with some reflections).

So I'm calibrating both Dark Room and Light Room to 170 cd/m2 peak.

Using Chad's suggested effective black level, I've used the BT,1886 spreadsheet to define BT.1886 luminance targets which is what I listed in my earlier post. I used these targets for my Datk Room Calibration.

Chad was kind enough to provide his luminance targets for 170 cd/m2 peak, and so I have calibrated my Bright Room gamma to these targets.

I'm switching back and forth on certain content, especially dark scenes, to see which gamma I prefer. So far, it's been difficult to see much difference.

For viewing blurays or streaming in a bright room, I've used Cinema, setting peak output to 250 cd/m2 and luminance targets set by a straight-up power-law-gamma of 2.2.

Dark Knight is a good movie to explore the impact of different near-black settings on shadow detail...
fafrd is offline  
post #25 of 1720 Old 11-26-2016, 03:16 PM
Member
 
bilditup1's Avatar
 
Join Date: Nov 2006
Location: brooklyn
Posts: 137
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 58 Post(s)
Liked: 18
Quote:
Originally Posted by fafrd View Post
I use Ted's Lightspace disk as my reference, but also have the GCD and AVSHD709 test patterns. I believe Ted's patterns are 10% window.
Aha, cool.
Quote:
Originally Posted by fafrd View Post
As far as ABL, you will occassionally see a real-time reading drop so you know ABL or AVSL kicked-in. When this happens, I just advance by one pattern and then shift back and reading returns to usual value.
So there's an easy fix. Brilliant.
Quote:
Originally Posted by fafrd View Post
But more and more, I am avoiding real-time measurements and adjustments while reading. I find it faster to make a full sweep reading using HCFR patterns, cycle through the entire 21-pt settings making adjustments as per the prior reading, and then make another sweep.
'cycle through the entire 21-pt settings making adjustments as per the prior reading' - in other words, you are using Chad's heuristic to make said adjustments? Otherwise, how do you know how much to adjust by without doing a reading at the same time?
Quote:
Originally Posted by fafrd View Post
If you read through this entire thread, you will see an early post from Chad where he suggests that. I have only started using that technique with this calibration but I find it to be much easier and more effective on these OLEDs.
Yeah, I saw it when Chad said it, and actually asked him in that long post to confirm that switching from 2-pt mode to 21-pt would not kill the settings made in 2-pt mode. But I thought that you were independently making the same recommendation and was thanking you all the same
Quote:
Originally Posted by fafrd View Post
It's brand new and I have not seen anyone recommend using it. I tried it yesterday and found it very effective for dialing-in gamma. Chad or someone else may eventually chime in with a reason it should be avoided, but my logic is even if that should be the case, backing off the single new control in exchange for appropriate adjustments to R, G and B should always be possible....
Right...
Quote:
Originally Posted by fafrd View Post
So my advice is: explore, enjoy .
Indeed!
Quote:
Originally Posted by fafrd View Post
After viewing Dolby Vision in the dark, I now find 120 cd/m2 peak too dark/dim on these OLEDs, even when watching in a pitch-black room (though with some reflections).

So I'm calibrating both Dark Room and Light Room to 170 cd/m2 peak.

Using Chad's suggested effective black level, I've used the BT,1886 spreadsheet to define BT.1886 luminance targets which is what I listed in my earlier post. I used these targets for my Datk Room Calibration.

Chad was kind enough to provide his luminance targets for 170 cd/m2 peak, and so I have calibrated my Bright Room gamma to these targets.
So to recap you're using the BT1886 spreadsheet for Dark Room and chad's targets for Bright Room, with the same peak output. In which case I don't think I understood why BT1886 should be used for dark and chad's for bright. Are you doing this arbitrarily just to be able to quickly compare the two gammas?
Quote:
Originally Posted by fafrd View Post
I'm switching back and forth on certain content, especially dark scenes, to see which gamma I prefer. So far, it's been difficult to see much difference.
So the difference is marginal. Noted.
Quote:
Originally Posted by fafrd View Post
For viewing blurays or streaming in a bright room, I've used Cinema, setting peak output to 250 cd/m2 and luminance targets set by a straight-up power-law-gamma of 2.2.

Dark Knight is a good movie to explore the impact of different near-black settings on shadow detail...
Thanks for the advice, I'll be keeping all this in mind. Sweet!
bilditup1 is offline  
post #26 of 1720 Old 11-26-2016, 04:38 PM
Member
 
Join Date: Feb 2016
Location: Australia
Posts: 171
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 89 Post(s)
Liked: 42
First attempt using a meter to get the greyscale right. I think I've made quite a bit of progress. The two extremes are rather off - is that worth doing the 20-point with?
Attached Thumbnails
Click image for larger version

Name:	pre cal.jpg
Views:	690
Size:	89.0 KB
ID:	1791161   Click image for larger version

Name:	post cal values.png
Views:	684
Size:	17.3 KB
ID:	1791169  

LG OLED55C6T 4K HDR TV, Yamaha RX-A2050 Receiver, Q Acoustics 3010, 3050, 3070 and 3090 speakers, Yamaha YST-SW030 rear sub, Sony BDP-S6500 Blu ray, LG UP970 UHD, Fetch TV Mighty, IQ2HD, Apple TV Series 4, Pioneer CLD2730K, eeColor LUT+Lightspace HTL+i1Display Pro and DreamScreen.
vollans is offline  
post #27 of 1720 Old 11-26-2016, 04:58 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by bilditup1 View Post
Aha, cool.
'cycle through the entire 21-pt settings making adjustments as per the prior reading' - in other words, you are using Chad's heuristic to make said adjustments? Otherwise, how do you know how much to adjust by without doing a reading at the same time?
Chad is a professional calibrator and can probably get 21-pt dialed-in in a single pass.

I just use some simple heuristics based on the reading. If red and blue are similar and a bit below green, I add a single click to both. Red way farther below than blue, then two clicks for red and one for blue. You'll see that your judgement gets dialed-in pretty quickly so in the end you'll be pretty close within 3 passes or so and then it's up to how much you want to let your OCD have its way with you .

This is slower when reading off of Bluray but with the HCFR generator, the actual re-reading is very fast - it's to parameter adjustment that takes the most time.


Quote:
Yeah, I saw it when Chad said it, and actually asked him in that long post to confirm that switching from 2-pt mode to 21-pt would not kill the settings made in 2-pt mode. But I thought that you were independently making the same recommendation and was thanking you all the same
Adjusting 1-pt first greatly reduces the amount of adjustment needed by 21-pt. the two are independant and additive.

Brightness adjusts brightness for all three primaries together while 2-pt low adjust brightness for each primary independantly.. Contrast adjusts gain for all 3 primaries together whole 2-pt high adjusts gain for each primary independantly.

21-pt then adds a further adjustment on top of that 2-pt baseline: Adjusting Luminance makes a local change for all 3 primaries together while R, G, and B each make a local adjustment for just one specific primary.

Quote:
So to recap you're using the BT1886 spreadsheet for Dark Room and chad's targets for Bright Room, with the same peak output. In which case I don't think I understood why BT1886 should be used for dark and chad's for bright. Are you doing this arbitrarily just to be able to quickly compare the two gammas?
I'm comparing two very similar calibrations for dark-room viewing. One is using BT.1886 luminance targets and stored as ISF Dark and the other is using Chad"s luminance targets and stored as ISF Bright. As long as the settings for Contrast and OLED Light are set in the same way, the two calibrations will be similar regardless of name. By default, Bright ISF is set for higher peak output than Dark (both defaults of which I overrode to be identical and delivering 170 cd/m2 peak.

And yes, I've dropped both in so any time I am unhappy with shadow detail and feel a scene is either too dark or too light, I can switch to the other ISF mode to see whether I like it better (I know, I know, first-world problem )

I think you'll be happy with either of those calibrations. The difference for me is, any time I want to calibrate for a different peak brightness, I know how to generate the BT.1886 targets off of the spreadsheet. For Chad's Calman targets, I need to pester him to check and share .
bilditup1 likes this.

Last edited by fafrd; 11-26-2016 at 05:08 PM.
fafrd is offline  
post #28 of 1720 Old 11-26-2016, 05:08 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by vollans View Post
First attempt using a meter to get the greyscale right. I think I've made quite a bit of progress. The two extremes are rather off - is that worth doing the 20-point with?
A lot of work needed here.

Restore to factory defaults.

Use Black Pluge patterns to set black level video level 16 invisible, video level 18 visible).

Use White pluge patterns and your meter to set peak putout (video level 235) - leave Contrast at 85.

Measure 21-pt (or 11-pt if that is all you have) and then open a second workspace.

Use 2-pt Low to adjust whitepoint of 5% to as close to perfect balance as you can get it; then use 2-pt high to adjust whitepoint of 100% to as close to perfect as you can get it. Repeat this cycle 3-5 times (2-ot low also affects 100% and 2-pt high also affects 5%).

When you've settled on the best 2-pt settings you can find, remeasire 21-Pt. you should see 5% and 100% close to perfect and the errors between will dictate whether you need to adjust 21-pt.
drapp1952 and vollans like this.
fafrd is offline  
post #29 of 1720 Old 11-26-2016, 06:24 PM
Member
 
bilditup1's Avatar
 
Join Date: Nov 2006
Location: brooklyn
Posts: 137
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 58 Post(s)
Liked: 18
Any way to reset to defaults without deleting all the apps, or no?
bilditup1 is offline  
post #30 of 1720 Old 11-26-2016, 10:06 PM - Thread Starter
AVS Forum Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 9,568
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 2731
Quote:
Originally Posted by bilditup1 View Post
Any way to reset to defaults without deleting all the apps, or no?
I believe 'reset to default values' in the picture settings and whitepoint settings only resets picture settings without changing app logins, etc...

Reset to factory defaults at the general/TV level is a different story, but that is not what I meant...
fafrd is offline  
Sponsored Links
Advertisement
 
Reply Display Calibration



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