Patterns don't work well and can mislead you into thinking you have a black crush issue when you really don't. I have had sets, plasma and OLED that would not show bar 17 or 1 (whatever pattern you use). This would lead you to believe there is a black crush problem, however, I have found that with actual content in many cases there is no black crush issue. Also on most OLED you will never get perfect shadow detail, it will never be perfect. The best thing a person can do is stop chasing perfection with an OLED. Its great tech but it also has flaws.I figured out how to fix the "black crush" issue on my c9 (should be no different with other models) using a meter without rising black luminance.
I use a white led 98% srgb lcd with standard ips black levels calibrated to flat 2.2 gamma d65 rec 709 to visually check and adjust the calibration results on the tv. OLED should look a bit darker at the low end compared to an lcd when both are calibrated to flat 2.2 gamma. I don't use 2.4 or bt 1886 (because why)?
There's actually no such thing as oled black crush, all the levels below 5 percent are still visible they are just darker than on lcd because the black point is much darker on oled than on lcd so the lower luminance points are darker as well, that's why I think it looks darker, but maybe I'm wrong idk, the fact is my c9 looks a bit darker than my lcd when both calibrated to flat 2.2 gamma.
Heres a solution I came up with using both my eyes and the measurements (I have i1d3 plus) to match the low end luminance to a reference lcd without rising black level.
First setting black levels correctly, I really don't understand why use the flashing bar pattern to set the black levels, what I did instead is sit in a completely dark room with the only light source being the tv, then display a full field 0% pattern and increase brightness until the pixels turn on, it's easy to see in complete darkness, my brightness ended up being 53 with 2.2 gamma preset, increasing it to 54 rised black luminance above 0cdm, also the eyes are far more sensitive at these near 0 cdm2 levels than the meter because i1d3 couldn't actually detect the luminance change from 0 cd/m2 at 54 brightness.
With the black level set I did the same with 2.5% IRE because messing with it will also rise black level, 2.5% ire with luminance set to 2 didn't rise the black level with 2.2 gamma preset so I set it to 2 as a compromise, setting it to 3 raised the black luminance, but to a lesser degree than setting brightness to 54.
Then what I did was to measure the black luminance on my macbook lcd which I use as a reference, it turned out to be around 0.087 cd/m2. Then in HCFR gamma preferences tab I checked the box "override black" and entered this value in the field. This way HCFR now thinks that c9 has the same black luminance as the lcd and won't even give you 0% pattern to measure and will build the gamma curve using this custom black luminance.
Then I measured and aligned the gamma curve to 2.2 using 22 points.
Then I measured the near black luminance with 7 points, this showed that 0% through 5% still tracked far below 0.087 cd/m2, that's basically where the oled "black crush" is. So I rised ire 5% luminance to 7 and ire 7.5% to 6.
this put luminance levels from 4% through 10% to above the lcd black level luminance of 0.087 cd/m2. The low end below 4% was still below 0.087 cd/m but rising ire 5 luminance above 7 would put 7%-10% way below 2.2 gamma without rising below 4% luminance as much so I figured this would be a good compromise.
This is it, looking at my c9 and macbook side by side they now look almost identical even in terms of low level detail, while c9 is still a bit darker because <4% is still below the lcd black level this is I think pretty much the best you can do without finer low end luminance controls without rising black level above 0 cd/m2 or messing up above 5%.
You can match oled to an lcd completely by rising black level to an lcd black luminance around 0.9 cd/m2 using brightness controls and then realigning the gamma with this new black point. But then what's the point of OLED in the first place considering that the OLED viewing angles are **** compared to a good IPS anyway?
TLDR:
using black luminance override in hcfr to put 0-7% grey above the lcd black luminance without rising black level
brightness 53
ire 2.5 luminance 2
ire 5 luminance 7
ire 7 luminance 6
ps would love to hear what I did wrong and if all of this isn't correct considering that I only recently got into tv calibration and only just learning after getting an lg c9 and being blown away how terrible the out of box color accuracy is, and I probably have no idea wtf I am doing
-
Get an exclusive sneak peek into our new project. >>> Click Here
-
Our native mobile app has a new name: Fora Communities. Learn more.
Correcting OLED "black crush"
I figured out how to fix the "black crush" issue on my c9 (should be no different with other models) using a meter without rising black luminance.
I use a white led 98% srgb lcd with standard ips black levels calibrated to flat 2.2 gamma d65 rec 709 to visually check and adjust the calibration results on the tv. OLED should look a bit darker at the low end compared to an lcd when both are calibrated to flat 2.2 gamma. I don't use 2.4 or bt 1886 (because why)?
There's actually no such thing as oled black crush, all the levels below 5 percent are still visible they are just darker than on lcd because the black point is much darker on oled than on lcd so the lower luminance points are darker as well, that's why I think it looks darker, but maybe I'm wrong idk, the fact is my c9 looks a bit darker than my lcd when both calibrated to flat 2.2 gamma.
Heres a solution I came up with using both my eyes and the measurements (I have i1d3 plus) to match the low end luminance to a reference lcd without rising black level.
First setting black levels correctly, I really don't understand why use the flashing bar pattern to set the black levels, what I did instead is sit in a completely dark room with the only light source being the tv, then display a full field 0% pattern and increase brightness until the pixels turn on, it's easy to see in complete darkness, my brightness ended up being 53 with 2.2 gamma preset, increasing it to 54 rised black luminance above 0cdm, also the eyes are far more sensitive at these near 0 cdm2 levels than the meter because i1d3 couldn't actually detect the luminance change from 0 cd/m2 at 54 brightness.
With the black level set I did the same with 2.5% IRE because messing with it will also rise black level, 2.5% ire with luminance set to 2 didn't rise the black level with 2.2 gamma preset so I set it to 2 as a compromise, setting it to 3 raised the black luminance, but to a lesser degree than setting brightness to 54.
Then what I did was to measure the black luminance on my macbook lcd which I use as a reference, it turned out to be around 0.087 cd/m2. Then in HCFR gamma preferences tab I checked the box "override black" and entered this value in the field. This way HCFR now thinks that c9 has the same black luminance as the lcd and won't even give you 0% pattern to measure and will build the gamma curve using this custom black luminance.
Then I measured and aligned the gamma curve to 2.2 using 22 points.
Then I measured the near black luminance with 7 points, this showed that 0% through 5% still tracked far below 0.087 cd/m2, that's basically where the oled "black crush" is. So I rised ire 5% luminance to 7 and ire 7.5% to 6.
this put luminance levels from 4% through 10% to above the lcd black level luminance of 0.087 cd/m2. The low end below 4% was still below 0.087 cd/m but rising ire 5 luminance above 7 would put 7%-10% way below 2.2 gamma without rising below 4% luminance as much so I figured this would be a good compromise.
This is it, looking at my c9 and macbook side by side they now look almost identical even in terms of low level detail, while c9 is still a bit darker because <4% is still below the lcd black level this is I think pretty much the best you can do without finer low end luminance controls without rising black level above 0 cd/m2 or messing up above 5%.
You can match oled to an lcd completely by rising black level to an lcd black luminance around 0.9 cd/m2 using brightness controls and then realigning the gamma with this new black point. But then what's the point of OLED in the first place considering that the OLED viewing angles are **** compared to a good IPS anyway?
TLDR:
using black luminance override in hcfr to put 0-7% grey above the lcd black luminance without rising black level
brightness 53
ire 2.5 luminance 2
ire 5 luminance 7
ire 7 luminance 6
ps would love to hear what I did wrong and if all of this isn't correct considering that I only recently got into tv calibration and only just learning after getting an lg c9 and being blown away how terrible the out of box color accuracy is, and I probably have no idea wtf I am doing
I use a white led 98% srgb lcd with standard ips black levels calibrated to flat 2.2 gamma d65 rec 709 to visually check and adjust the calibration results on the tv. OLED should look a bit darker at the low end compared to an lcd when both are calibrated to flat 2.2 gamma. I don't use 2.4 or bt 1886 (because why)?
There's actually no such thing as oled black crush, all the levels below 5 percent are still visible they are just darker than on lcd because the black point is much darker on oled than on lcd so the lower luminance points are darker as well, that's why I think it looks darker, but maybe I'm wrong idk, the fact is my c9 looks a bit darker than my lcd when both calibrated to flat 2.2 gamma.
Heres a solution I came up with using both my eyes and the measurements (I have i1d3 plus) to match the low end luminance to a reference lcd without rising black level.
First setting black levels correctly, I really don't understand why use the flashing bar pattern to set the black levels, what I did instead is sit in a completely dark room with the only light source being the tv, then display a full field 0% pattern and increase brightness until the pixels turn on, it's easy to see in complete darkness, my brightness ended up being 53 with 2.2 gamma preset, increasing it to 54 rised black luminance above 0cdm, also the eyes are far more sensitive at these near 0 cdm2 levels than the meter because i1d3 couldn't actually detect the luminance change from 0 cd/m2 at 54 brightness.
With the black level set I did the same with 2.5% IRE because messing with it will also rise black level, 2.5% ire with luminance set to 2 didn't rise the black level with 2.2 gamma preset so I set it to 2 as a compromise, setting it to 3 raised the black luminance, but to a lesser degree than setting brightness to 54.
Then what I did was to measure the black luminance on my macbook lcd which I use as a reference, it turned out to be around 0.087 cd/m2. Then in HCFR gamma preferences tab I checked the box "override black" and entered this value in the field. This way HCFR now thinks that c9 has the same black luminance as the lcd and won't even give you 0% pattern to measure and will build the gamma curve using this custom black luminance.
Then I measured and aligned the gamma curve to 2.2 using 22 points.
Then I measured the near black luminance with 7 points, this showed that 0% through 5% still tracked far below 0.087 cd/m2, that's basically where the oled "black crush" is. So I rised ire 5% luminance to 7 and ire 7.5% to 6.
this put luminance levels from 4% through 10% to above the lcd black level luminance of 0.087 cd/m2. The low end below 4% was still below 0.087 cd/m but rising ire 5 luminance above 7 would put 7%-10% way below 2.2 gamma without rising below 4% luminance as much so I figured this would be a good compromise.
This is it, looking at my c9 and macbook side by side they now look almost identical even in terms of low level detail, while c9 is still a bit darker because <4% is still below the lcd black level this is I think pretty much the best you can do without finer low end luminance controls without rising black level above 0 cd/m2 or messing up above 5%.
You can match oled to an lcd completely by rising black level to an lcd black luminance around 0.9 cd/m2 using brightness controls and then realigning the gamma with this new black point. But then what's the point of OLED in the first place considering that the OLED viewing angles are **** compared to a good IPS anyway?
TLDR:
using black luminance override in hcfr to put 0-7% grey above the lcd black luminance without rising black level
brightness 53
ire 2.5 luminance 2
ire 5 luminance 7
ire 7 luminance 6
ps would love to hear what I did wrong and if all of this isn't correct considering that I only recently got into tv calibration and only just learning after getting an lg c9 and being blown away how terrible the out of box color accuracy is, and I probably have no idea wtf I am doing
1 - 5 of 5 Posts
Patterns don't work well and can mislead you into thinking you have a black crush issue when you really don't. I have had sets, plasma and OLED that would not show bar 17 or 1 (whatever pattern you use). This would lead you to believe there is a black crush problem, however, I have found that with actual content in many cases there is no black crush issue. Also on most OLED you will never get perfect shadow detail, it will never be perfect. The best thing a person can do is stop chasing perfection with an OLED. Its great tech but it also has flaws.
When you want to check your near black and have an X-Rite meter, you must use a pattern with flashing bars.
When you have a high-end colorimeter, you can create a custom patchlist with 8 or 10-bit code values in ColourSpace (10-bit if you have Accupel 6000, Murideo SEVEN, or Pi4 1.3 ISO), and after using a pattern with flashing bars, you can evaluate the near-black performance with some measurements about how different Brightness settings are affecting your near black.
The calibration controls of the low end can affect your near black also.
When you look at a pattern with a flashing bar, watch the TV from a close distance and keep an eye to prevent a 'glowing' or lift of your native blacks level.
For PGenerator, I have removed the text indications from brightness patterns with flashing bars, and a lot of users have found that they are working better.
Its more important to retain native black level and prevent 'glowing blacks' than trying to see 17 flashing bar.
When you have a high-end colorimeter, you can create a custom patchlist with 8 or 10-bit code values in ColourSpace (10-bit if you have Accupel 6000, Murideo SEVEN, or Pi4 1.3 ISO), and after using a pattern with flashing bars, you can evaluate the near-black performance with some measurements about how different Brightness settings are affecting your near black.
The calibration controls of the low end can affect your near black also.
When you look at a pattern with a flashing bar, watch the TV from a close distance and keep an eye to prevent a 'glowing' or lift of your native blacks level.
For PGenerator, I have removed the text indications from brightness patterns with flashing bars, and a lot of users have found that they are working better.
Its more important to retain native black level and prevent 'glowing blacks' than trying to see 17 flashing bar.
"Its more important to retain native black level and prevent 'glowing blacks' than trying to see 17 flashing bar"When you want to check your near black and have an X-Rite meter, you must use a pattern with flashing bars.
When you have a high-end colorimeter, you can create a custom patchlist with 8 or 10-bit code values in ColourSpace (10-bit if you have Accupel 6000, Murideo SEVEN, or Pi4 1.3 ISO), and after using a pattern with flashing bars, you can evaluate the near-black performance with some measurements about how different Brightness settings are affecting your near black.
The calibration controls of the low end can affect your near black also.
When you look at a pattern with a flashing bar, watch the TV from a close distance and keep an eye to prevent a 'glowing' or lift of your native blacks level.
For PGenerator, I have removed the text indications from brightness patterns with flashing bars, and a lot of users have found that they are working better.
Its more important to retain native black level and prevent 'glowing blacks' than trying to see 17 flashing bar.
I agree 100%
I usually end up with bar 18 just barely visible if visible at all. On both OLED and plasma (my preference) it is very difficult to achieve reference black without making sacrifices in other areas of the picture quality or without raising black level. In my opinion a small amount of black crush on test patterns is fine because 99% of real content will not utilize the very low scale of blacks.
You can use shadow detail in the ddc menus of autocal if you have that program. It raises just the near blacks. It would be easier to do that if you use calman.
1 - 5 of 5 Posts
- This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
AVS Forum
A forum community dedicated to home theater owners and enthusiasts. Come join the discussion about home audio/video, TVs, projectors, screens, receivers, speakers, projects, DIY’s, product reviews, accessories, classifieds, and more!
Full Forum Listing
Top Contributors this Month
View All
WilliamR
833 Replies
BASSAHOLIC
488 Replies
mrtickleuk
404 Replies
Recommended Communities
