AVS Forum banner

1 - 20 of 70 Posts

·
Registered
Joined
·
137 Posts
Discussion Starter #1 (Edited)
I just received my HDFury today and am working on calibrating my Samsung Q70R HDR. Is this the correct settings to send the proper metadata?

Isn't it supposed to output legal and not video range? I got the metadata from here...https://www.avsforum.com/threads/dedicated-raspberry-pi-pgenerator-thread-set-up-configuration-updates-special-features-general-usage-tips.3167475/#post-60090109

Do I use DCI-P3 or BT2020 metadata?

3109921
3109922
3109923
 

·
Registered
Joined
·
1,605 Posts
You want to use 1000 nit metadata to avoid having the Q70 inject its own tone mapping. AFAIK the mastering primaries are ignored by the Q70 (you can take measurements both ways to confirm), but you might as well use DCI-P3 as I believe that's the most common for actual content.
 

·
Registered
Joined
·
1,605 Posts
Interesting, your peak brightness is higher than rtings measured (488) and lower than flatpanelshd measured (829). Did you leave Contrast at 50, or lower it some?

(Just my $.02, it's always better to take 21pt readings, to be more sure you don't miss a hidden bump.)
 

·
Registered
Joined
·
137 Posts
Discussion Starter #7
Interesting, your peak brightness is higher than rtings measured (488) and lower than flatpanelshd measured (829). Did you leave Contrast at 50, or lower it some?

(Just my $.02, it's always better to take 21pt readings, to be more sure you don't miss a hidden bump.)
This is with contrast at 48. I'm currently only doing 2pt. I'll dig into the 20pt. this weekend. I just had the 10 pt. on to see a better representation of how the 2pt. effects the rest of the graph.

Does it matter with the 82"? I think the 82" used a different panel than the other sizes? Maybe. I can't recall.

There's a lot of error up the middle and 2pt. isn't fixing that any.

3109998
 

·
Registered
Joined
·
137 Posts
Discussion Starter #8
Interesting, your peak brightness is higher than rtings measured (488) and lower than flatpanelshd measured (829). Did you leave Contrast at 50, or lower it some?

(Just my $.02, it's always better to take 21pt readings, to be more sure you don't miss a hidden bump.)
CMS is tracking properly now.
 

·
Registered
Joined
·
1,605 Posts
Ah, rtings measured a 55", flatpanelshd measured a 65", so the 82" could easily be different from both.
 

·
Registered
Joined
·
1,605 Posts
You might want to take a full set of HDR measurements with PGen & HD Fury running 1080p24, just to make sure there's no difference with 1080p60.
 

·
Registered
Joined
·
137 Posts
Discussion Starter #12
You might want to take a full set of HDR measurements with PGen & HD Fury running 1080p24, just to make sure there's no difference with 1080p60.
When I try to inject the 1080p24 metadata the TV still says it's in 1080p60.
 

·
Registered
Joined
·
137 Posts
Discussion Starter #13 (Edited)
Got it through device control for pgen lol. Somewhere along the lines though. I feel like I'm using the wrong settings. Grayscale tracks correct at 1080p24. Color tracks better at 1080p60.
 

·
Registered
Joined
·
1,605 Posts
Just to check, you changed PGen to 24 and also changed the HD Fury AVI IF string to 24, right?

My Q90R tracks consistently for HDR at 1080p24, 1080p60 and 2160p24, where it goes wonky is at 2160p60, you can see an example in that post. The Q70 is mid-range, so I guess it's possible it differs between 1080p24 and 1080p60. If you can get your laptop to output 2160p24 and 2160p60, then you might try measuring all four output types using madTPG, to see how consistent they are with each other. You should also compare PGen SDR between 1080p24 and 1080p60; my guess is they will be consistent. (2160p60 SDR tracks fine on my Q90R.) Assuming that's the case, and assuming you've configured HDR correctly, then I'd suggest doing all of your calibration using 1080p24, because 24fps is most common for HDR. (And make sure your source devices and AVR are doing frame rate passthrough, not converting up to 60fps.)
 

·
Registered
Joined
·
1,605 Posts
Oh, one other random tidbit. With my Q90R, when doing the default ColorChecker read series in HDR, the Light Skin measurement always results in a high dE, so afterwards I go back to do a single read for that one color to get a more reasonable measurement. Some weird quirk of the Q90R; maybe the Q70 has a similar quirk.
 

·
Registered
Joined
·
137 Posts
Discussion Starter #17
Just to check, you changed PGen to 24 and also changed the HD Fury AVI IF string to 24, right?

My Q90R tracks consistently for HDR at 1080p24, 1080p60 and 2160p24, where it goes wonky is at 2160p60, you can see an example in that post. The Q70 is mid-range, so I guess it's possible it differs between 1080p24 and 1080p60. If you can get your laptop to output 2160p24 and 2160p60, then you might try measuring all four output types using madTPG, to see how consistent they are with each other. You should also compare PGen SDR between 1080p24 and 1080p60; my guess is they will be consistent. (2160p60 SDR tracks fine on my Q90R.) Assuming that's the case, and assuming you've configured HDR correctly, then I'd suggest doing all of your calibration using 1080p24, because 24fps is most common for HDR. (And make sure your source devices and AVR are doing frame rate passthrough, not converting up to 60fps.)
I did set both to 1080p24. I get inconsistent results across everything. I even received different results when I plugged everything back in this morning using the same settings.

When I do 2pt on manual, I must do signal read at 30% or it does a double read and gives me the 25% reading on full screen.
 

·
Registered
Joined
·
1,605 Posts
Are you getting inconsistent results with SDR too, or only HDR? What are your meter exposure settings? For HDR, is Full Field Pattern Insertion (under Application Measurement Options) enabled or disabled?
 

·
Registered
Joined
·
137 Posts
Discussion Starter #19
Are you getting inconsistent results with SDR too, or only HDR? What are your meter exposure settings? For HDR, is Full Field Pattern Insertion (under Application Measurement Options) enabled or disabled?
I get consistent results across sdr via Pgenerator, MadTPG and Client3. SDR tracks fine at both 24 and 60.

Standard exposure is 1.5 secs and low light handler is 3secs below 1 nit.
 

·
Registered
Joined
·
1,605 Posts
Is the inconsistency across the entire grayscale and colors, or more specific?

In my experience with the Manual Calibration workflow in HDR, CalMAN silently enables Full Screen Pattern Insertion behind your back when you get to the Meter Positioning page. If you've been working with it enabled, you might try disabling it to see if it makes any difference.
 
1 - 20 of 70 Posts
Top