Projector Calibration (6040UB) - Second Look - Page 4 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 76Likes
Reply
 
Thread Tools
post #91 of 244 Old 07-05-2019, 01:29 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
I decided to disable the screen offset when adjusting the two point controls & the enabled it again for adjusting the gamma controls.


This is the best I can do with maxing out the gamma & contrast controls. Still hard clipping at 60 nits if I am correct?
Gregg Loewen’s reply says “This is the value you will enter in the large Y value for screen offset. So every measurement you perform will be multiplied by 7.“ but I don’t see that happening in your screenshot. What did you see that’s different, when you turn screen offset on and off?

EDIT: Just saw your reply regarding not applying screen offset correctly (y instead of Y). No wonder the grey scale is off, and the luminance didn’t scale. Will wait for updated result before commenting.

Looks like using screen offset Y is a kludge for getting a multiplier for HDR. No wonder not everyone agrees with it.

The multiplier/scale factor is used by LightSpace and HCFR. All the JVC projector calibrations also use it. You should check with Tyler to see if there’s anything equivalent in CalMAN.
AdamAttewell likes this.

Last edited by Dominic Chan; 07-05-2019 at 01:44 PM.
Dominic Chan is online now  
Sponsored Links
Advertisement
 
post #92 of 244 Old 07-05-2019, 01:56 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
Gregg Loewen’s reply says “This is the value you will enter in the large Y value for screen offset. So every measurement you perform will be multiplied by 7.“ but I don’t see that happening in your screenshot. What did you see that’s different, when you turn screen offset on and off?

EDIT: Just saw your reply regarding not applying screen offset correctly (y instead of Y). No wonder the grey scale is off, and the luminance didn’t scale. Will wait for updated result before commenting.

Looks like using screen offset Y is a kludge for getting a multiplier for HDR. No wonder not everyone agrees with it.

The multiplier/scale factor is used by LightSpace and HCFR. All the JVC projector calibrations also use it. You should check with Tyler to see if there’s anything equivalent in CalMAN.
OK so after using the correct Y for screen offset grayscale is back in line & I went ahead & adjusted the gamma controls.

I ended up increasing the contrast to 60 from 50 as maxing out the gamma controls I could not get close to the curve where it rolls off so I increased the contrast & used the gamma controls to bring the other parts of the curve into line.

I could not do anything about the bottom of the curve as the gamma controls are pretty useless at the ends of the curve. I just ended up increasing error trying to get it inline so I left it alone

Not looked at the CMS yet, just wanted to see if I am on the right track first.

One thing did spring to mind, up to this point I have not been adjusting the brightness control, should I be doing so with a pattern before starting?


If this is the case I will need to start all over as I am sure this will effect everything else.


If I can proceed to CMS adjustments shall I do so with the Screen Offset & the White Point at 524 nits?

Last edited by AdamAttewell; 07-05-2019 at 02:02 PM.
AdamAttewell is offline  
post #93 of 244 Old 07-06-2019, 05:41 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
I went ahead & watched some content with the Screen Offset calibration & it was way to dim.

I would say it was even darker than when you engage the color filter, its just way to dark.

I then tried a Screen Offset of 3 with a White Level of 393 nits & it was still to dark, only thing I can think of now is trying a Screen Offset of 2 & a White Level of 262 to see what that looks like?

I have been watching content from the calibration I did with the White Level set to the peak luminance of the display (131 nits) & its looks really good. It's very bright but the only thing that concerns me is that I am clipping a lot of the brighter content.

I have been using The Revenant to see how things are looking & I am pleased with effect that curve creates but I then watched some of The Grand Tour on Amazon & whites seemed blown out.

@WiFi-Spy Tyler is there any other way I can adjust the EOTF curve so hard clipping does not occur so soon? This Screen Offset method seems to be very hit & miss.

Last edited by AdamAttewell; 07-06-2019 at 07:12 PM.
AdamAttewell is offline  
Sponsored Links
Advertisement
 
post #94 of 244 Old 07-06-2019, 06:24 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
I have been watching content from the calibration I did with the White Level set to the peak luminance of the display (131 nits) & its looks really good. It's very bright but the only thing that concerns me is that I am clipping a lot of the brighter content.
If your peak luminance is 131 nits and you if use 100 nits for diffuse white without a multiplier, there’s only 31 nits left for highlights. HDR highlights go to 1000, 4000, or even 10,000 nits.

How does the luminance curve look? It may be possible to avoid “hard clipping” but you won’t get much gradation in the highlights.
AdamAttewell likes this.
Dominic Chan is online now  
post #95 of 244 Old 07-06-2019, 07:15 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
If your peak luminance is 131 nits and you if use 100 nits for diffuse white without a multiplier, there’s only 31 nits left for highlights. HDR highlights go to 1000, 4000, or even 10,000 nits.

How does the luminance curve look? It may be possible to avoid “hard clipping” but you won’t get much gradation in the highlights.
Are we talking about the luminance curve with the screen offset Y set too 4 & white level at 524 nits?

What does a screen offset of 4 give me for diffuse white?

I am getting really confused with this multiplier.

Last edited by AdamAttewell; 07-06-2019 at 07:45 PM.
AdamAttewell is offline  
post #96 of 244 Old 07-06-2019, 08:35 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
Are we talking about the luminance curve with the screen offset Y set too 4 & white level at 524 nits?

What does a screen offset of 4 give me for diffuse white?

I am getting really confused with this multiplier.
No, I was to the part I quoted, i.e., “with the White Level set to the peak luminance of the display (131 nits) & its looks really good“. That is equivalent to turning off the multiplier.

If you set the screen offset to 4 and turn on tone mapping, you shouldn’t be seeing much hard clipping, at least for 1000 nits titles. The overall brightness will be lower with a screen offset >1.

If you prefer a bright overall picture, you may want to try one of the settings floating around in the 5040UB thread (Hypervision etc). Also, many people like the SDR2020 mapping of the Panasonic UB420/820 players.
AdamAttewell likes this.

Last edited by Dominic Chan; 07-07-2019 at 08:15 AM.
Dominic Chan is online now  
post #97 of 244 Old 07-07-2019, 08:34 AM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
No, I was to the part I quoted, i.e., “with the White Level set to the peak luminance of the display (131 nits) & its looks really good“. That is equivalent to turning off the multiplier.

If you set the screen offset to 4 and turn on tone mapping, you shouldn’t be seeing much hard clipping, at least for 1000 nits titles. The overall brightness will be lower with a screen offset >1.

If you prefer a bright overall picture, you may want to try one of the settings floating around in the 5040UB thread (Hypervision etc). Also, many people like the SDR2020 mapping of the Panasonic UB420/820 players.
I was starting to think that about Harper Vision as it seems if you try to calibrate these Epson's "normally" for HDR the overall image brightness is very dim.


Can you tell me if my understanding of the screen offset/multiplier is correct.

When using a screen offset of 4, let say the meter reads 100 nits CalMAN takes that reading & multiplies it by 4 so it turns into 400 nits?

Then multiplying the white level by four allows us to adjust the EOTF curve to roll off later allowing for clipping to occur much later. So without the multiplier hard clipping occurs around 60 nits? But with it hard clipping occurs around 100nits?

So we just essentially fooling the software into thinking we are measuring a higher nit level than we actually are?

Last edited by AdamAttewell; 07-07-2019 at 08:50 AM.
AdamAttewell is offline  
post #98 of 244 Old 07-07-2019, 08:50 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
So we just essentially fooling the software into thinking we are measuring a higher nit level than we actually are?
That is how the multiplier works, but it not as ridiculous as it sounds. Even for SDR calibration, the white level is generally 100-200 nits for direct view TVs, but 50 nits for projectors in dim viewing environments, so that’s a multiplier of 4.

HDR PQ curve is supposed to be “absolute” but many people find 100 nits diffuse white too bright for projectors (even setting aside the fact that highlights willl be limited).
Dominic Chan is online now  
post #99 of 244 Old 07-07-2019, 09:14 AM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
That is how the multiplier works, but it not as ridiculous as it sounds. Even for SDR calibration, the white level is generally 100-200 nits for direct view TVs, but 50 nits for projectors in dim viewing environments, so that’s a multiplier of 4.

HDR PQ curve is supposed to be “absolute” but many people find 100 nits diffuse white too bright for projectors (even setting aside the fact that highlights willl be limited).
Just so I am on the same page diffuse white is for our purposes the 50% grayscale point?

I will try again with an offset of 4 & a white level of 524 but everything on screen just looked really dim, even things that would be bright like the sun.

So what are settings like Harper Vision doing to get a brighter picture?

I presume using the SDR setting under dynamic range plays a big part in that the projector isn't using its internal tone mapping?

I have owned this projector for a few years now & over the last year I have been using others settings like Harper Vision but I have always wondered if I could do better my self, seems not

If I were to go back to using other peoples settings as a baseline would I follow the same steps as I have been doing but without the multiplier?

I was thinking of loading in the Harper Vision settings & then using BT 2390 & D65 P3 lining up the grayscale, gamma & then the CMS to see what it looks like?
AdamAttewell is offline  
post #100 of 244 Old 07-07-2019, 10:08 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
Just so I am on the same page diffuse white is for our purposes the 50% grayscale point?
In my previous posts I was referring to the 50% input level as diffuse white, but recent ITU reports (e.g. BT.2408) use 203 nits (58% input) for diffuse white. However, the actually value is not important for the current discussion, as the multiplier is applied across the entire curve (e.g., with a multiplier of 4, 25 measured nits is equivalent to 100 nits on the PQ curve, 50 nits is equivalent to 200 nits, etc.)

Quote:
I will try again with an offset of 4 & a white level of 524 but everything on screen just looked really dim, even things that would be bright like the sun.
The peak luminance of your projector is fixed at 130 nits (actual), regardless of the multiplier. Which scene were you looking at?

Quote:
So what are settings like Harper Vision doing to get a brighter picture?
I have no idea. If you end up using it, I would be interested to see what the luminance curve looks like.

Quote:
I presume using the SDR setting under dynamic range plays a big part in that the projector isn't using its internal tone mapping?
The projector won't use its internal tone mapping if you're using a custom curve.
Attached Thumbnails
Click image for larger version

Name:	ST2084PQ.jpg
Views:	12
Size:	143.3 KB
ID:	2588476  
Dominic Chan is online now  
post #101 of 244 Old 07-07-2019, 10:18 AM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
The peak luminance of your projector is fixed at 130 nits (actual), regardless of the multiplier. Which scene were you looking at?

The battle scene at the start of The Revenant has some shots of the sun. Even when the tree that has been set of fire comes crashing down it looks very dim.

Last edited by AdamAttewell; 07-07-2019 at 11:04 AM.
AdamAttewell is offline  
post #102 of 244 Old 07-07-2019, 01:56 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Had a look at the grayscale with Harper Vision settings.



Had a go at adjsuting the gamma controls to match the EOTF curve & well I wasted my time to say the least Still you only learn by your mistakes.


I wonder what these guys were targeting when dailing in their settings.
AdamAttewell is offline  
post #103 of 244 Old 07-07-2019, 02:32 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
Had a look at the grayscale with Harper Vision settings.

I wonder what these guys were targeting when dailing in their settings.
What mode did you use? I don’t think Hypervision is meant to go with 60 nits peak.
Dominic Chan is online now  
post #104 of 244 Old 07-07-2019, 02:47 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
What mode did you use? I don’t think Hypervision is meant to go with 60 nits peak.

Just BT 2390 I had no screen offset active & I was using the "used measured white level" option.
AdamAttewell is offline  
post #105 of 244 Old 07-07-2019, 02:50 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
Just BT 2390 I had no screen offset active & I was using the "used measured white level" option.
I was asking about the Epson’s picture mode.

EDIT: Apparently he uses Digital Cinema, High Power.

Last edited by Dominic Chan; 07-07-2019 at 03:00 PM.
Dominic Chan is online now  
post #106 of 244 Old 07-07-2019, 03:05 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
I was asking about the Epson’s picture mode.

EDIT: Apparently he uses Digital Cinema, High Power.

Yes Digital Cinema, high power consumption. I drop this down to medium as the projector is above my head.


I have tried both & while I agree it does look better on high I can live with medium.


I just wanted to check to see how accurate these settings are and bring them into line for my setup.
AdamAttewell is offline  
post #107 of 244 Old 07-07-2019, 03:10 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
Yes Digital Cinema, high power consumption. I drop this down to medium as the projector is above my head.
You’re getting only 60 nits even with Medium Power? I would switch to Bright Cinema instead.

Last edited by Dominic Chan; 07-07-2019 at 03:17 PM.
Dominic Chan is online now  
post #108 of 244 Old 07-07-2019, 06:06 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
You’re getting only 60 nits even with Medium Power? I would switch to Bright Cinema instead.

I would agree, but what if I were to tell you that using these settings the image "appears" brighter than the Bright Cinema calibration I did with the screen offset/multiplier active? This mode also uses the colour filter hence the dramatic drop in nits but at least I get the benefit of the expanded colour gamut.


Also using the Dynamic Range Adjustment slider on the Panasonic UB820 allows me to increase the overall image brightness without crushing blacks or blowing out whites if use sparingly.
AdamAttewell is offline  
post #109 of 244 Old 07-07-2019, 06:27 PM
AVS Forum Special Member
 
WiFi-Spy's Avatar
 
Join Date: Feb 2004
Location: Seattle, WA
Posts: 3,881
Mentioned: 233 Post(s)
Tagged: 0 Thread(s)
Quoted: 1603 Post(s)
Liked: 2192
Unless you are getting 120+ nits of peak brightness, it is kinda a waste to try to calibrate an “HDR” mode. Honestly I don’t really consider any of these projectors actually “HDR”, Yes they are better than regular SDR, but even Dolby considers Dolby Cinema to be EDR , not HDR.

I don’t think anyone has completely narrowed in on how these projectors should be set up when it comes to EOTF. I really think the projector manufacturers really need to invest R&D into advanced dynamic tone mapping algorithms. I don’t think we’re ever going to be at the point where one static tone map or even two, are appropriate for all HDR movies. For the time being BT.2390 Is a good place to start, but it’s not appropriate for every set up.

I think it’s telling that there is no Dolby Vision or HDR10+ home theater projectors yet. This is not an easy nut to crack.

Taking content that was mastered on a 31 inch 1000 nit direct view display, and having that look good on 200 nit Projector in a black room, while maintaining artistic intent, is not trivial. The TV manufacturers have a hard enough time doing that on a 500 nit display.
latet likes this.

Tyler Pruitt - Technical Evangelist - for CalMAN

10 Bit Gradient Test Patterns (HEVC) - Free Download
WiFi-Spy is offline  
post #110 of 244 Old 07-07-2019, 07:10 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by WiFi-Spy View Post
Unless you are getting 120+ nits of peak brightness, it is kinda a waste to try to calibrate an “HDR” mode. Honestly I don’t really consider any of these projectors actually “HDR”, Yes they are better than regular SDR, but even Dolby considers Dolby Cinema to be EDR , not HDR.

I don’t think anyone has completely narrowed in on how these projectors should be set up when it comes to EOTF. I really think the projector manufacturers really need to invest R&D into advanced dynamic tone mapping algorithms. I don’t think we’re ever going to be at the point where one static tone map or even two, are appropriate for all HDR movies. For the time being BT.2390 Is a good place to start, but it’s not appropriate for every set up.

I think it’s telling that there is no Dolby Vision or HDR10+ home theater projectors yet. This is not an easy nut to crack.

Taking content that was mastered on a 31 inch 1000 nit direct view display, and having that look good on 200 nit Projector in a black room, while maintaining artistic intent, is not trivial. The TV manufacturers have a hard enough time doing that on a 500 nit display.
I completely agree that these projectors are by no means HDR capable but they can be "calibrated" to produce a great looking image.

Whether that image is "accurate" or not is another question.

I am by no means expecting specular highlights & the like, I just want a fairly bright accurate image that uses the wider color gamut if possible.

I could go down the SDR BT.2020 route which seems the logical choice if I can get a bright enough image with the color filter in place which seems to be the case with other peoples settings I have tried.

I suppose if you want to cut to the chase I am happy with other users settings I have applied I would just like a way to verify they are as accurate as possible but I guess if they are not conforming to any known "standard" then this is impossible.
AdamAttewell is offline  
post #111 of 244 Old 07-07-2019, 07:44 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
I would agree, but what if I were to tell you that using these settings the image "appears" brighter than the Bright Cinema calibration I did with the screen offset/multiplier active?
I’m not surprised at all. If you look at my signature, there are links to “bright curves” for low nits JVC projectors (50-60 nits peak). All you need to do is to keep diffuse white (and below) at a high level, the trade-off being (relatively) diminished highlights. ST2390 even compensates the low end to enhance shadow details.

Here’s are some comparison shots between 75 nits peak and 120 nits peak, taken with the same camera exposure. You can see how close the overall images can get:
https://www.avsforum.com/forum/24-di...l#post55874834

Quote:
This mode also uses the colour filter hence the dramatic drop in nits but at least I get the benefit of the expanded colour gamut.
Most owners would rather have 130 nits for HDR, without the filter. It’s up to each individual. Note that the low luminance curve can lose colour saturation for bright scenes, in spite of the colour filter.

Assuming the curve in your post 102 is Hypervision with 60 nits peak, it gives you only 22 nits at 50%, comparable to a multiplier or 4~5.

Quote:
Also using the Dynamic Range Adjustment slider on the Panasonic UB820 allows me to increase the overall image brightness without crushing blacks or blowing out whites if use sparingly.
If I only get 60 nits I would use the UB820 in the SDR2020 mode; i.e., let the UB820 handle all the HDR tone mapping. In the HDR mode the UB820 has absolutely no knowledge of your projector’s tone mapping, and can’t perform proper tone mapping
AdamAttewell likes this.

Last edited by Dominic Chan; 07-08-2019 at 07:34 AM.
Dominic Chan is online now  
post #112 of 244 Old 07-08-2019, 09:02 AM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
I’m not surprised at all. If you look at my signature, there are links to “bright curves” for low nits JVC projectors (50-60 nits peak). All you need to do is to keep diffuse white (and below) at a high level, the trade-off being (relatively) diminished highlights. ST2390 even compensates the low end to enhance shadow details.

Here’s are some comparison shots between 75 nits peak and 120 nits peak, taken with the same camera exposure. You can see how close the overall images can get:
https://www.avsforum.com/forum/24-di...l#post55874834


Most owners would rather have 130 nits for HDR, without the filter. It’s up to each individual. Note that the low luminance curve can lose colour saturation for bright scenes, in spite of the colour filter.

Assuming the curve in your post 102 is Hypervision with 60 nits peak, it gives you only 22 nits at 50%, comparable to a multiplier or 4~5.


If I only get 60 nits I would use the UB820 in the SDR2020 mode; i.e., let the UB820 handle all the HDR tone mapping. In the HDR mode the UB820 has absolutely no knowledge of your projector’s tone mapping, and can’t perform proper tone mapping
So if I understand correctly you would go back to Bright Cinema mode & using BT 2390 with a screen offset of 4 & the white level at 524 nits calibrate from there?

Or do I need to play around with the screen offset & the multiplier added to the white level until I get an acceptable amount of nits at the 50% point?

What would you consider this value to be with the multiplier active??

Last edited by AdamAttewell; 07-08-2019 at 09:56 AM.
AdamAttewell is offline  
post #113 of 244 Old 07-08-2019, 01:42 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
So if I understand correctly you would go back to Bright Cinema mode & using BT 2390 with a screen offset of 4 & the white level at 524 nits calibrate from there?

Or do I need to play around with the screen offset & the multiplier added to the white level until I get an acceptable amount of nits at the 50% point?

What would you consider this value to be with the multiplier active??
With the multiplier active the value should come out to be about 94 nits at 50%.
I use a multiplier of 4 (23.5 nits at 50%) myself, but your previous post seemed to indicate that you found it too dim. On the other hand, you were happy with Harpervision’s diffuse white which is only 22 nits, so I’m not sure what to suggest.

Maybe SDR2020 is the best way to go.
AdamAttewell likes this.

Last edited by Dominic Chan; 07-08-2019 at 01:48 PM.
Dominic Chan is online now  
post #114 of 244 Old 07-08-2019, 02:09 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
With the multiplier active the value should come out to be about 94 nits at 50%.
I use a multiplier of 4 (23.5 nits at 50%) myself, but your previous post seemed to indicate that you found it too dim. On the other hand, you were happy with Harpervision’s diffuse white which is only 22 nits, so I’m not sure what to suggest.

Maybe SDR2020 is the best way to go.

I will give the offset of 4 with a white level of 524 nits another go to make sure I did not screw anything up the first time.


That is what I am struggling to understand myself as to why these brighter curves seem to provide the same luminance when measured but appear brighter when viewed.
AdamAttewell is offline  
post #115 of 244 Old 07-08-2019, 03:53 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
I will give the offset of 4 with a white level of 524 nits another go to make sure I did not screw anything up the first time.


That is what I am struggling to understand myself as to why these brighter curves seem to provide the same luminance when measured but appear brighter when viewed.
50% is just a single data point on the curve. When analyzing other (non-standard) curves, you need to find a multiplier that would make the 50% measurement match the target, and look at the rest of the curve. A “proper” HDR curve should follow the ST.2084 PQ curve up to at least the diffuse white level (possibly with a multiplier). The highlight tracking is less critical and can be customized according to your preference, deviating from ST2390.
AdamAttewell likes this.

Last edited by Dominic Chan; 07-08-2019 at 05:31 PM.
Dominic Chan is online now  
post #116 of 244 Old 07-08-2019, 11:00 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
50% is just a single data point on the curve. When analyzing other (non-standard) curves, you need to find a multiplier that would make the 50% measurement match the target, and look at the rest of the curve. A “proper” HDR curve should follow the ST.2084 PQ curve up to at least the diffuse white level (possibly with a multiplier). The highlight tracking is less critical and can be customized according to your preference, deviating from ST2390.
So I have been experimenting & I think I have found some settings that give me a bright enough image.

I tried the offset at 4 again with the multiplier applied to the white level but still found the resulting image to dim.

I decided to try the HDR mode 1 instead of HDR mode 2 & after checking contrast, brightness I went ahead with the grayscale & tried my best to match the EOTF curve.

I watched some content & am happy with how it looks, here is the grayscale. This is with the offset & multiplier in place.

I have not done any CMS adjustments yet as I wanted to see what content looked like first.

One question I do have is with white clipping, should I be targeting around 1000 nits by adjusting the contrast control?

I could use the contrast control to get the EOTF curve closer but would this not lower the point at which clipping occurs?

Last edited by AdamAttewell; 07-09-2019 at 04:28 PM.
AdamAttewell is offline  
post #117 of 244 Old 07-09-2019, 08:10 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Was unsure how to proceed so went ahead & set contrast so level 724 just barely flashes & brightness with level 68 barely flashing I touched up the grayscale & EOTF again. I could not get it as good as I did before which is disappointing but I presume this is from lowering the contrast further.



CMS was pretty good with only magenta not falling under a delta E of 2 at all the 25, 50 & 80% points.



ColorChecker does not look to bad either.



I was back to back watching the SDR & HDR versions of The Revenant & I will be honest I could not tell much of a difference in terms of HDR effect so I am going to have a go at an SDR BT2020 calibration now.

WiFi-Spy likes this.
AdamAttewell is offline  
post #118 of 244 Old 07-09-2019, 08:29 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
Was unsure how to proceed so went ahead & set contrast so level 724 just barely flashes & brightness with level 68 barely flashing[SIZE=3] I touched up the grayscale & EOTF again. I could not get it as good as I did before which is disappointing but I presume this is from lowering the contrast further.
With tone mapping the slope at 1000 nits (level 724) is so shallow that it’s almost impossible to see the flashing. Looks like in order to see the flashing you reduced the Contrast so much that it’s falling off the curve. I would just follow the target curve and not worry about the flashing.

Quote:
I was back to back watching the SDR & HDR versions of The Revenant & I will be honest I could not tell much of a difference in terms of HDR effect so I am going to have a go at an SDR BT2020 calibration now.
What multiplier did you end up using? Was the overall brightness of the HDR version comparable to the SDR version?
AdamAttewell likes this.

Last edited by Dominic Chan; 07-09-2019 at 08:35 PM.
Dominic Chan is online now  
post #119 of 244 Old 07-09-2019, 10:03 PM - Thread Starter
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 987
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 766 Post(s)
Liked: 161
Quote:
Originally Posted by Dominic Chan View Post
With tone mapping the slope at 1000 nits (level 724) is so shallow that it’s almost impossible to see the flashing. Looks like in order to see the flashing you reduced the Contrast so much that it’s falling off the curve. I would just follow the target curve and not worry about the flashing.

What multiplier did you end up using? Was the overall brightness of the HDR version comparable to the SDR version?
Damn, I wish I had not looked at that white clipping pattern now

I used a screen offset of 4 with a white level of 524 nits.

I would say yes, the SDR looks very close to the HDR version in terms of brightness. I was wondering if I should try a different multiplier to get the image to be a little brighter?
AdamAttewell is offline  
post #120 of 244 Old 07-09-2019, 10:17 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,115
Mentioned: 133 Post(s)
Tagged: 0 Thread(s)
Quoted: 5007 Post(s)
Liked: 1927
Quote:
Originally Posted by AdamAttewell View Post
I used a screen offset of 4 with a white level of 524 nits.

I would say yes, the SDR looks very close to the HDR version in terms of brightness.
What changed? In post 99 you said “everything looked really dim” with a multiplier of 4.

Quote:
I was wondering if I should try a different multiplier to get the image to be a little brighter?
HDR is not supposed to look overall brighter than SDR. If you use a lower multiplier to boost the overall brightness, the picture will look less like HDR as the highlights will be dimmer.
Dominic Chan is online now  
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