Forum Jump: 
 121Likes
Reply
 
Thread Tools
post #61 of 1081 Old 11-27-2017, 05:30 AM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
Quote:
Originally Posted by viperlogic View Post
Unfortunately with the Epson 5040 we are very limited in our controls, we only have two point control of offset and gain for RGB. For gamma we have 9, known as colour tone 1-9. Then for colour we have hue, sat, and brightness for each of the RGBCYM colours
ok, so the two point would be used to do gray scale tracking after matching the 2084 curve as best you can, hue/sat/brightness to align CMS around 92 nits, and then contrast/9-point to tailor curve to one of the tone mapping suggestions.
viperlogic likes this.
zoyd is offline  
Sponsored Links
Advertisement
 
post #62 of 1081 Old 11-27-2017, 05:46 AM - Thread Starter
Senior Member
 
Join Date: Jul 2017
Posts: 225
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 197 Post(s)
Liked: 27
Something else I'm struggling with is the concept of the "multiplier" as discussed here http://www.lightillusion.com/hdr_calibration.html

If its scaled down and a curve produced for that and calibrated to, how does the projector know this eg there is no way to tell the projector that your peak is x and therefore treat this peak as level 940 and scale all content to this level. Eg content is mastered to 1000 nits, a bunch of pixels are at level 609 which is 300 nits, the projector cannot produce that, where as a scale value would be 110 nits which the projector can.

Epson EH-TW7300, 104" borderless ReAct 3, Denon AVR-X2300W in 5.1.2, Cambridge Audio Minx Min21's & Min12's, Monitor Audio Pro-65, BK P12-300 sub, Nvidia Shield, Vero 4K, Panasonic UB400, SKY Q, PS3, Wii, RPi, LG 42LS575T, Denon Heos soundbar, Amazon Echo, Logitech Harmony Elite
viperlogic is offline  
post #63 of 1081 Old 11-27-2017, 05:58 AM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
That multiplier is similar to your contrast control which you can set such that an incoming 1000 nit code (723) is converted to your peak luminance of 140 nits. All codes under 723 will then also produce luminance levels your projector can display and the recommendation for best perceptual mapping of these codes is to use a curve based on the BT.2390 formulas.
zoyd is offline  
Sponsored Links
Advertisement
 
post #64 of 1081 Old 11-27-2017, 06:05 AM - Thread Starter
Senior Member
 
Join Date: Jul 2017
Posts: 225
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 197 Post(s)
Liked: 27
OK, I think I'm beginning to see the light, parden the pun!!

Epson EH-TW7300, 104" borderless ReAct 3, Denon AVR-X2300W in 5.1.2, Cambridge Audio Minx Min21's & Min12's, Monitor Audio Pro-65, BK P12-300 sub, Nvidia Shield, Vero 4K, Panasonic UB400, SKY Q, PS3, Wii, RPi, LG 42LS575T, Denon Heos soundbar, Amazon Echo, Logitech Harmony Elite
viperlogic is offline  
post #65 of 1081 Old 11-27-2017, 06:30 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by viperlogic View Post
This is where I'm getting lost, especially in terms of Epson controls, so please bear with me, is the st.2084 curve the white dotted line in the luminance view? When doing a 11 or 21 grayscale sweep it produces a yellow solid line. What Epson controls should be used to "match" the yellow solid line to the white dotted line, contrast or gamma? When you say match do you actually mean match the shape of it not its actual points?
In ST.2084 calibration is absolute, so each code value maps to a specific luminance, e.g., code value 502 (50% stimulus) maps to 92.25 nits. This is the white dotted line in the HCFR luminance view (although currently HCFR uses code value 504 and 94.4 nits, due to the use of 8 bits).

Quote:
If Epson gamma controls are being used then isnt that tone mapping and hence the two steps are the same thing of adjusting to ST.2084 and tone mapping?
No, as I mentioned in my previous post, ST.2084 is the first column in the spreadsheet and goes all the way to 10,000 nits, whereas the tone mapping curves are the ones in the second and third columns, For masters of 1000 nits and 4000 nits respectively.

Quote:
If contrast is being used then at what should we set contrast to because for me with contrast at 50 up to level 642 flashes. If I want higher levels to flash then I have to reduce contrast eg contrast level 29 causes upto level 712 to flash.
Since the Epson 5040UB cannot hit the absolute curve with the gamma controls, I suggested boosting the Contrast control as an interim step to set the basis for calibrating the colours, even though that leads to pre-mature clipping (as long as clipping still occurs beyond code value 502). After that you would re-adjust the contrast to give you the proper clipping point.

Quote:
Using Dominics steps (and contrast at 50 and near same gamma settings on Epson) as our max nits are nearly the same (141 vs 147), I get the following
At that time, I was only focusing on the EOTF and was not adjusting the colours. In hind sight, what I said was a step in the right direction but I would do it differently now, given what zoyd said regarding CMS calibration.

To hit the actual values of ST.2084 (not just the "shape"), I would now do the following:
  1. reduce the projected image size to about 50% in area (or 70% in width and height). This will allow us to bring the luminance to 92.2 nits at code value 502
  2. adjust the gamma controls to match ST.2084
  3. adjust the CMS to match to primary and secondary colours
  4. readjust the Gamma and Contrast controls to match the tone mapping curve (1000 or 4000 nits)
When everything is done, zoom the image back to the size of your choice (ideally fitting the screen). Essentially you will get a dimmer version of the proper calibration, with the correct colours and the "correct shape" of tone mapping (correct in "shape" but scaled down in absolute nits).
The end result will still look like the yellow line above, but hopefully won’t have the “terrible” cololours that you had last time.
viperlogic likes this.

Last edited by Dominic Chan; 11-27-2017 at 08:47 AM.
Dominic Chan is offline  
post #66 of 1081 Old 11-27-2017, 08:48 AM
Advanced Member
 
jwhn's Avatar
 
Join Date: Dec 2013
Posts: 773
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 646 Post(s)
Liked: 160
Quote:
Originally Posted by viperlogic View Post
Will have to check again as can't remember! Changed alot last night!


Okay. Because if you just start measuring colors without doing a 50-50 white pattern your colors will likely be crushed the way you are describing. I accidentally forgot once and had the same issue.


Sent from my iPhone using Tapatalk
jwhn is offline  
post #67 of 1081 Old 11-27-2017, 08:50 AM - Thread Starter
Senior Member
 
Join Date: Jul 2017
Posts: 225
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 197 Post(s)
Liked: 27
Quote:
Originally Posted by Dominic Chan View Post

To hit the actual values of ST.2084 (not just the "shape"), I would not do the following:
  1. reduce the projected image size to about 50% in area (or 70% in width and height). This will allow us to bring the luminance to 92.2 nits at code value 502
  2. adjust the gamma controls to match ST.2084
  3. adjust the CMS to match to primary and secondary colours
  4. readjust the Gamma and Contrast controls to match the tone mapping curve (1000 or 4000 nits)
When everything is done, zoom the image back to the size of your choice (ideally fitting the screen). Essentially you will get a dimmer version of the proper calibration, with the correct colours and the "correct shape" of tone mapping (correct in "shape" but scaled down in absolute nits).
Thanks, will give this a go if I get a chance tonight.

What your personal opinion of the outcome? While technically calibrated to the spec is it just too dark? Though in saying that, that's quite subjective to what everyone deems too dim.

Has anyone tried yet to calibrate using dynamic mode and high lamp, while get great peak brightness (I gain 100 nits for a peak of 248), are the colour so far off they can't be dialled in? I will try when I get a chance though I think I have more hours clocked on this PJ in calibration than actually viewing content!!!

Epson EH-TW7300, 104" borderless ReAct 3, Denon AVR-X2300W in 5.1.2, Cambridge Audio Minx Min21's & Min12's, Monitor Audio Pro-65, BK P12-300 sub, Nvidia Shield, Vero 4K, Panasonic UB400, SKY Q, PS3, Wii, RPi, LG 42LS575T, Denon Heos soundbar, Amazon Echo, Logitech Harmony Elite

Last edited by viperlogic; 11-27-2017 at 09:16 AM.
viperlogic is offline  
post #68 of 1081 Old 11-27-2017, 08:54 AM
Advanced Member
 
jwhn's Avatar
 
Join Date: Dec 2013
Posts: 773
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 646 Post(s)
Liked: 160
Quote:
Originally Posted by Dominic Chan View Post
In ST.2084 calibration is absolute, so each code value maps to a specific luminance, e.g., code value 502 (50% stimulus) maps to 92.25 nits. This is the white dotted line in the HCFR luminance view (although currently HCFR uses code value 504 and 94.4 nits, due to the use of 8 bits).





No, as I mentioned in my previous post, ST.2084 is the first column in the spreadsheet and goes all the way to 10,000 nits, whereas the tone mapping curves are the ones in the second and third columns, For masters of 1000 nits and 4000 nits respectively.





Since the Epson 5040UB cannot hit the absolute curve with the gamma controls, I suggested boosting the Contrast control as an interim step to set the basis for calibrating the colours, even though that leads to pre-mature clipping (as long as clipping still occurs beyond code value 502). After that you would re-adjust the contrast to give you the proper clipping point.





At that time, I was only focusing on the EOTF and was not adjusting the colours. In hind sight, what I said was a step in the right direction but I would do it differently now, given what zoyd said regarding CMS calibration.



To hit the actual values of ST.2084 (not just the "shape"), I would now do the following:
  1. reduce the projected image size to about 50% in area (or 70% in width and height). This will allow us to bring the luminance to 92.2 nits at code value 502
  2. adjust the gamma controls to match ST.2084
  3. adjust the CMS to match to primary and secondary colours
  4. readjust the Gamma and Contrast controls to match the tone mapping curve (1000 or 4000 nits)

When everything is done, zoom the image back to the size of your choice (ideally fitting the screen). Essentially you will get a dimmer version of the proper calibration, with the correct colours and the "correct shape" of tone mapping (correct in "shape" but scaled down in absolute nits).

The end result will still look like the yellow line above, but hopefully won’t have the “terrible” cololours that you had last time.


Makes sense. Wouldn't you want to zoom back out and then do the tone mapping curve? Seems like you would need a different curve if you are zoomed in and getting higher luminance.


Sent from my iPhone using Tapatalk
jwhn is offline  
post #69 of 1081 Old 11-27-2017, 09:19 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by jwhn View Post
Makes sense. Wouldn't you want to zoom back out and then do the tone mapping curve? Seems like you would need a different curve if you are zoomed in and getting higher luminance.
No, when I zoom in (or is it out? I'm referring to a smaller image) I used the 300 nits curve that zoyd calculated. When zooming back to the actual screen size I get ~150 nits.
It's the same as what I was doing before, but instead of scaling in HCFR, I'm now "scaling" the image size instead.

Last edited by Dominic Chan; 11-27-2017 at 09:28 AM.
Dominic Chan is offline  
post #70 of 1081 Old 11-27-2017, 09:26 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by viperlogic View Post
What your personal opinion of the outcome? While technically calibrated to the spec is it just too dark? Though in saying that, that's quite subjective to what everyone deems too dim.
When watch X-Men Apocalyse I didn't find the picture dim. That's not surprising given that I use 50 nits for SDR, compared with 100 nits "standard" for SDR.
However, some movies do look darker. That's a problem the JVC people also mentioned. Given that HDR "expects" 1000 nits peak, there's bound to be compromes when trying to do that with projectors. Unlike OLED TVs (for example), projectors cannot have localized high luminance.

Quote:
Has anyone tried yet to calibrate using dynamic mode and high lamp, while get great peak brightness (I gain 100 nits for a peak of 248), are the colour so far off they can't be dialled in? I will try when I get a chance though I think I have more hours clocked on this PJ in calibration than actually viewing content!!!
Dynamic is greenish. You can cut down the Green Gain for a more neutral white, and still get higher luminance than Bright Cinema. I didn't check the colour gamut, however.
Dominic Chan is offline  
post #71 of 1081 Old 11-27-2017, 09:55 AM
Advanced Member
 
jwhn's Avatar
 
Join Date: Dec 2013
Posts: 773
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 646 Post(s)
Liked: 160
Quote:
Originally Posted by Dominic Chan View Post
No, when I zoom in (or is it out? I'm referring to a smaller image) I used the 300 nits curve that zoyd calculated. When zooming back to the actual screen size I get ~150 nits.
It's the same as what I was doing before, but instead of scaling in HCFR, I'm now "scaling" the image size instead.
Okay. But it would be fine to zoom back to the actual screen and do the final tone mapping, right? Trying to understand if there is an advantage one way or the other.
jwhn is offline  
post #72 of 1081 Old 11-27-2017, 10:04 AM
Advanced Member
 
jwhn's Avatar
 
Join Date: Dec 2013
Posts: 773
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 646 Post(s)
Liked: 160
Quote:
Originally Posted by viperlogic View Post
Thanks, will give this a go if I get a chance tonight.

What your personal opinion of the outcome? While technically calibrated to the spec is it just too dark? Though in saying that, that's quite subjective to what everyone deems too dim.

Has anyone tried yet to calibrate using dynamic mode and high lamp, while get great peak brightness (I gain 100 nits for a peak of 248), are the colour so far off they can't be dialled in? I will try when I get a chance though I think I have more hours clocked on this PJ in calibration than actually viewing content!!!
I think I've mentioned a few times that compared with all the other settings I've tried, including ones on the 5040 thread, using this approach is producing much better results. It looks essentially the same as SDR content in terms of average brightness. No issue of it being too dark.

I am using dynamic mode + high lamp as that is the only way I can get the luminance high enough. I've been able to minimize the grey scale errors and am getting good CMS results (although I do need to try the new approach suggested by Zoyd). When viewing content, colors look pretty good.
jwhn is offline  
post #73 of 1081 Old 11-27-2017, 10:52 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by jwhn View Post
Okay. But it would be fine to zoom back to the actual screen and do the final tone mapping, right? Trying to understand if there is an advantage one way or the other.
The tone mapping curves zoyd calculated for me are for 300 nits peak, which I can only get on the reduced size. On the actual screen it becomes about 140 nits peak.
Dominic Chan is offline  
post #74 of 1081 Old 11-27-2017, 11:16 AM
Advanced Member
 
jwhn's Avatar
 
Join Date: Dec 2013
Posts: 773
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 646 Post(s)
Liked: 160
Quote:
Originally Posted by Dominic Chan View Post
The tone mapping curves zoyd calculated for me are for 300 nits peak, which I can only get on the reduced size. On the actual screen it becomes about 140 nits peak.
I see what you are saying. Better to do the calibration on the original tone map vs. the scaled version.
jwhn is offline  
post #75 of 1081 Old 11-27-2017, 04:12 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by zoyd View Post
ok, so the two point would be used to do gray scale tracking after matching the 2084 curve as best you can, hue/sat/brightness to align CMS around 92 nits, and then contrast/9-point to tailor curve to one of the tone mapping suggestions.
Hi zoyd,

In post 65 above I outlined the steps following your process as closely as the projector allows. However, I wonder what the consequence would be, if the CMS is adjusted without first boosting the luminance to 92 nits at 50% stimulus. Wouldn't the colours still look "right", assuming the colour lumnance targets are relative to the white at the same stimulus level?

That brings up another question when calibrating SDR. In the Primary & Secondary Colours screen, the white level is also measured to form the basis for the relative luminance of the colours. However, this does not seem to be done in the saturation sweep. Wouldn't that lead to luminance errors in the colours, if the gamma is off?
Dominic Chan is offline  
post #76 of 1081 Old 11-27-2017, 05:19 PM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
Quote:
Originally Posted by Dominic Chan View Post
Hi zoyd,

In post 65 above I outlined the steps following your process as closely as the projector allows. However, I wonder what the consequence would be, if the CMS is adjusted without first boosting the luminance to 92 nits at 50% stimulus. Wouldn't the colours still look "right", assuming the colour lumnance targets are relative to the white at the same stimulus level?
Technically in HDR if the color has the wrong absolute luminance it is not correct, with a dE value based on how far off that luminance is. What you are talking about is analogous to a "dE w/o gamma" calculation where only dC and dH are considered. I suppose I could add a switch to turn off dL in the dE calculation but it's just a feel good situation. With dL included you can still minimize dE at 50% white luminance below 92 nits, you'll just be minimizing a larger number.

Quote:
That brings up another question when calibrating SDR. In the Primary & Secondary Colours screen, the white level is also measured to form the basis for the relative luminance of the colours. However, this does not seem to be done in the saturation sweep. Wouldn't that lead to luminance errors in the colours, if the gamma is off?
The saturation sweeps and the color checker targets are calculated from the white measured during the primaries sweep (or failing that, your gray scale white) plus your target EOTF. So yes, if your actual EOTF does not match the target EOTF you will have color errors. You can sort of get rid of that component of the error if you check "use measured gamma", although that is power law approximation of your measurements and won't remove all the "dL".
zoyd is offline  
post #77 of 1081 Old 11-27-2017, 05:45 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by zoyd View Post
Technically in HDR if the color has the wrong absolute luminance it is not correct, with a dE value based on how far off that luminance is. What you are talking about is analogous to a "dE w/o gamma" calculation where only dC and dH are considered. I suppose I could add a switch to turn off dL in the dE calculation but it's just a feel good situation. With dL included you can still minimize dE at 50% white luminance below 92 nits, you'll just be minimizing a larger number.
Perhaps this concept of “Absolute Luminance” hasn’t quite sunk in for me.
With SDR, if I calibrate a projector perfectly on a small screen and then zoom out the image onto a larger screen, all the calibrations remain correct even though the overall luminance drops, i.e., the image gets dimmer but is still accurate (dE=0 for all measurements, whether we use dE w/ gamma or w/o gamma).
From what you said, this is no longer the case with HDR.

Last edited by Dominic Chan; 11-28-2017 at 05:41 AM.
Dominic Chan is offline  
post #78 of 1081 Old 11-27-2017, 05:50 PM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
Quote:
Originally Posted by Dominic Chan View Post
From what you said, this is no longer the case with HDR.
Correct, if 50% white doesn't measure 92 nits it is technically wrong.
zoyd is offline  
post #79 of 1081 Old 11-27-2017, 05:52 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by zoyd View Post
Correct, if 50% white doesn't measure 92 nits it is technically wrong.
Yes, I understood that part in terms of EOTF, but didn’t realize that the colors are also considered wrong.
Dominic Chan is offline  
post #80 of 1081 Old 11-27-2017, 05:56 PM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
zoyd is offline  
post #81 of 1081 Old 11-27-2017, 06:15 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by zoyd View Post
The saturation sweeps and the color checker targets are calculated from the white measured during the primaries sweep (or failing that, your gray scale white) plus your target EOTF.
I just tested that and it seems to be the other round. Saturation sweep refuses to run when the grey scale white is absent, even if the primaries sweep is done.
BTW, there's a minor quirk that I hope you can fix in future version. When running a series of sweeps (e.g., saturation sweeps of all primaries), the last column of each sweep is not updated prior to the start of the next sweep. I can add it to your editable spreadsheet.
Dominic Chan is offline  
post #82 of 1081 Old 11-27-2017, 06:31 PM
Advanced Member
 
jwhn's Avatar
 
Join Date: Dec 2013
Posts: 773
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 646 Post(s)
Liked: 160
Quote:
Originally Posted by zoyd View Post
Correct, if 50% white doesn't measure 92 nits it is technically wrong.

Thanks for your patience educating people in this thread.

Question - I get your point here, but then doesn't it follow that if the display can't hit the absolute targets at higher luminance, say up to 1,000 or 4,000 nits, it is technically wrong also? Just trying to understand if there is something different about the 50% point. Does this affect colors more so than the higher points on the greyscale?



Sent from my iPhone using Tapatalk
jwhn is offline  
post #83 of 1081 Old 11-27-2017, 06:42 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 5,430
Mentioned: 88 Post(s)
Tagged: 0 Thread(s)
Quoted: 3727 Post(s)
Liked: 1157
Quote:
Originally Posted by jwhn View Post
Question - I get your point here, but then doesn't it follow that if the display can't hit the absolute targets at higher luminance, say up to 1,000 or 4,000 nits, it is technically wrong also? Just trying to understand if there is something different about the 50% point. Does this affect colors more so than the higher points on the greyscale
That’s the role of tone mapping - making the wrong right through the highlight roll off
Dominic Chan is offline  
post #84 of 1081 Old 11-27-2017, 06:46 PM
Advanced Member
 
jwhn's Avatar
 
Join Date: Dec 2013
Posts: 773
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 646 Post(s)
Liked: 160
Quote:
Originally Posted by Dominic Chan View Post
That’s the role of tone mapping - making the wrong right through the highlight roll off


Understood - just trying to understand if there is something magic about the 50% point. ; )


Sent from my iPhone using Tapatalk
jwhn is offline  
post #85 of 1081 Old 11-27-2017, 06:50 PM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
Quote:
Originally Posted by jwhn View Post
Thanks for your patience educating people in this thread.

Question - I get your point here, but then doesn't it follow that if the display can't hit the absolute targets at higher luminance, say up to 1,000 or 4,000 nits, it is technically wrong also? Just trying to understand if there is something different about the 50% point. Does this affect colors more so than the higher points on the greyscale?



Sent from my iPhone using Tapatalk
The luminance range up to 100 nits (sometimes called diffuse white) will still be the vast majority of levels that get exercised with HDR content, so if you can nail that range down, most of the content will have correct colors.
zoyd is offline  
post #86 of 1081 Old 11-27-2017, 07:13 PM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
Quote:
Originally Posted by Dominic Chan View Post
I just tested that and it seems to be the other round. Saturation sweep refuses to run when the grey scale white is absent, even if the primaries sweep is done.
No, it still uses the primaries sweep white if it's available. The mandatory grayscale run was added after BT.1886 because you need a black measurement to calculate targets and not everyone runs a black during the primaries sweep.

Quote:
BTW, there's a minor quirk that I hope you can fix in future version. When running a series of sweeps (e.g., saturation sweeps of all primaries), the last column of each sweep is not updated prior to the start of the next sweep. I can add it to your editable spreadsheet.
ok
zoyd is offline  
post #87 of 1081 Old 11-28-2017, 04:11 AM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
Quote:
Originally Posted by Dominic Chan View Post
Perhaps this concept of “Absolute Luminance” hasn’t quite sunk in for me.
With SDR, if I calibrate a projector perfectly on a small screen and then zoom out the image onto a larger screen, all the calibrations remain correct even though the overall luminance drops, i.e., the image gets dimmer but is still accurate (dE=0 for all measurements, whether we use dE w/ gamma or w/o gamma.
From what you said, this is no longer the case with HDR.
I should add that the steps I suggested are formal steps to first meet the HDR requirement which is typically not a problem for flat panels up to 100 nits where there is no desire to then lower the diffuse white response. If in practice you end up at the same CMS settings by minimizing dE at lower contrast then of course you could skip the step of matching the PQ function at 50% = 92 nits. But I'm not completely sure what the target EOTF should be with diffuse white = 47 nits for example, is it just a scaled version of PQ?

edit: I will add a diffuse white target nits to HCFR that will rescale the targets.

Last edited by zoyd; 11-28-2017 at 04:26 AM.
zoyd is offline  
post #88 of 1081 Old 11-28-2017, 04:38 AM - Thread Starter
Senior Member
 
Join Date: Jul 2017
Posts: 225
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 197 Post(s)
Liked: 27
@zoyd

Any ETA on the next HCFR build as I'm going to give calibration another go once available rather than asking for a curve to be manually created for me. Eg will adjust screen size till 50% pattern produces 92 nits, then see what 100% produces and create a curve from that.

Epson EH-TW7300, 104" borderless ReAct 3, Denon AVR-X2300W in 5.1.2, Cambridge Audio Minx Min21's & Min12's, Monitor Audio Pro-65, BK P12-300 sub, Nvidia Shield, Vero 4K, Panasonic UB400, SKY Q, PS3, Wii, RPi, LG 42LS575T, Denon Heos soundbar, Amazon Echo, Logitech Harmony Elite
viperlogic is offline  
post #89 of 1081 Old 11-28-2017, 04:45 AM - Thread Starter
Senior Member
 
Join Date: Jul 2017
Posts: 225
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 197 Post(s)
Liked: 27
FYI, any one using Cosmos Laundromat or any other HDR clip, do a reality check by just playing it as SDR. I taught it looked great on the "calibrated" HDR until I watched it on SDR and while it doesn't have highlights it was much better looking. I guess its easy to get so tied up with HDR to see the highlights that it can cause the rest of the picture to suffer.

Here is a HDR copy (~6G) of it http://drive.google.com/uc?id=0B7Gr...xport=download

Epson EH-TW7300, 104" borderless ReAct 3, Denon AVR-X2300W in 5.1.2, Cambridge Audio Minx Min21's & Min12's, Monitor Audio Pro-65, BK P12-300 sub, Nvidia Shield, Vero 4K, Panasonic UB400, SKY Q, PS3, Wii, RPi, LG 42LS575T, Denon Heos soundbar, Amazon Echo, Logitech Harmony Elite
viperlogic is offline  
post #90 of 1081 Old 11-28-2017, 06:03 AM
AVS Forum Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 6,804
Mentioned: 144 Post(s)
Tagged: 0 Thread(s)
Quoted: 1783 Post(s)
Liked: 1342
Quote:
Originally Posted by viperlogic View Post
@zoyd

Any ETA on the next HCFR build as I'm going to give calibration another go once available rather than asking for a curve to be manually created for me. Eg will adjust screen size till 50% pattern produces 92 nits, then see what 100% produces and create a curve from that.
I might get something out for testing later today.
viperlogic likes this.
zoyd 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