New Lumagen Radiance Pro Series - Page 150 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 3113Likes
Reply
 
Thread Tools
post #4471 of 4870 Old 04-26-2019, 05:45 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 3,420
Mentioned: 57 Post(s)
Tagged: 0 Thread(s)
Quoted: 2860 Post(s)
Liked: 2049
Quote:
Originally Posted by Soulnight View Post
@Wookii

Thanks for bringing the topic here.
No problem, I thought it was the best move, then @jrp Jim or Patrick (I forget his forum handle), or maybe even @Kris Deering or @Gordon Fraser (who will be much more technically versed on this than me) can come back to you.

Quote:
Originally Posted by Soulnight View Post
But the improvement with dynamic display target nits was even bigger in my opinion. As it impacts the global brightness on the fly. If done well to be invisible, you can 1:1 brightness in some cases with real nits=target nits or preserve picture dynamic range (at the cost of a dimmer picture) instead of compressing too much by raising the target nits as much as necessary. The smart resides in how to choose this dynamic target nits since no paper is describing it.
Can you clarify what you mean by "dynamic display target nits" Flo? I don't understand how you are differentiating this from "dynamic frame nits"? Feel free to use a practical example to explain the difference as that might be easier to follow.
Wookii is offline  
Sponsored Links
Advertisement
 
post #4472 of 4870 Old 04-26-2019, 06:02 AM
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,279
Mentioned: 194 Post(s)
Tagged: 0 Thread(s)
Quoted: 941 Post(s)
Liked: 1384
Quote:
Originally Posted by Wookii View Post
Can you clarify what you mean by "dynamic display target nits" Flo? I don't understand how you are differentiating this from "dynamic frame nits"? Feel free to use a practical example to explain the difference as that might be easier to follow.
Easy. For each frame of the movie, you stop the picture, and adjust the display target nits, or what lumagen calls "Display Max Light" until the picture look the best for you for this single frame. You will end up with very different value for dark content and for bright content.

Please have a look at page 22 of the Rep. ITU-R BT.2390-5 if you want to understand the math behind.

In the document, display target nits is refered to as:
"target display maximum (Lmax) luminances".


What the "basic" dynamic tone mapping is adjusting on the other hand, is called "mastering display white luminances: LW". It happens on the fly with the measured current frame peak.

Of course both the dynamic Lw and Lmax need to be smooth / resetted at scene cut to have an enjoyable experience without visible pumping.

MadVR have both Lw and Lmax dynamic. And from your description, Lumagen has only Lw.
Soulnight is offline  
post #4473 of 4870 Old 04-26-2019, 07:22 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 3,420
Mentioned: 57 Post(s)
Tagged: 0 Thread(s)
Quoted: 2860 Post(s)
Liked: 2049
Quote:
Originally Posted by Soulnight View Post
Easy. For each frame of the movie, you stop the picture, and adjust the display target nits, or what lumagen calls "Display Max Light" until the picture look the best for you for this single frame. You will end up with very different value for dark content and for bright content.

Please have a look at page 22 of the Rep. ITU-R BT.2390-5 if you want to understand the math behind.

In the document, display target nits is refered to as:
"target display maximum (Lmax) luminances".


What the "basic" dynamic tone mapping is adjusting on the other hand, is called "mastering display white luminances: LW". It happens on the fly with the measured current frame peak.

Of course both the dynamic Lw and Lmax need to be smooth / resetted at scene cut to have an enjoyable experience without visible pumping.

MadVR have both Lw and Lmax dynamic. And from your description, Lumagen has only Lw.
I'm struggling to understand how this would work in practical terms Flo (so please bear with me). Are you saying you raise the average luminance of a dark scene so that it occupies a greater proportion of the displays available brightness range, i.e. output nits > input nits?

As I say, its probably best demonstrating with some work examples:

So say you have a projector set-up with a measured peak nits of 100 nits. Say you want to target average brightness (excluding highlights) in movies to look similar to their SDR counterparts, so roughly 2 nits input to 1 nit output (on screen).

You have four frames (or Scenes) with a MaxCLL of 50nits, 150nits, 1000 nits and 2500nits respectively.

In my (basic) understanding of dynamic tone mapping, the first two frames/scenes, would effectively be shown nit for nit (or 2 nits for 1 nit on our projector) with no tone mapping since they both fall within the displays physical luminance range.

The 1000nit frame falls beyond the displays brightness range so the DTM system has to choose a proportion of the input range to maintain as nit-for-nit (2 nits input for 1 nit output on our projector), and a proportion to allocate for tone mapping the highlights (obviously with some sort of smooth transition between the two) - for example, maybe the first 150 nits input to the displays 0-75nit output range, and reserves the remaining 25 nits output range to tone map the highlights (151-1000nits input) into.

The 2500nit frame falls even further beyond the displays brightness range so the DTM system has to choose a smaller proportion of the input to maintain as nit-for-nit (2 nits for 1 nit on our projector) to allow greater display output range to map the highlights into. For example, maybe the first 100 nits input to the displays 0-50nit output range, and reserves the remaining 50 nits output range to tone map the highlights (101-2500 input nits) into.

It's obviously a very over-simplified example, with simplified round numbers just to get the idea across, but that is my (very basic) understanding of how dynamic tone mapping works. That prioritises the nit-for-nit range and maintains the average brightness level of that nit for nit range where most of the content exists, and as it would have been viewed during mastering (as in one movie scene to the next).

Now eventually to the question, which is the purpose of my post, how does changing the Display Max Light (Lmax) dynamically, affect how those scenes are rendered on screen?

Last edited by Wookii; 04-26-2019 at 07:30 AM.
Wookii is offline  
Sponsored Links
Advertisement
 
post #4474 of 4870 Old 04-26-2019, 07:36 AM
AVS Forum Special Member
 
SJHT's Avatar
 
Join Date: May 2002
Location: SF Bay Area
Posts: 3,461
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 1022 Post(s)
Liked: 681
Interesting stuff. I think Jim specifically mentioned he would be not be describing in too much detail how they do it for IP reasons. Call it magic!
SJHT is online now  
post #4475 of 4870 Old 04-26-2019, 07:58 AM
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,279
Mentioned: 194 Post(s)
Tagged: 0 Thread(s)
Quoted: 941 Post(s)
Liked: 1384
Quote:
Originally Posted by Wookii View Post
I'm struggling to understand how this would work in practical terms Flo (so please bear with me). Are you saying you raise the average luminance of a dark scene so that it occupies a greater proportion of the displays available brightness range, i.e. output nits > input nits?

As I say, its probably best demonstrating with some work examples:

So say you have a projector set-up with a measured peak nits of 100 nits. Say you want to target average brightness (excluding highlights) in movies to look similar to their SDR counterparts, so roughly 2 nits input to 1 nit output (on screen).

You have four frames (or Scenes) with a MaxCLL of 50nits, 150nits, 1000 nits and 2500nits respectively.

In my (basic) understanding of dynamic tone mapping, the first two frames/scenes, would effectively be shown nit for nit (or 2 nits for 1 nit on our projector) with no tone mapping since they both fall within the displays physical luminance range.

The 1000nit frame falls beyond the displays brightness range so the DTM system has to choose a proportion of the input range to maintain as nit-for-nit (2 nits input for 1 nit output on our projector), and a proportion to allocate for tone mapping the highlights (obviously with some sort of smooth transition between the two) - for example, maybe the first 150 nits input to the displays 0-75nit output range, and reserves the remaining 25 nits output range to tone map the highlights (151-1000nits input) into.

The 2500nit frame falls even further beyond the displays brightness range so the DTM system has to choose a smaller proportion of the input to maintain as nit-for-nit (2 nits for 1 nit on our projector) to allow greater display output range to map the highlights into. For example, maybe the first 100 nits input to the displays 0-50nit output range, and reserves the remaining 50 nits output range to tone map the highlights (101-2500 input nits) into.

It's obviously a very over-simplified example, with simplified round numbers just to get the idea across, but that is my (very basic) understanding of how dynamic tone mapping works. That prioritises the nit-for-nit range and maintains the average brightness level of that nit for nit range where most of the content exists, and as it would have been viewed during mastering (as in one movie scene to the next).

Now eventually to the question, which is the purpose of my post, how does changing the Display Max Light (Lmax) dynamically, affect how those scenes are rendered on screen?

In your initial post, you mentioned at the beginning that you had 150nits on the screen but where using a fixed display target nits (display max light: Lmax) of 500nits instead.

500/150 gives you a factor 3.3.
It means that instead of having your diffuse white of 100nits shown at 100nits, or even at 50nits if we speak about old sdr projector standard, you have effectively your diffuse white at 33nits.

In your example if you were using 300 target nits instead, you would be getting a nice 2:1 ratio for diffuse white brightness. Would look very nice for all dark scene.
Why did you not do it? Simple. Because you had to compromise. If you play the movie the Meg with some frame with a FALL over 1000nits (!!!), a display target nits of 300 would be too low and everything would get flattened, blown-out.
Why? Because you have to compress everything avove 300nits and put it below. If you compress 300nits to 4000nits within let's say 100(bt2390 knee start) to 300nits, you are compressed extremely heavily. And if like in the Meg 90% of the picture is above 300nits to begin with, you are basically heavily compressing the whole picture!
To compress less and retain some hdr effect, you would have to raise the target nits to something like 1000 or higher in some scene of the meg.

But at the same time, if you have 100nits, and you want a 2:1 mapping up to the diffuse white, you need to choose 200 Target nits. This will give you great results for any frame peak below 200. But 200 Target nits will be very inadequate for bright scenes. Depending on how bright you will want to go up with your target nits to retain hdr and image plasticity.

Therefore, dynamic target nits is a great solution to display dark content as bright as your projector can (1:1 or 2:1 depending on your taste) and bright content with a still a great in picture contrast/hdr with a higher target nits.

With a static, you have to pick a middle and it's never optimal.

Edit:

And the smaller the real nits you get on the screen, the worse the compromise. With only 50nits on the screen for example, it would be a veeeery bad idea to pick a display target nits of 2:1 because most movie would look completely blown out. From experience a 300 static target nits is a good compromise with 50 real nits. But that's a factor 6 on the diffuse white!!
Here dynamic target nits is even more important so that you can have 100 target nits (2:1) for scene with a peak lower than 100nits. And higher target nits 200,300,400,... 1000 and more for brighter scene to retain some hdr effect and picture depth.

Last edited by Soulnight; 04-26-2019 at 08:22 AM.
Soulnight is offline  
post #4476 of 4870 Old 04-26-2019, 08:19 AM
AVS Forum Special Member
 
Wookii's Avatar
 
Join Date: Aug 2008
Posts: 3,420
Mentioned: 57 Post(s)
Tagged: 0 Thread(s)
Quoted: 2860 Post(s)
Liked: 2049
Quote:
Originally Posted by Soulnight View Post
500/150 gives you a factor 3.3.
It means that instead of having your diffuse white of 100nits shown at 100nits, or even at 50nits if we speak about old sdr projector standard, you have effectively your diffuse white at 33nits.
Thanks for the explanation, that’s a bit clearer.

If I wanted to test that would measuring 18% windowed test pattern at 100 nit input be sufficient? In that case are you anticipating that would be 33nits?

Quote:
Originally Posted by Soulnight View Post
In your example if you were using 300 target nits instead, you would be getting a nice 2:1 ratio for diffuse white brightness. Would look very nice for all dark scene.
Why did you not do it? Simple. Because you had to compromise. If you play the movie the Meg with some frame with a FALL over 1000nits (!!!), a display target nits of 300 would be too low and everything would get flattened, blown-out.
I have the Meg, but I’ve not tried it under DTM, do you have a time stamp I can take a look at?
Wookii is offline  
post #4477 of 4870 Old 04-26-2019, 10:31 AM
AVS Forum Addicted Member
 
beastaudio's Avatar
 
Join Date: Dec 2010
Location: Western NC
Posts: 14,403
Mentioned: 392 Post(s)
Tagged: 1 Thread(s)
Quoted: 5504 Post(s)
Liked: 5321
Ok so we all know that the new paladin DCR lens will add 30% more brightness to a scope setup vs. zooming. My question is, with the Lumagen, do you get more brightness preserved in a scope presentation as well? Do we know the percentage?

Also, how well does the lumagen handle mixed aspect ratio movies?

I know these questions have likely been asked many times but I couldn't turn up in search a distinct answer for the first one at least...

European Models do not accept banana plugs
Belly of the Beast: Bass Bunker Theater
beastaudio is offline  
post #4478 of 4870 Old 04-26-2019, 12:38 PM
AVS Forum Addicted Member
 
Craig Peer's Avatar
 
Join Date: Aug 2003
Location: Sacramento, CA
Posts: 15,775
Mentioned: 106 Post(s)
Tagged: 0 Thread(s)
Quoted: 6613 Post(s)
Liked: 7687
Quote:
Originally Posted by beastaudio View Post
Ok so we all know that the new paladin DCR lens will add 30% more brightness to a scope setup vs. zooming. My question is, with the Lumagen, do you get more brightness preserved in a scope presentation as well? Do we know the percentage?

Also, how well does the lumagen handle mixed aspect ratio movies?

I know these questions have likely been asked many times but I couldn't turn up in search a distinct answer for the first one at least...
I only use the Lumagen and scope screen with the DCR lens, which gives me 38% more light ( as actually measured ) when using 4096 x 2160, vs zooming using 3840 x 2160. Used this way, with a DCR lens, mixed aspect ratio movies just become scope movies. Period. Nobody in my theater even notices. If I want to watch them as aspect ratio changing movies ( Dunkirk comes to mind since over 1/2 of that film is 16:9 ), I just watch them on my 16:9 screen.

I just want to add, the scenes I looked at last night with the Lumagen DTM - several in Passengers, and Oblivion - including the dark interrogation scenes with Morgan Freeman in the dark - looked profoundly better to me than without the HDR tone mapping.

I'm still surprised how much my RS4500 has evolved over 2 years, with the JVC firmware updates, the Panamorph DCR lens, the Lumagen Radiance Pro and now dynamic HDR tone mapping. It's like having a completely new and improved projector when it comes to watching 4K HDR. The difference is astounding !
audioguy, cargen, OzHDHT and 1 others like this.

Last edited by Craig Peer; 04-26-2019 at 12:57 PM.
Craig Peer is online now  
post #4479 of 4870 Old 04-26-2019, 01:19 PM
AVS Forum Addicted Member
 
beastaudio's Avatar
 
Join Date: Dec 2010
Location: Western NC
Posts: 14,403
Mentioned: 392 Post(s)
Tagged: 1 Thread(s)
Quoted: 5504 Post(s)
Liked: 5321
Quote:
Originally Posted by Craig Peer View Post
I only use the Lumagen and scope screen with the DCR lens, which gives me 38% more light ( as actually measured ) when using 4096 x 2160, vs zooming using 3840 x 2160. Used this way, with a DCR lens, mixed aspect ratio movies just become scope movies. Period. Nobody in my theater even notices. If I want to watch them as aspect ratio changing movies ( Dunkirk comes to mind since over 1/2 of that film is 16:9 ), I just watch them on my 16:9 screen.

I just want to add, the scenes I looked at last night with the Lumagen DTM - several in Passengers, and Oblivion - including the dark interrogation scenes with Morgan Freeman in the dark - looked profoundly better to me than without the HDR tone mapping.

I'm still surprised how much my RS4500 has evolved over 2 years, with the JVC firmware updates, the Panamorph DCR lens, the Lumagen Radiance Pro and now dynamic HDR tone mapping. It's like having a completely new and improved projector when it comes to watching 4K HDR. The difference is astounding !
Thanks Craig but my question was that if you chose NOT to use the DCR, would the lumagen's aspect manipulation software preserve some brightness just like the DCR regains by not having to zoom the PJ. It doesn't sound like this is the case, since using the lumagen you would still have to use the zoom method for a scope source. I was hoping since you could use the lumagen to stretch a 16:9 image out to scope, that you could then further stretch a scope source vertically and not have to use the zoom method.

European Models do not accept banana plugs
Belly of the Beast: Bass Bunker Theater
beastaudio is offline  
post #4480 of 4870 Old 04-26-2019, 01:39 PM
AVS Forum Addicted Member
 
Craig Peer's Avatar
 
Join Date: Aug 2003
Location: Sacramento, CA
Posts: 15,775
Mentioned: 106 Post(s)
Tagged: 0 Thread(s)
Quoted: 6613 Post(s)
Liked: 7687
Quote:
Originally Posted by beastaudio View Post
Thanks Craig but my question was that if you chose NOT to use the DCR, would the lumagen's aspect manipulation software preserve some brightness just like the DCR regains by not having to zoom the PJ. It doesn't sound like this is the case, since using the lumagen you would still have to use the zoom method for a scope source. I was hoping since you could use the lumagen to stretch a 16:9 image out to scope, that you could then further stretch a scope source vertically and not have to use the zoom method.
You kind of need either a HE lens to expand the picture horizontally back into the correct aspect ratio, or a VC lens to compress it vertically back into the correct aspect ratio.
Craig Peer is online now  
post #4481 of 4870 Old 04-26-2019, 01:43 PM
AVS Forum Addicted Member
 
beastaudio's Avatar
 
Join Date: Dec 2010
Location: Western NC
Posts: 14,403
Mentioned: 392 Post(s)
Tagged: 1 Thread(s)
Quoted: 5504 Post(s)
Liked: 5321
Quote:
Originally Posted by Craig Peer View Post
You kind of need either a HE lens to expand the picture horizontally back into the correct aspect ratio, or a VC lens to compress it vertically back into the correct aspect ratio.
Im sure my trusty ole uh380 could handle that then, but what I was hoping to do was accomplish everything just within a lumagen processor and avoid a lens.

European Models do not accept banana plugs
Belly of the Beast: Bass Bunker Theater
beastaudio is offline  
post #4482 of 4870 Old 04-26-2019, 02:15 PM
AVS Forum Club Gold
 
Mike_WI's Avatar
 
Join Date: Mar 2008
Location: SE Wisconsin
Posts: 3,446
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 434 Post(s)
Liked: 411
Arrow FW Beta 042019

http://www.lumagen.com/testindex.php...ncepro_updates

Beta 042019
Found and fixed issues with HLG tone mapping and also now have it working with DTM.
Fixed an issue causing a flickering effect in some scenes with DTM.
DTM improved to avoid some undesirable tonemap updates.
New 18 Ghz input chip firmware from chip vendor which fixes an Apple TV audio dropout issue.
Please continue giving us your detailed feedback on issues via email at [email protected] .
Update time ~1 minutes @230k from previous firmware---systems with 18Ghz input cards will have ~1 minute bootup time on first power up after this update

HT: Oppo UDP-203 -> Lumagen RadiancePro 4446 {18 GHz input x2 & 18 GHz output x1 cards} - "new (112818 FW)" 18 GHz microcode - parallel outs to --> [Audio: Denon 5308CI] --> [Video: JVC RS520 FW v30.1]
HT Details: link
Mike_WI is offline  
post #4483 of 4870 Old 04-26-2019, 03:26 PM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 2,869
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 2327 Post(s)
Liked: 1477
Quote:
Originally Posted by Mike_WI View Post
http://www.lumagen.com/testindex.php...ncepro_updates

Beta 042019
Found and fixed issues with HLG tone mapping and also now have it working with DTM.
Fixed an issue causing a flickering effect in some scenes with DTM.
DTM improved to avoid some undesirable tonemap updates.
New 18 Ghz input chip firmware from chip vendor which fixes an Apple TV audio dropout issue.
Please continue giving us your detailed feedback on issues via email at [email protected] .
Update time ~1 minutes @230k from previous firmware---systems with 18Ghz input cards will have ~1 minute bootup time on first power up after this update
I've just installed and given this a go, looks great, ATV4k audio fix is working well, and has fixed up a few issues I'd reported in to Lumagen. Hats off to the guys at Lumagen, many thanks Patrick for the continued updates. Keeps getting better.
bobof is offline  
post #4484 of 4870 Old 04-26-2019, 03:42 PM
AVS Forum Addicted Member
 
Craig Peer's Avatar
 
Join Date: Aug 2003
Location: Sacramento, CA
Posts: 15,775
Mentioned: 106 Post(s)
Tagged: 0 Thread(s)
Quoted: 6613 Post(s)
Liked: 7687
Quote:
Originally Posted by bobof View Post
I've just installed and given this a go, looks great, ATV4k audio fix is working well, and has fixed up a few issues I'd reported in to Lumagen. Hats off to the guys at Lumagen, many thanks Patrick for the continued updates. Keeps getting better.
And you were just talking about this issue in another thread. " Voila' " - there's a fix !
Mike_WI likes this.
Craig Peer is online now  
post #4485 of 4870 Old 04-26-2019, 10:01 PM
jrp
AVS Forum Special Member
 
jrp's Avatar
 
Join Date: Jan 2003
Location: Portland, Oregon, USA
Posts: 1,462
Mentioned: 57 Post(s)
Tagged: 0 Thread(s)
Quoted: 374 Post(s)
Liked: 938
I am not a photographer but in response to requests here I took a couple pictures of the Lumagen Demo Theater.

The knickknacks on shelves are LOTR reproductions include "Sting" and a couple pipes, etc. The glasses have etched text in "LOTR Elvish." We have two additional LOTR swords and a shield in the hall not pictured. In the theater we have four faux "hand hewn" look beams each with a hand forged sconce to give a slight castle look (okay very slight).

Certainly a very mild rendition of a LOTR theme but we like it.
Attached Thumbnails
Click image for larger version

Name:	Lumagen Theater Entrance.JPG
Views:	142
Size:	203.3 KB
ID:	2558684   Click image for larger version

Name:	Lumagen Theater.JPG
Views:	142
Size:	155.0 KB
ID:	2558686   Click image for larger version

Name:	Lumagen Theater front to back.JPG
Views:	144
Size:	208.4 KB
ID:	2558688  
turls, giomania, dlinsley and 9 others like this.

Jim Peterson
Lumagen
jrp is offline  
post #4486 of 4870 Old 04-26-2019, 11:23 PM
jrp
AVS Forum Special Member
 
jrp's Avatar
 
Join Date: Jan 2003
Location: Portland, Oregon, USA
Posts: 1,462
Mentioned: 57 Post(s)
Tagged: 0 Thread(s)
Quoted: 374 Post(s)
Liked: 938
Quote:
Originally Posted by Soulnight View Post
@Wookii

Thanks for bringing the topic here.

Here the begining of the discussion:

https://www.avsforum.com/forum/24-di...l#post57954362

Again from your new explanation, it seems that Lumagen only implemented the first stone of dynamic tone mapping.

And that is using the frame peak instead of the movie peak in the bt2390 formulas.
There is obviously a misunderstanding of how the Radiance Pro Dynamic Tone Mapping works. It definitely dynamically adjusts both the scene max light (a.k.a. Lw), and the display max target (a.k.a. LMax) on a scene by scene basis. Doing this is not optional for dynamic tone mapping.

Perhaps what was missed is that the HDR parameter "Ratio" adjusts the display target max on an input and input memory basis and separately for darker movies and brighter movies (low set and high set). The Radiance Pro allows the user to adjust these to match the TV/projector and their personal taste. We have refined our defaults to the point where we feel that almost all users will be very happy with the defaults. So that the only adjustment needed by most users is the "Display Max Light" in the CMS. Dynamic Tone Mapping makes it even less necessary to adjust the HDR Parameters (IMO).

Someone complained that there was some trial and error setting Display Max Light, but I think this is an advantage and not a detriment. In any case it is what we choose to do and I am very happy with how it works. It is very straight forward to set Display Max Light once you get the hang of it.

Current Dynamic Tone Mapping uses the low set Ratio (LMax if you prefer) for darker scenes, and the High set Ratio (LMax again) for bright movies. The Low and High set Shape and Transition also play a key role. It then dynamically interpolates, the low/high Display Max Light (LMax), Shape, Transition, and of course the dynamically calculated scene max light (Lw), and so takes user settings/preferences into account appropriately for each scene's brightness. Stated more explicitly display target max is not the same for dark scenes and bright scenes.

====

I have completed C code for a new low/high blend algorithm and have handed it off for Patrick to work his magic in integrating it in the software and in the FPGA code. I likely should not mention this since while it looks like a nice improvement in plots we have yet to see it on the screen. It may not prove visibly better (but I think it will).

*If* we decide to switch to the new algorithm it has three control points for Display Max Light (LMax again). These are at the display max, 1000 nits, and 5000 nits, and then the >5000 range is handled somewhat like the display max to 1000 nit range only going from 5000 nits to 10000 nits.

In the current implementation the Low Set Shape and Transition have no effect at the display max (but some effect as the scene max rises). The current implementation interpolates the highly non-linear changes in the transfer function in one section from the display max up to the high end of the curve. While this is working very well, I think the new algorithm will be better, and I have it completed and ready to implement. It breaks the interpolation into Display Max up to 1000 nits, then 1000 nits to 5000 nits (and then there is additional processing for over 5000 nits). This significantly improves the accuracy of the blend for dark scenes (whether this will be visible is unknown). It also uses a different algorithm from Display Max to 1000, than it does from 1000 to 5000 nits, which gives us more accurate interpolation.

I probably should not have mentioned this until we know if we will use it or not, but it seems appropriate for this post. So, you may, or may not, see this new blend algorithm in a release in a couple weeks.

====

I believe Patrick has done an amazing job on the his algorithms and implementation of scene detection. I have been told by someone who's opinion I respect that it is better than professional editing suite's scene detection. Patrick has been able to completely resolve (or in one or two cases significantly improve) each scene sent to us to look at. This is a real tightrope walk since the scene being worked on has to be improved without breaking any other scenes. In my testing (looking at a debug version with the scene detection information visible) i continue to be impressed how the Pro scene detection catches cases I didn't think it could. It also does an excellent job of not having false detection of a scene change that causes a visible change in the middle of a scene. Patrick is still working on a couple scenes with bright flashes that "look" like a new scene, but has already significantly improved the ones we know about.

As you find new scenes for us to work on please send them to us at the Lumagen support email. It is impossible to have perfect scene detection, and so we know as good as it is already we can improve it with your help finding the scenes that need attention.

Jim Peterson
Lumagen
jrp is offline  
post #4487 of 4870 Old 04-27-2019, 02:11 AM
Senior Member
 
Join Date: Feb 2016
Posts: 363
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 174 Post(s)
Liked: 67
Quote:
Originally Posted by jrp View Post
I am not a photographer but in response to requests here I took a couple pictures of the Lumagen Demo Theater.



The knickknacks on shelves are LOTR reproductions include "Sting" and a couple pipes, etc. The glasses have etched text in "LOTR Elvish." We have two additional LOTR swords and a shield in the hall not pictured. In the theater we have four faux "hand hewn" look beams each with a hand forged sconce to give a slight castle look (okay very slight).



Certainly a very mild rendition of a LOTR theme but we like it.


Love it Will done


Sent from my iPhone using Tapatalk
A7mad78 is online now  
post #4488 of 4870 Old 04-27-2019, 02:23 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 2,869
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 2327 Post(s)
Liked: 1477
Quote:
Originally Posted by jrp View Post
We have refined our defaults to the point where we feel that almost all users will be very happy with the defaults. So that the only adjustment needed by most users is the "Display Max Light" in the CMS. Dynamic Tone Mapping makes it even less necessary to adjust the HDR Parameters (IMO).

Someone complained that there was some trial and error setting Display Max Light, but I think this is an advantage and not a detriment. In any case it is what we choose to do and I am very happy with how it works. It is very straight forward to set Display Max Light once you get the hang of it.
Guilty as charged! . Indeed now I have good P3 LUTS loaded in the Pro (using the hack discovered to make the X7900 units have a linear mode with the WCG filter enabled) I don't change any of the settings (other than the level of DTM for some sources) and am very happy with the results seen so far. I too have been very pleasantly surprised with what has been achieved on what is clearly a hard problem.

A bit of experience has shown me that for my peak white, display size / viewing angle and room lighting the "correct" Display Max Light setting is about 300-400 for 75 nits peak and about 500-600 for 130 nits peak. I do wonder if a curve could be fitted to that control to make it represent something more like the real measured max nits, though I guess that would deter experimentation that might be beneficial if there were all of a sudden only one "right" answer for it. And probably it would need to be at least two controls really, one that defined the measured max nits and one that defined the field of view that those nits were occupying (maybe even a third control to define the room ambient...). When I think in those terms, actually just fiddling the Display Max Light (which takes moments) is perhaps the easier approach. It is easy for us in our projection bubble to think that everyone is doing projection, but of course this isn't true.

I think my comments with respect to the rest of the controls were really more related to "paradox of choice" matters, which being a self confessed "maximizer" affects me more than most (interesting reading and insight into the way many of us on these forums are; book exec summary here: https://keithdwalker.ca/wp-content/s...hwartz.EBS.pdf).
bobof is offline  
post #4489 of 4870 Old 04-27-2019, 02:34 AM
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,279
Mentioned: 194 Post(s)
Tagged: 0 Thread(s)
Quoted: 941 Post(s)
Liked: 1384
Quote:
Originally Posted by jrp View Post
There is obviously a misunderstanding of how the Radiance Pro Dynamic Tone Mapping works. It definitely dynamically adjusts both the scene max light (a.k.a. Lw), and the display max target (a.k.a. LMax) on a scene by scene basis. Doing this is not optional for dynamic tone mapping.

Perhaps what was missed is that the HDR parameter "Ratio" adjusts the display target max on an input and input memory basis and separately for darker movies and brighter movies (low set and high set). The Radiance Pro allows the user to adjust these to match the TV/projector and their personal taste. We have refined our defaults to the point where we feel that almost all users will be very happy with the defaults. So that the only adjustment needed by most users is the "Display Max Light" in the CMS. Dynamic Tone Mapping makes it even less necessary to adjust the HDR Parameters (IMO).

Someone complained that there was some trial and error setting Display Max Light, but I think this is an advantage and not a detriment. In any case it is what we choose to do and I am very happy with how it works. It is very straight forward to set Display Max Light once you get the hang of it.

Current Dynamic Tone Mapping uses the low set Ratio (LMax if you prefer) for darker scenes, and the High set Ratio (LMax again) for bright movies. The Low and High set Shape and Transition also play a key role. It then dynamically interpolates, the low/high Display Max Light (LMax), Shape, Transition, and of course the dynamically calculated scene max light (Lw), and so takes user settings/preferences into account appropriately for each scene's brightness. Stated more explicitly display target max is not the same for dark scenes and bright scenes.
Thank you for the explanation. You are correct that I did not know of the "Ratio" settings since it seems that most Lumagen user keep default settings currently and therefore do not mention it.
This raises some other questions.

So, to take back the case of @Wookii :
He has a projector with an on screen brightness of 150nits. He has set-up the "display max light" to 500nits.
From what he wrote earlier in this thread, he is using the "default" settings for everything else including Ratio setting for both low and high sets.

A few questions:
1) What is the default ratio setting?
2) Can this ratio both adjust from the display max light of 500nits to a lower value for a dark scene AND a higher value of Lmax for a bright scene? What would be the range of acceptable value for Lmax for Wookii calculated by the DTM?
To make the question a bit more precise: can Wookii get an dynamic "Display max Target" lower than 500nits? Can it even get lower than 300nits (2:1 ratio for diffuse white at 50nits?) or lower than his real nits (150) on screen (1:1 ratio with diffuse white at 100nits?)

3) How does Lumagen DTM ensures that the picture will never be displayed brighter than was encoded in nits on the UHD Bluray? Can the user input somewhere his Real nits?

4) How would DTM differentiate between:
a) a TV User with a real on screen brightness of 500nits and who sets the "display max light" to 500nits and is using default settings for everything else
b) a projector user like @Wookii with a real on screen brightness of 150nits but who set 500nits to "the display max light" as well.

In case a), you don't want ever to use a Lmax (Display Target nits) lower than 500nits so that the content is never shown brighter than coded on the UHD bluray
In case b), @Wookii would like to have 2:1 ratio whenever possible which means that the target nits (Lmax) should be able to drop to 300nits. Some other user might even want to have 1:1 whenever possible which would mean than Lmax should be able to go down up to 150nits Lmax (Target Nits).

Thank you.
I think the clarification you provide will help the Lumagen users to better understand how to use DTM in itself.
It will also explain differences in rendering between madVR DTM and Lumagen DTM when compared.

Florian
Roland Janus, Javs and bobof like this.

Last edited by Soulnight; 04-27-2019 at 02:39 AM.
Soulnight is offline  
post #4490 of 4870 Old 04-27-2019, 03:06 AM
AVS Forum Special Member
 
audioguy's Avatar
 
Join Date: Jun 2001
Location: Not far from Atlanta - but far enough!
Posts: 9,063
Mentioned: 89 Post(s)
Tagged: 0 Thread(s)
Quoted: 4567 Post(s)
Liked: 3335
I have yet to receive my Lumagen and have it calibrated, but I sure hope there is a "set it and forget it" default mode, even if it does not always produce the most optimized results in all cases. I really prefer to not have to diddle with any Lumagen settings when we go to the theater to watch a movie. I wouldn't mind if there was a single setting to adjust if I were in my "tweak mode" but I'm trying very hard to make our theater as wife friendly as possible - and requiring her to diddle with Lumagen settings would be moving in the opposite direction.

From reading much of the posting above, it is clearl that the majority of users who post on this thread are about 1 million times more knowledgeable than I am in all matters of video.

Last edited by audioguy; 04-27-2019 at 03:13 AM.
audioguy is online now  
post #4491 of 4870 Old 04-27-2019, 04:52 AM
AVS Forum Club Gold
 
Mike_WI's Avatar
 
Join Date: Mar 2008
Location: SE Wisconsin
Posts: 3,446
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 434 Post(s)
Liked: 411
Quote:
Originally Posted by audioguy View Post
I have yet to receive my Lumagen and have it calibrated, but I sure hope there is a "set it and forget it" default mode, even if it does not always produce the most optimized results in all cases. I really prefer to not have to diddle with any Lumagen settings when we go to the theater to watch a movie. I wouldn't mind if there was a single setting to adjust if I were in my "tweak mode" but I'm trying very hard to make our theater as wife friendly as possible - and requiring her to diddle with Lumagen settings would be moving in the opposite direction.

From reading much of the posting above, it is clearl that the majority of users who post on this thread are about 1 million times more knowledgeable than I am in all matters of video.
I think most people set and forget. Or hire someone to set it.
The DTM should make the dynamic adjustment happen in the background.
There will always be room for the tweakers that like to see what adjustments do.
mikela, A7mad78, dinamigym and 1 others like this.

HT: Oppo UDP-203 -> Lumagen RadiancePro 4446 {18 GHz input x2 & 18 GHz output x1 cards} - "new (112818 FW)" 18 GHz microcode - parallel outs to --> [Audio: Denon 5308CI] --> [Video: JVC RS520 FW v30.1]
HT Details: link
Mike_WI is offline  
post #4492 of 4870 Old 04-27-2019, 05:52 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 2,869
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 2327 Post(s)
Liked: 1477
Quote:
Originally Posted by Soulnight View Post
A few questions:
1) What is the default ratio setting?
2) Can this ratio both adjust from the display max light of 500nits to a lower value for a dark scene AND a higher value of Lmax for a bright scene? What would be the range of acceptable value for Lmax for Wookii calculated by the DTM?
To make the question a bit more precise: can Wookii get an dynamic "Display max Target" lower than 500nits? Can it even get lower than 300nits (2:1 ratio for diffuse white at 50nits?) or lower than his real nits (150) on screen (1:1 ratio with diffuse white at 100nits?)
Hi Florian,

I believe the following will help your understanding for the above points; these are two shots I just took of DML400 with DTM enabled and disabled on the Masciola ramps. Deliberately underexposed to capture the highlights, so completely ignore the shadow detail (it is resolved almost all the way down with this 10,000 nit image, which I guess is as you would expect). Clearly there is detail resolved above the DML400 "limit" with DTM enabled. A couple of other shots also of DML set to 9900 and 50 to show the extremes. You can see with DML 50 it still appears to resolve up to well over 1000 nits, but that if you up it to 9900 you are resolving the few patches that DTM did clip at the lower settings (it actually starts to resolve all those boxes around 6000 if I recall).

Everything else in this image is factory I recall believe.

In terms of your other questions, I understand the point you are making; from what I can see if you had a 500 nit TV in a living room environment you'd actually want to set DML higher than 500 nits, and / or modify the Ratio I believe (don't have one so haven't tried to know exactly what). I think there are some levels of indirection here with the way the settings interact that make it a little harder to achieve a specific numeric goal, but at the same time you have to bear in mind that this has been built up in such a way that the significant number of folk who have had units professionally calibrated previously and set up with the old Intensity Mapping settings have been able to just upgrade their SW and it has added DTM seemingly without any additional configuration needed. Of course, at the moment most MadVR users of DTM are enthusiasts who are right "in the guts" of their tone mapping solution.

FWIW I am outputting P3 SDR with a 3DLUT to convert the native PJ gamut (x7900 profile off / filter on) to P3 2.4 gamma per Radiance requirements, 75 real nits on screen. Photos are just a Sony mobile phone so apologies for the quality, but I think they probably give you the idea.
Attached Thumbnails
Click image for larger version

Name:	DML400_DynOn.JPG
Views:	86
Size:	189.9 KB
ID:	2558880   Click image for larger version

Name:	DML400_DynOff.JPG
Views:	83
Size:	222.2 KB
ID:	2558882   Click image for larger version

Name:	DML50_DynOn.JPG
Views:	76
Size:	219.1 KB
ID:	2558892   Click image for larger version

Name:	DML9900_DynOn.JPG
Views:	79
Size:	124.8 KB
ID:	2558894  
bobof is offline  
post #4493 of 4870 Old 04-27-2019, 06:48 AM
Member
 
loggeo's Avatar
 
Join Date: Aug 2011
Location: Rhodes, Greece
Posts: 175
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 138 Post(s)
Liked: 71
Quote:
Originally Posted by Soulnight View Post
Thank you for the explanation. You are correct that I did not know of the "Ratio" settings since it seems that most Lumagen user keep default settings currently and therefore do not mention it.
This raises some other questions.

So, to take back the case of @Wookii :
He has a projector with an on screen brightness of 150nits. He has set-up the "display max light" to 500nits.
From what he wrote earlier in this thread, he is using the "default" settings for everything else including Ratio setting for both low and high sets.

A few questions:
1) What is the default ratio setting?
2) Can this ratio both adjust from the display max light of 500nits to a lower value for a dark scene AND a higher value of Lmax for a bright scene? What would be the range of acceptable value for Lmax for Wookii calculated by the DTM?
To make the question a bit more precise: can Wookii get an dynamic "Display max Target" lower than 500nits? Can it even get lower than 300nits (2:1 ratio for diffuse white at 50nits?) or lower than his real nits (150) on screen (1:1 ratio with diffuse white at 100nits?)

3) How does Lumagen DTM ensures that the picture will never be displayed brighter than was encoded in nits on the UHD Bluray? Can the user input somewhere his Real nits?

4) How would DTM differentiate between:
a) a TV User with a real on screen brightness of 500nits and who sets the "display max light" to 500nits and is using default settings for everything else
b) a projector user like @Wookii with a real on screen brightness of 150nits but who set 500nits to "the display max light" as well.

In case a), you don't want ever to use a Lmax (Display Target nits) lower than 500nits so that the content is never shown brighter than coded on the UHD bluray
In case b), @Wookii would like to have 2:1 ratio whenever possible which means that the target nits (Lmax) should be able to drop to 300nits. Some other user might even want to have 1:1 whenever possible which would mean than Lmax should be able to go down up to 150nits Lmax (Target Nits).

Thank you.
I think the clarification you provide will help the Lumagen users to better understand how to use DTM in itself.
It will also explain differences in rendering between madVR DTM and Lumagen DTM when compared.

Florian
This is more to validate the method we have been using so far and hopefully it helps with your questions. I never used just the default settings, so it will be interesting to recheck now. If only jvc would deliver my projector..
Lumagen has an adjustable MaxCLL crossover for static tone mapping. It defaults at 2500 nits. Therefore you can have different settings (presets) for above and below crossover settings.
Its preset consists of some sub settings to optimize static tone mapping (ratio, shape, transition, hdsaturation, gamma, black).
While you set the display max light once (defaults at 500 nits but it is also adjustable), you can use the ratio sub setting to adjust max display light per preset.
Ideally you optimize the 2 presets and then enable the dynamic tone mapping.
At least this is what I have been doing with a temporary projector I have here.
giomania likes this.
loggeo is offline  
post #4494 of 4870 Old 04-27-2019, 10:46 AM
Member
 
Join Date: Jan 2019
Posts: 43
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Liked: 12
Quote:
Originally Posted by audioguy View Post
I have yet to receive my Lumagen and have it calibrated, but I sure hope there is a "set it and forget it" default mode, even if it does not always produce the most optimized results in all cases. I really prefer to not have to diddle with any Lumagen settings when we go to the theater to watch a movie. I wouldn't mind if there was a single setting to adjust if I were in my "tweak mode" but I'm trying very hard to make our theater as wife friendly as possible - and requiring her to diddle with Lumagen settings would be moving in the opposite direction.

From reading much of the posting above, it is clearl that the majority of users who post on this thread are about 1 million times more knowledgeable than I am in all matters of video.
Once you have it installed and your theatre calibrated you won't need to adjust anything. Everyone's theatre is unique and the smallest variable can create large problems - hdmi cables for example - and what you read may not be applicable to your theatre. I personally was so happy with my video that I did not install a single update for nearly 2 years - until the creation of DTM. A couple emails to Patrick and Jim prior to installing the update to confirm the process were answered nearly instantaneously. The customer service offered by Jim and Patrick is second to none. I installed the update in 15 minutes and the new DTM has breathed new life into my projector.

So, yes, if you want it to be set it and forget it then it certainly can be.
nbynw is online now  
post #4495 of 4870 Old 04-27-2019, 05:04 PM
AVS Forum Special Member
 
audioguy's Avatar
 
Join Date: Jun 2001
Location: Not far from Atlanta - but far enough!
Posts: 9,063
Mentioned: 89 Post(s)
Tagged: 0 Thread(s)
Quoted: 4567 Post(s)
Liked: 3335
Quote:
Originally Posted by nbynw View Post
So, yes, if you want it to be set it and forget it then it certainly can be.
That is good to know. Thanks. As it turns out, a calibration client I am working with has a Lumagen and he has never touched it!!
audioguy is online now  
post #4496 of 4870 Old 04-27-2019, 05:39 PM
Member
 
Join Date: Jan 2019
Location: Dayton, Ohio
Posts: 88
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 86 Post(s)
Liked: 62
Quote:
Originally Posted by jrp View Post
There is obviously a misunderstanding of how the Radiance Pro Dynamic Tone Mapping works. It definitely dynamically adjusts both the scene max light (a.k.a. Lw), and the display max target (a.k.a. LMax) on a scene by scene basis. Doing this is not optional for dynamic tone mapping.

Perhaps what was missed is that the HDR parameter "Ratio" adjusts the display target max on an input and input memory basis and separately for darker movies and brighter movies (low set and high set). The Radiance Pro allows the user to adjust these to match the TV/projector and their personal taste. We have refined our defaults to the point where we feel that almost all users will be very happy with the defaults. So that the only adjustment needed by most users is the "Display Max Light" in the CMS. Dynamic Tone Mapping makes it even less necessary to adjust the HDR Parameters (IMO).

Someone complained that there was some trial and error setting Display Max Light, but I think this is an advantage and not a detriment. In any case it is what we choose to do and I am very happy with how it works. It is very straight forward to set Display Max Light once you get the hang of it.

Current Dynamic Tone Mapping uses the low set Ratio (LMax if you prefer) for darker scenes, and the High set Ratio (LMax again) for bright movies. The Low and High set Shape and Transition also play a key role. It then dynamically interpolates, the low/high Display Max Light (LMax), Shape, Transition, and of course the dynamically calculated scene max light (Lw), and so takes user settings/preferences into account appropriately for each scene's brightness. Stated more explicitly display target max is not the same for dark scenes and bright scenes.

====

I have completed C code for a new low/high blend algorithm and have handed it off for Patrick to work his magic in integrating it in the software and in the FPGA code. I likely should not mention this since while it looks like a nice improvement in plots we have yet to see it on the screen. It may not prove visibly better (but I think it will).

*If* we decide to switch to the new algorithm it has three control points for Display Max Light (LMax again). These are at the display max, 1000 nits, and 5000 nits, and then the >5000 range is handled somewhat like the display max to 1000 nit range only going from 5000 nits to 10000 nits.

In the current implementation the Low Set Shape and Transition have no effect at the display max (but some effect as the scene max rises). The current implementation interpolates the highly non-linear changes in the transfer function in one section from the display max up to the high end of the curve. While this is working very well, I think the new algorithm will be better, and I have it completed and ready to implement. It breaks the interpolation into Display Max up to 1000 nits, then 1000 nits to 5000 nits (and then there is additional processing for over 5000 nits). This significantly improves the accuracy of the blend for dark scenes (whether this will be visible is unknown). It also uses a different algorithm from Display Max to 1000, than it does from 1000 to 5000 nits, which gives us more accurate interpolation.

I probably should not have mentioned this until we know if we will use it or not, but it seems appropriate for this post. So, you may, or may not, see this new blend algorithm in a release in a couple weeks.

====

I believe Patrick has done an amazing job on the his algorithms and implementation of scene detection. I have been told by someone who's opinion I respect that it is better than professional editing suite's scene detection. Patrick has been able to completely resolve (or in one or two cases significantly improve) each scene sent to us to look at. This is a real tightrope walk since the scene being worked on has to be improved without breaking any other scenes. In my testing (looking at a debug version with the scene detection information visible) i continue to be impressed how the Pro scene detection catches cases I didn't think it could. It also does an excellent job of not having false detection of a scene change that causes a visible change in the middle of a scene. Patrick is still working on a couple scenes with bright flashes that "look" like a new scene, but has already significantly improved the ones we know about.

As you find new scenes for us to work on please send them to us at the Lumagen support email. It is impossible to have perfect scene detection, and so we know as good as it is already we can improve it with your help finding the scenes that need attention.
JRP

You posted - So that the only adjustment needed by most users is the "Display Max Light" in the CMS.

? Is the CMS one of the settings in the lumagen - or is this something in the projector, i am pretty new to all this but trying to learn and keep up.

I do have the RS4500 DCR and Lumagen 4240 that i just purchased. I did the recent update this afternoon on the lumagen.

Many thanks
SSnarski is offline  
post #4497 of 4870 Old 04-27-2019, 06:13 PM
Rise & Grind
 
dinamigym's Avatar
 
Join Date: Mar 2017
Location: SW Burb of Chicago
Posts: 472
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 295 Post(s)
Liked: 277
Quote:
Originally Posted by SSnarski View Post
JRP



You posted - So that the only adjustment needed by most users is the "Display Max Light" in the CMS.



? Is the CMS one of the settings in the lumagen - or is this something in the projector, i am pretty new to all this but trying to learn and keep up.



I do have the RS4500 DCR and Lumagen 4240 that i just purchased. I did the recent update this afternoon on the lumagen.



Many thanks


CMS is found in the Lumagen and stands for Color Management System. Stores settings used to calibrate display color.


Sent from my iPad using Tapatalk

Trinnov Altitude 16, Magico A3, Magico ACC, Magico A1s, Dual Rythmik FV25HP
Amps- ATI AT523NC, AT524NC, AT528NC
Kaleidescape Terra Server/Strato, Oppo UDP-203, Apple TV4k, XFinity X1
Sony VPL-VW5000, 136" 2.35:1 Stewart Studiotek 100 screen, Lumagen Pro 4242
dinamigym is offline  
post #4498 of 4870 Old 04-28-2019, 04:21 AM
Advanced Member
 
Roland Janus's Avatar
 
Join Date: Mar 2005
Location: Switzerland
Posts: 535
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 131 Post(s)
Liked: 33
Quote:
Originally Posted by bobof View Post
Hi Florian,

I believe the following will help your understanding for the above points; these are two shots I just took of DML400 with DTM enabled and disabled on the Masciola ramps. Deliberately underexposed to capture the highlights, so completely ignore the shadow detail (it is resolved almost all the way down with this 10,000 nit image, which I guess is as you would expect). Clearly there is detail resolved above the DML400 "limit" with DTM enabled. A couple of other shots also of DML set to 9900 and 50 to show the extremes. You can see with DML 50 it still appears to resolve up to well over 1000 nits, but that if you up it to 9900 you are resolving the few patches that DTM did clip at the lower settings (it actually starts to resolve all those boxes around 6000 if I recall).

Everything else in this image is factory I recall believe.

In terms of your other questions, I understand the point you are making; from what I can see if you had a 500 nit TV in a living room environment you'd actually want to set DML higher than 500 nits, and / or modify the Ratio I believe (don't have one so haven't tried to know exactly what). I think there are some levels of indirection here with the way the settings interact that make it a little harder to achieve a specific numeric goal, but at the same time you have to bear in mind that this has been built up in such a way that the significant number of folk who have had units professionally calibrated previously and set up with the old Intensity Mapping settings have been able to just upgrade their SW and it has added DTM seemingly without any additional configuration needed. Of course, at the moment most MadVR users of DTM are enthusiasts who are right "in the guts" of their tone mapping solution.

FWIW I am outputting P3 SDR with a 3DLUT to convert the native PJ gamut (x7900 profile off / filter on) to P3 2.4 gamma per Radiance requirements, 75 real nits on screen. Photos are just a Sony mobile phone so apologies for the quality, but I think they probably give you the idea.

Would that be a reliable method to determine MaxLight, using a ramp?

-roland
Roland Janus is offline  
post #4499 of 4870 Old 04-28-2019, 05:44 AM
aka jfinnie
 
Join Date: Aug 2015
Location: Norwich, UK
Posts: 2,869
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 2327 Post(s)
Liked: 1477
Quote:
Originally Posted by Roland Janus View Post
Would that be a reliable method to determine MaxLight, using a ramp?
Hi Roland,

I spent much of yesterday locked away in the cinema room trying to get my head around it and what should be the "correct" settings for my room. much of it prompted by trying to get my head around the questions @Soulnight mentions ("thanks"... no really I mean it, it is good to learn ) Still not really there and hopefully we could get a definitive guide to these parameters at some point.
I'm not sure(!) is the honest answer.

Per my understanding; if you had a movie scene which had all the pixels between 0 and 100 nits (notionally the diffuse white level if it was SDR) you'd want that SDR level to be around what you would have had your diffuse level for projection previously (so I guess around 41-55 nit depending on your preference?).

I've previously just set DML by eye, which recently with DTM has been around the 400 level, but we all know how fallible the Calibrated Eyeball Mk 1 is, so my honest preference would be for a solid method to set these parameters correctly via measurement.

With the default settings showing these kind of ramp images the level of the 100 nits patch gets dropped right down, and if you then removed the highlights and showed the image again you'd see the 100 nits using a much greater amount of the dynamic range.

If I leave the settings at default and just play with DML parameter, and enabling / disabling DTM (also checked the DTM settings), displaying just a 92 nit patch (it was the closest I had on UHD disc) then I actually have to raise my DML much higher than I had previously expected to get the diffuse white to be around the typical SDR levels I would have. I'd need to have my DML set to around 700 with my paltry 74 nits peak for the patch to look similar to how it would under SDR at typical levels. I spent some time yesterday watching some content at 700 and it looked great to be honest.

If you have, say, 150 nits like @Wookii has and DML at 500 then an SDR-esque scene within an HDR10 container (100 nits max content) would be much brighter than the equivalent from an SDR source configured for "normal" SDR projection levels.

DTM results in a much brighter image for that kind of scene with the same DML settings vs DTM off with just intensity mapping. See the table I did below.


For what it is worth (little really, but was interesting to look into) this is the MadVR tonemapped image from Lucy (courtesy of @Javs ) on my macbook photographed in front of my projection screen showing the same image from UHD source. At 700 nits DML it was a much closer match for how MadVR it seems would have mapped that image. As you lower the DML it gets much brighter in the midtones as expected; which perhaps looks more pleasing in isolation, but what about the intent?
Edit: I should add, in reality there is much more shadow detail in both images, but the camera crushes it to avoid clipping the whites.
Attached Thumbnails
Click image for larger version

Name:	Screenshot 2019-04-28 at 13.23.59.png
Views:	253
Size:	74.0 KB
ID:	2559368   Click image for larger version

Name:	DSC01149.jpg
Views:	271
Size:	395.1 KB
ID:	2559370  

Last edited by bobof; 04-28-2019 at 08:37 AM.
bobof is offline  
post #4500 of 4870 Old 04-28-2019, 07:36 AM
Senior Member
 
KarlKlammer's Avatar
 
Join Date: Apr 2010
Location: Germany
Posts: 299
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 250 Post(s)
Liked: 243
Quote:
Originally Posted by bobof View Post
I spent some time yesterday watching some content at 700 and it looked great to be honest.
That's interesting.

I'm in the range of 110-130 nits and use 700 since IM. That was a decision I made when I tested several ST.2084 LUTs. But for IM to work with 700 you have to tweak the other parameters a bit.

With DTM I can leave all parameters where they were.
bobof likes this.

Projection: JVC DLA-NX9
VP/Calibration: Lumagen Radiance Pro, LightSpace CMS, x-rite i1 Pro 2, x-rite i1 Display 3
KarlKlammer is offline  
Sponsored Links
Advertisement
 
Reply Video Processors

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