Improving Madvr HDR to SDR mapping for projector - Page 296 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 8448Likes
Reply
 
Thread Tools
post #8851 of 8884 Old 03-12-2020, 03:55 PM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,562
Mentioned: 373 Post(s)
Tagged: 0 Thread(s)
Quoted: 5799 Post(s)
Liked: 6253
Quote:
Originally Posted by *Mori* View Post
I would just copy my settings and see whether you still have flickering with these. If yes (and that's what I fear), then it's either the JVC or some settings in the madVR processing.
No, the new JVC 4K models don't have the instability of the older models on white or near white, so definitely not a projector issue

I don't have the time to do much testing at the moment, so I simply played with peak brightness as you have an exceptionally high one for a projector (175nits).

It looks like you're lucky. I tested with Neo-XP 1.0 (usual Lucy @ 24190) and the flicker is still there (though reduced) up to 160 nits.

However, set peak brightness to 175 nits, and... the flicker is gone

So it looks like it's a compression issue. When there is not enough headroom for the highlights and the curve is too agressive, it leads to flicker.

I'm sure that other parameters might play a role as well.

Anyway, I'm back to Neo-XP log conservative, as it doesn't flicker with my real peak of 115nits.

Could you try lowering your peak brightness to 115nits (possibly your dynamic target to 100nits as well) and confirm that it gives you flicker too, without changing any of the other parameters?

Hopefully it might help @madshi to resolve the flicker issue with more aggressive curves.

Thanks!
*Mori* likes this.

Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders
JVC Macro feature on Vertex/Vertex2/Integral2/Maestro
Manni01 is offline  
Sponsored Links
Advertisement
 
post #8852 of 8884 Old 03-13-2020, 05:12 AM
Senior Member
 
*Mori*'s Avatar
 
Join Date: Feb 2009
Posts: 326
Mentioned: 25 Post(s)
Tagged: 0 Thread(s)
Quoted: 342 Post(s)
Liked: 514
Quote:
Originally Posted by Manni01 View Post

It looks like you're lucky. I tested with Neo-XP 1.0 (usual Lucy @ 24190) and the flicker is still there (though reduced) up to 160 nits.

However, set peak brightness to 175 nits, and... the flicker is gone

So it looks like it's a compression issue. When there is not enough headroom for the highlights and the curve is too agressive, it leads to flicker.

I'm sure that other parameters might play a role as well.

Anyway, I'm back to Neo-XP log conservative, as it doesn't flicker with my real peak of 115nits.

Could you try lowering your peak brightness to 115nits (possibly your dynamic target to 100nits as well) and confirm that it gives you flicker too, without changing any of the other parameters?

Hopefully it might help @madshi to resolve the flicker issue with more aggressive curves.

Thanks!
Very interesting findings, Manni !

On my VW760 I have 2 brightness levels:

A) One with 185 real nits and dynamic target set at 170 nits. Here the laser otuput is set at 80%. Above 80% the acoustic noise becomes significantly louder then with the 760.

B) One with laser 55% giving me 155 nits. I am not sure which DT value I have set here since I prefer A) and I haven't used B) for a while now. But I think it is 140 nits. This setup was mainly for learning / testing.

115 nits: not sure whether I can go as low as that but it should be possible. For 120 nits I had to set the laser at 10% as far as I remember.

I really wish that your hypothesis turns out to be correct however I have a friend having only about 80 nits with his 760 in his setup. There we have done a lot of testing with DT values from 75-100 nits. Never we have noticed any flickering. Also with my Nero 3 plus (my reserve projektor) I didn't see any: here I got a really nice picture with 85 nits but only when I set the DT at 85 nits as well. With DT at 60 nits the contrast did not look good to me. Side story: My Sim2 dealer - normally selling much higher priced models - could not believe how good HDR looked with that little Nero 3 ... he keeps asking me to be allowed to show this to some of his customers. I told him to just set up a PC including madVR or to buy an Envy. This way he could have it all of the time


But I will check 115 real nits / DT 100 nits tonight with Lucy @ 24190 and report back.
*Mori* is offline  
post #8853 of 8884 Old 03-13-2020, 07:12 AM
AVS Forum Special Member
 
Slickman's Avatar
 
Join Date: Aug 2004
Location: New Jersey
Posts: 1,782
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 541 Post(s)
Liked: 504
I tried build 95 with the measurement files but they still didn't work. 🤷

Last edited by Slickman; 03-13-2020 at 07:54 AM.
Slickman is offline  
Sponsored Links
Advertisement
 
post #8854 of 8884 Old 03-13-2020, 07:47 AM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,562
Mentioned: 373 Post(s)
Tagged: 0 Thread(s)
Quoted: 5799 Post(s)
Liked: 6253
Quote:
Originally Posted by *Mori* View Post
Very interesting findings, Manni !

On my VW760 I have 2 brightness levels:

A) One with 185 real nits and dynamic target set at 170 nits. Here the laser otuput is set at 80%. Above 80% the acoustic noise becomes significantly louder then with the 760.

B) One with laser 55% giving me 155 nits. I am not sure which DT value I have set here since I prefer A) and I haven't used B) for a while now. But I think it is 140 nits. This setup was mainly for learning / testing.

115 nits: not sure whether I can go as low as that but it should be possible. For 120 nits I had to set the laser at 10% as far as I remember.

I really wish that your hypothesis turns out to be correct however I have a friend having only about 80 nits with his 760 in his setup. There we have done a lot of testing with DT values from 75-100 nits. Never we have noticed any flickering. Also with my Nero 3 plus (my reserve projektor) I didn't see any: here I got a really nice picture with 85 nits but only when I set the DT at 85 nits as well. With DT at 60 nits the contrast did not look good to me. Side story: My Sim2 dealer - normally selling much higher priced models - could not believe how good HDR looked with that little Nero 3 ... he keeps asking me to be allowed to show this to some of his customers. I told him to just set up a PC including madVR or to buy an Envy. This way he could have it all of the time


But I will check 115 real nits / DT 100 nits tonight with Lucy @ 24190 and report back.
Just to clarify, you don't need to get your PJ to match the peak brightness specified in madVR settings (say 115nits). If caused by the settings, the flicker should appear even if your actual peak is still 175nits and your laser is still at 55%.

I didn't use high lamp and put a new bulb in to reach 175nits when I tested that setting. The flicker went away while my PJ actual peak was still in low lamp, with 1000 hours, outputing 115nits... It was a dimmer picture of course (so I wouldn't use these settings), but it was flicker-free.

Not sure why you would not get flicker with 85nits if my theory is true, unless you didn't test specifically for flicker with these settings.

Looking forward to hearing more from you when you've had the time to do a quick test!
*Mori* likes this.

Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders
JVC Macro feature on Vertex/Vertex2/Integral2/Maestro
Manni01 is offline  
post #8855 of 8884 Old 03-13-2020, 08:00 AM
Senior Member
 
*Mori*'s Avatar
 
Join Date: Feb 2009
Posts: 326
Mentioned: 25 Post(s)
Tagged: 0 Thread(s)
Quoted: 342 Post(s)
Liked: 514
Ah, yes you are right and I hope to have understood it now: if it is a matter of how much compression then the real nits do not really matter. It is rather a matter of the ratio (or difference) between real nits and DT-value.
Manni01 likes this.
*Mori* is offline  
post #8856 of 8884 Old 03-13-2020, 08:08 AM
AVS Forum Special Member
 
claw's Avatar
 
Join Date: Jun 2001
Location: L'Etoile du Nord
Posts: 3,454
Mentioned: 124 Post(s)
Tagged: 0 Thread(s)
Quoted: 2738 Post(s)
Liked: 2581
Quote:
Originally Posted by Slickman View Post
I tried build 95 with the measurement files but they still didn't work. 🤷
Then maybe try with the current release v0.92.17.

http://madshi.net/madVR.zip

CJ
JVC RS500 | LG B7A OLED | Denon X6400H/X4200W | Panasonic UB820 | Two Oppo 203 | Samsung K8500 | Apple TV 4K | HDfury Diva/Vertex/Linker/Integral | MadVR with EVGA RTX-2080 Ti
claw is offline  
post #8857 of 8884 Old 03-13-2020, 08:25 AM
AVS Forum Special Member
 
Slickman's Avatar
 
Join Date: Aug 2004
Location: New Jersey
Posts: 1,782
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 541 Post(s)
Liked: 504
Quote:
Originally Posted by claw View Post
Then maybe try with the current release v0.92.17.

http://madshi.net/madVR.zip
I was using beta 40 which is newer than the official release. I will try today and see if I can find something newer that uses the measurement files.
Slickman is offline  
post #8858 of 8884 Old 03-14-2020, 07:18 AM
Newbie
 
Join Date: Sep 2017
Posts: 9
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 0
Quote:
Originally Posted by Slickman View Post
I was using beta 40 which is newer than the official release. I will try today and see if I can find something newer that uses the measurement files.

I'm using 112b and the measurement files work. Please check that you have "ignore all measurement files" unchecked in your settings.


Attached Thumbnails
Click image for larger version

Name:	ignoremeasurement.jpg
Views:	188
Size:	45.6 KB
ID:	2697498  
wow_deathbringer is offline  
post #8859 of 8884 Old 03-14-2020, 01:46 PM
AVS Forum Special Member
 
Slickman's Avatar
 
Join Date: Aug 2004
Location: New Jersey
Posts: 1,782
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 541 Post(s)
Liked: 504
They don't work for me, the image just displays extremely dark to where you can barely see anything. Oh well, the live algorithm is really really good now anyway.
Slickman is offline  
post #8860 of 8884 Old 03-14-2020, 01:56 PM
AVS Forum Addicted Member
 
zombie10k's Avatar
 
Join Date: Dec 2007
Posts: 13,149
Mentioned: 180 Post(s)
Tagged: 1 Thread(s)
Quoted: 5619 Post(s)
Liked: 6087
Quote:
Originally Posted by Slickman View Post
They don't work for me, the image just displays extremely dark to where you can barely see anything. Oh well, the live algorithm is really really good now anyway.
hi, I would skip the effort for now. if you have a GPU that can run everything at max settings the live algorithms are amazing in it's current state. I am still finding it hard to believe every time I watch a UHD movie.

We just finished watching the new Jumanji movie. HDR looked amazing and the wide color gamut was really impressive in a number of scenes. great detail in the low APL scenes as well, worth checking out!
stevenjw and Manni01 like this.
zombie10k is offline  
post #8861 of 8884 Old 03-14-2020, 02:19 PM
AVS Forum Special Member
 
Slickman's Avatar
 
Join Date: Aug 2004
Location: New Jersey
Posts: 1,782
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 541 Post(s)
Liked: 504
Quote:
Originally Posted by zombie10k View Post
hi, I would skip the effort for now. if you have a GPU that can run everything at max settings the live algorithms are amazing in it's current state. I am still finding it hard to believe every time I watch a UHD movie.

We just finished watching the new Jumanji movie. HDR looked amazing and the wide color gamut was really impressive in a number of scenes. great detail in the low APL scenes as well, worth checking out!
I watched it as well, looked great. I don't need the live algorithm for most movies since I use the program with a Samsung Q90R. In fact so far I've only found 3 movies that needed it and that was Tomb Raider, Batman vs Superman, and (especially)The Meg. Jumanji showed it was mastered to 4,000 nits but it looks more like the 1,000 nit movies and there was no highlight detail I could regain by using the live algorithm.
Slickman is offline  
post #8862 of 8884 Old 03-14-2020, 02:29 PM
AVS Forum Addicted Member
 
zombie10k's Avatar
 
Join Date: Dec 2007
Posts: 13,149
Mentioned: 180 Post(s)
Tagged: 1 Thread(s)
Quoted: 5619 Post(s)
Liked: 6087
Quote:
Originally Posted by Slickman View Post
I watched it as well, looked great. I don't need the live algorithm for most movies since I use the program with a Samsung Q90R. In fact so far I've only found 3 movies that needed it and that was Tomb Raider, Batman vs Superman, and (especially)The Meg. Jumanji showed it was mastered to 4,000 nits but it looks more like the 1,000 nit movies and there was no highlight detail I could regain by using the live algorithm.
hi, many of the folks in this specific thread are working with much lower nit capable projector setups and helping to improve the HSTM curves over the last few months. The live algorithms have made a huge impact for the projector community.
zombie10k is offline  
post #8863 of 8884 Old 03-14-2020, 09:06 PM
Senior Member
 
catav's Avatar
 
Join Date: Feb 2019
Location: Iowa
Posts: 347
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 361 Post(s)
Liked: 211
Quote:
Originally Posted by zombie10k View Post
hi, The live algorithms have made a huge impact for the projector community.
I agree whole hardily on my RS540, but I was also surprised to see an improvement on my LG HDR 400 nit monitor as well. Especially when I dialed down the "real nits" to around 150 or so. HSTM kicked in nicely. Same added 3D effect as projection, but with more nits. Real nice.
zombie10k likes this.

Last edited by catav; 03-14-2020 at 10:07 PM.
catav is offline  
post #8864 of 8884 Old 03-15-2020, 01:19 AM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 2,535
Mentioned: 69 Post(s)
Tagged: 0 Thread(s)
Quoted: 1385 Post(s)
Liked: 404
Where there any issues with Jumanji in the desert scenes?

What are your settings?

LE:

The issue in the scenes in the desert is the brightness goes up and down, like a dimming function.

Tried manipulating the settings.

Peak nits 43, dynamic nits 170. The only thing that reduced the effect (partially) is chaining the dynamic nits to 70.

Last edited by noob00224; 03-15-2020 at 03:28 AM.
noob00224 is online now  
post #8865 of 8884 Old 03-15-2020, 06:39 AM
Senior Member
 
*Mori*'s Avatar
 
Join Date: Feb 2009
Posts: 326
Mentioned: 25 Post(s)
Tagged: 0 Thread(s)
Quoted: 342 Post(s)
Liked: 514
Quote:
Originally Posted by *Mori* View Post
But I will check 115 real nits / DT 100 nits tonight with Lucy @ 24190 and report back.
Again no flickering with Neo-XP 1.0 even with that.

In that context: Since I was experimenting I had entered for min target / real display peak nits a wrong number of 120 nits (it should be actually 185 nits in my case), and set the compression limit also to 120 nits. Then I set the dynamic target nits to 200 nits. Guess what ? Despite I know it is not the correct way to do I liked it a lot . Maybe a way to avoid flickering ?
Manni01 likes this.
*Mori* is offline  
post #8866 of 8884 Old 03-18-2020, 08:53 AM
Member
 
Join Date: Apr 2016
Posts: 26
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 3
Hi!

I am also testing beta 113 and the new profiles. However, I have a question:

I am using a VW760 and get 136 nits on my screen. So basically what I did was to put the value 136 as „display peak luminance“ into the madVR settings. Was this the right place to input this value?

And for the value „dynamic target nits“ I entered a value of 100.

Is this correct for a proper comparison of the different curves?

If someone could shortly comment on this I would be very pleased.

Thanks in advance,
Chris
ix400 is offline  
post #8867 of 8884 Old 03-18-2020, 09:40 AM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 2,535
Mentioned: 69 Post(s)
Tagged: 0 Thread(s)
Quoted: 1385 Post(s)
Liked: 404
Quote:
Originally Posted by ix400 View Post
Hi!

I am also testing beta 113 and the new profiles. However, I have a question:

I am using a VW760 and get 136 nits on my screen. So basically what I did was to put the value 136 as „display peak luminance“ into the madVR settings. Was this the right place to input this value?

And for the value „dynamic target nits“ I entered a value of 100.

Is this correct for a proper comparison of the different curves?

If someone could shortly comment on this I would be very pleased.

Thanks in advance,
Chris
How have the brightness/nits of the screen measured?

If that is the correct nits then it's in the right place.

Dynamic target nits is a slider that adjusts HDR effect. Less for less HDR effect, more for more. It's a personal choice.

Set no compr limit to 0 for now.

Test the Contrast Recovery versions.

Do you use a 3DLUT?

Set the gamma:
https://forum.kodi.tv/showthread.php?tid=223175&page=37
noob00224 is online now  
post #8868 of 8884 Old 03-18-2020, 10:30 AM
Member
 
Join Date: Apr 2016
Posts: 26
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 3
Thank you!!!

I determined the nits with an X-Rite i1Display Pro.

I have activated „tone map HDR using pixel shaders“, 3DLUT is not activated. Gamma is set correctly to 2.2.

„No compression limit“ is set to 100. Seems to be wrong then...
Can you shortly comment in one or two sentences on the effect of the „No compression limit“ setting?
ix400 is offline  
post #8869 of 8884 Old 03-18-2020, 10:41 AM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 2,535
Mentioned: 69 Post(s)
Tagged: 0 Thread(s)
Quoted: 1385 Post(s)
Liked: 404
Quote:
Originally Posted by ix400 View Post
Thank you!!!

I determined the nits with an X-Rite i1Display Pro.

I have activated „tone map HDR using pixel shaders“, 3DLUT is not activated. Gamma is set correctly to 2.2.

„No compression limit“ is set to 100. Seems to be wrong then...
Can you shortly comment in one or two sentences on the effect of the „No compression limit“ setting?
No compression limit means the madvr tone mapping algorithm will not compress under the specified nits. If it's set to 100, madvr will not compress pixels under 100 nits, leaving 36 nits (in your case) to compress the rest.

If you want to save details under 50 for example set it to 50, and madvr will have 86 nits for the rest.

Most detail is under 300 nits, the rest are bright items and highlights. But 300 nits is still over whatever a projector can do.

A more detailed explanation:
https://forum.kodi.tv/showthread.php...309#pid2046309
Nombz likes this.
noob00224 is online now  
post #8870 of 8884 Old 03-18-2020, 10:50 AM
Member
 
Join Date: Apr 2016
Posts: 26
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 3
But with that explanation a value of 100 for no compression limit sounds not to bad.

Can a value of 0 for No compression limit then be seen as sort of an „auto“ value? Since compression is then allowed for the whole range?
ix400 is offline  
post #8871 of 8884 Old 03-18-2020, 10:52 AM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 2,535
Mentioned: 69 Post(s)
Tagged: 0 Thread(s)
Quoted: 1385 Post(s)
Liked: 404
Quote:
Originally Posted by ix400 View Post
But with that explanation a value of 100 for no compression limit sounds not to bad.

Can a value of 0 for No compression limit then be seen as sort of an „auto“ value? Since compression is then allowed for the whole range?
Can be set to whatever you like, it's not a rule.

0 means no compression limit, madvr has the projectors entire nit value to tone map to.
noob00224 is online now  
post #8872 of 8884 Old 03-18-2020, 11:10 AM
Member
 
Join Date: Apr 2016
Posts: 26
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 3
Quote:
Originally Posted by noob00224 View Post
Can be set to whatever you like, it's not a rule.

0 means no compression limit, madvr has the projectors entire nit value to tone map to.
Okay, thank you.

Sorry for all these questions. I‘m not the brightest guy, I‘m a bit stupid and have severe problems in understanding all this.

However, even with my probably not perfect madvr settings for my 136 nits projector I can already imagine how great madvr obviously is and I really would like to use it regularly.

Another little question: The dynamic target nits is then sort of a mean value to which the mapping is performed? Or how does this value relate to the display peak luminance value?
ix400 is offline  
post #8873 of 8884 Old 03-18-2020, 11:26 AM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 2,535
Mentioned: 69 Post(s)
Tagged: 0 Thread(s)
Quoted: 1385 Post(s)
Liked: 404
Quote:
Originally Posted by ix400 View Post
Okay, thank you.

Sorry for all these questions. I‘m not the brightest guy, I‘m a bit stupid and have severe problems in understanding all this.

However, even with my probably not perfect madvr settings for my 136 nits projector I can already imagine how great madvr obviously is and I really would like to use it regularly.

Another little question: The dynamic target nits is then sort of a mean value to which the mapping is performed? Or how does this value relate to the display peak luminance value?
It's not a literal nits value, more of a slider. Not sure what the upper limit is. Start with 100 with HDR files and see what works.

There's formulas for these things, but it just confuses new users.

Last edited by noob00224; 03-18-2020 at 11:29 AM.
noob00224 is online now  
post #8874 of 8884 Old 03-18-2020, 11:35 AM
Member
 
Join Date: Apr 2016
Posts: 26
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 3
... thank you!
ix400 is offline  
post #8875 of 8884 Old 03-30-2020, 06:54 PM
Member
 
ryrynz's Avatar
 
Join Date: Jan 2009
Posts: 40
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 15
Quote:
Originally Posted by noob00224 View Post
So I did some tests.
Really need to be using the same frame for comparison.
ryrynz is offline  
post #8876 of 8884 Old 04-02-2020, 08:52 PM
Member
 
ausvette's Avatar
 
Join Date: Jun 2007
Location: Gold Coast, Australia
Posts: 129
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 15
Quote:
Originally Posted by ryrynz View Post
Really need to be using the same frame for comparison.
I've been using madmeasurehdr for a long time. But development seems to have finished and the thread seems to be dead.

Is the dynamic toning mapping superior in the latest madvr beta to the madmeasurehdr method?
ausvette is offline  
post #8877 of 8884 Old 04-02-2020, 09:56 PM
Member
 
ryrynz's Avatar
 
Join Date: Jan 2009
Posts: 40
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 15
Based on the small amount of reading I've done here the dynamic tone mapping is good enough that all but the most hardcore image quality fanatics are using the dynamic tone mapping predominantly.
stevenjw likes this.
ryrynz is offline  
post #8878 of 8884 Old 04-02-2020, 11:51 PM
Advanced Member
 
hasta666's Avatar
 
Join Date: Apr 2016
Posts: 749
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 564 Post(s)
Liked: 202
I switched to the live algo a long time ago and don’t miss the measurement files.
hasta666 is online now  
post #8879 of 8884 Old 04-03-2020, 01:26 AM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,562
Mentioned: 373 Post(s)
Tagged: 0 Thread(s)
Quoted: 5799 Post(s)
Liked: 6253
Quote:
Originally Posted by ausvette View Post
I've been using madmeasurehdr for a long time. But development seems to have finished and the thread seems to be dead.

Is the dynamic toning mapping superior in the latest madvr beta to the madmeasurehdr method?
This thread isn’t dead, it is only active when madshi needs feedback on a specific build. He is currently busy working on getting the Envy ready for launch, so this thread is paused. It might get some action very soon...

The live algo works very well 99% of the time. Apart from a few blink-it-and-miss-it situations only visible if you watch the content frame by frame, the measurement files might only help with the flicker present with contrast enhancement algos that are too strong in some situations. This is not an issue if you use a conservative algo, such as Neo XP Log conservative (the one I use). Others can use stronger algos in their setup, but they flicker in mine (110 peak nits).

If madshi manages to “look into the future” for a few frames, which is an improvement that has been discussed for the live algo, it will make measurements files even less relevant, so I’d rather see this implemented when the time comes.

Anyway, I don’t miss measurements files and it makes sense for madshi to keep them disabled for now, as he needs feedback to improve the live algo even further, not to improve measurement files results for people who never use the live algo, hence never help to improve it.

IMO the files were useful when the live algo had big shortcomings, but it’s not the case anymore, provided your config is powerful enough, as enabling the files free a bit of resouces for those who were borderline.

So give the live algo a try, and let us know where you think it still falls short compared to measurement files.

That would be useful feedback!
stevenjw and Abdo Esmaeal like this.

Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders
JVC Macro feature on Vertex/Vertex2/Integral2/Maestro
Manni01 is offline  
post #8880 of 8884 Old 04-03-2020, 09:35 AM
AVS Forum Special Member
 
SamuriHL's Avatar
 
Join Date: Jan 2005
Posts: 5,259
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 279 Post(s)
Liked: 343
Manni, I think they meant the thread for the measurement tool but I could have misread.

LG C8 OLED | Pioneer VSX-LX303 | HTPC (9900k, RTX 2070) | nVidia SHIELD 2019 | Panasonic UB820
SamuriHL is online now  
Sponsored Links
Advertisement
 
Reply Digital Hi-End Projectors - $3,000+ USD MSRP

Tags
dynamic tone mapping , hdr , madvr , sdr , ton mapping

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