madVR Tool: MadmeasureHDR Optimizer (Measurements, Dynamic Clipping & Target Nits) - Page 37 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 990Likes
Reply
 
Thread Tools
post #1081 of 1179 Old 07-21-2019, 04:57 AM
Advanced Member
 
jmone's Avatar
 
Join Date: May 2007
Posts: 549
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 44
Quote:
Originally Posted by 3ll3d00d View Post
mad-measure-path should be the directory that contains the madMeasureHDR.exe, you can see the effect of what you've set in the lines above
First issued, is that madMeasureHDR.exe is now called madMeasureHDRDynamicOptimizer.exe. So I renamed it madMeasureHDR.exe and it pops up but the path to the MPLS is not passed correctly.
jmone is offline  
Sponsored Links
Advertisement
 
post #1082 of 1179 Old 07-21-2019, 04:59 AM
Advanced Member
 
jmone's Avatar
 
Join Date: May 2007
Posts: 549
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 44
From the Log I see: "2019-07-21 21:57:47,123 - Triggering : ['C:\\Users\\natha\\Downloads\\madMeasureHDRDynamic OptimizerV4.0.0\\madMeasureHDR.exe', 'M:\\Media\\Discs\\10 Cloverfield Lane\\BDMV\\PLAYLIST\\00800.mpls']" and the GUI opens but there it shows "Please select a file or folder"
jmone is offline  
post #1083 of 1179 Old 07-21-2019, 05:05 AM
AVS Forum Special Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 3,881
Mentioned: 255 Post(s)
Tagged: 0 Thread(s)
Quoted: 2324 Post(s)
Liked: 2471
You are getting confused I think (or I misunderstand you). You need to use madmeasurehdr.exe to create a measurements file, this is part of the madvr test build and is what my app does. Once you have a measurement file then you can use the GUI in this thread if you want to optimise using the tool provided by this thread.
3ll3d00d is online now  
Sponsored Links
Advertisement
 
post #1084 of 1179 Old 07-21-2019, 05:05 AM
Advanced Member
 
jmone's Avatar
 
Join Date: May 2007
Posts: 549
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 44
Quote:
Originally Posted by Soulnight View Post
That's easy.
Once you have generated the measurements files.
Just select the path(s) where all your BDMV are saved.

Enter your Real Nits.
Press Start.
Our tool will automatically find all the measurement files which have not been optimized yet in the chosen path(s).

That's it.
I have 1 common set of Folders that is then played by different HTPCs connected to a JVC X7500 PJ, Epson 9400, Sony OLED, LG OLED, and a Philips LED. Each HTPC has it's own madVR install with it's own settings. When you say "Enter your Real Nits", the Question is for what?
Thanks
Nathan
jmone is offline  
post #1085 of 1179 Old 07-21-2019, 05:10 AM
Advanced Member
 
jmone's Avatar
 
Join Date: May 2007
Posts: 549
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 44
Quote:
Originally Posted by 3ll3d00d View Post
You are getting confused I think (or I misunderstand you). You need to use madmeasurehdr.exe to create a measurements file, this is part of the madvr test build and is what my app does. Once you have a measurement file then you can use the GUI in this thread if you want to optimise using the tool provided by this thread.
No I'm confused! To many similar files names.... but I now get it.

running this over 1,000+ discs (but only a few '00 will have HDR)

madmeasurer.exe -d1 -vvv --main-by-jriver --mad-measure-path "C:\Users\natha\AppData\Roaming\J River\Media Center 25\Plugins\madvr" -m "M:\Media\Discs"
jmone is offline  
post #1086 of 1179 Old 07-21-2019, 05:11 AM
Advanced Member
 
jmone's Avatar
 
Join Date: May 2007
Posts: 549
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 44
I'll let it run overnight.....it's gonna take awhile!!! Thanks for the help.
jmone is offline  
post #1087 of 1179 Old 07-21-2019, 05:16 AM - Thread Starter
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,312
Mentioned: 196 Post(s)
Tagged: 0 Thread(s)
Quoted: 968 Post(s)
Liked: 1435
Quote:
Originally Posted by jmone View Post
First issued, is that madMeasureHDR.exe is now called madMeasureHDRDynamicOptimizer.exe. So I renamed it madMeasureHDR.exe and it pops up but the path to the MPLS is not passed correctly.
Once again: the measurement process itself was created by madshi and can be started from the file "madmeasurehdr.exe" that everybody can find in the madVR beta folder.

What Manni01 script or 3ll3d00d measurer tool are doing is call this exe from madshi to start the measurement.

The same is true for our tool which calls this "madmeasurehdr.exe" in madVR folder to start a measurement.

So of course, you should not rename our tool executable since it has nothing to do with it. ;-)

Once you have successfully generated a measurement file, either with the naked "madmeasurehdr.exe" from madshi, or with the different tool build around it which are calling the process in a more automatic fashion, you can then optimize the said measurement with our tool. :-)
Manni01 likes this.
Soulnight is offline  
post #1088 of 1179 Old 07-21-2019, 05:33 AM - Thread Starter
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,312
Mentioned: 196 Post(s)
Tagged: 0 Thread(s)
Quoted: 968 Post(s)
Liked: 1435
Quote:
Originally Posted by jmone View Post
I'll let it run overnight.....it's gonna take awhile!!! Thanks for the help.
Measuring is slow ~15min per movie.

Optimizing is fast ~10s per movie.

Once you have done the hard part "generate a measurement ", you can try various optimization pretty quickly. ;-)

Flo :-)
Soulnight is offline  
post #1089 of 1179 Old 07-21-2019, 05:38 AM
Advanced Member
 
jmone's Avatar
 
Join Date: May 2007
Posts: 549
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Liked: 44
I'm still on measuring my first movie and after 20min it is at 60%! At this rate is is '00s of Hours. Lets see how it goes overnight.
jmone is offline  
post #1090 of 1179 Old 07-21-2019, 06:39 AM
Member
 
vp4lifehva's Avatar
 
Join Date: Oct 2009
Posts: 43
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 18
Hi, I apologize if this has already been asked. I used the previous version of this tool and ran all of my 4k MKV files. Do I have to delete all of the measurement files and generate new ones with this current version?
vp4lifehva is offline  
post #1091 of 1179 Old 07-21-2019, 06:43 AM - Thread Starter
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,312
Mentioned: 196 Post(s)
Tagged: 0 Thread(s)
Quoted: 968 Post(s)
Liked: 1435
Quote:
Originally Posted by vp4lifehva View Post
Hi, I apologize if this has already been asked. I used the previous version of this tool and ran all of my 4k MKV files. Do I have to delete all of the measurement files and generate new ones with this current version?
Hi there. Of course not!
No new measurements needed.

Just re-run this new version of the tool.
It will re-optimize existing measurements files saved in the 'Original' folders (~10s per movie).
Soulnight is offline  
post #1092 of 1179 Old 07-21-2019, 06:50 AM
Member
 
vp4lifehva's Avatar
 
Join Date: Oct 2009
Posts: 43
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 18
Quote:
Originally Posted by Soulnight View Post
Hi there. Of course not!
No new measurements needed.

Just re-run this new version of the tool.
It will re-optimize existing measurements files saved in the 'Original' folders (~10s per movie).
I knew it haha. I should have asked beforehand. Luckily, I only deleted my #-E movie measurement folder so all is not lost. Thank you!
vp4lifehva is offline  
post #1093 of 1179 Old 07-21-2019, 10:05 AM
Member
 
Michael2907's Avatar
 
Join Date: Nov 2007
Posts: 63
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 0
There are some movies that dont measures.
Alien
Akita
Robin Hood
All Files Run normal

Gesendet von meinem SM-N960F mit Tapatalk
Michael2907 is online now  
post #1094 of 1179 Old 07-21-2019, 12:27 PM
Member
 
Join Date: Oct 2017
Posts: 76
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 57 Post(s)
Liked: 36
Thank's for the update Soulnight, it sounds great. I'm going to try it right now

I'm watching the latest Jurassic Park and the picture seem brighter than what im used to with the live algo

both pictures are dynamic tuning 100, target nits 100, dynamic clipping off.










Quote:
Originally Posted by Michael2907 View Post
There are some movies that dont measures.
Alien
Akita
Robin Hood
All Files Run normal

Gesendet von meinem SM-N960F mit Tapatalk

I didnt have any issue with Alita & Robin Hood. I never Tried Alien but I had some titles in the past that wouldnt work. the problem was related to the fps and mkvtoolnix. They were in 23.976 (23976/1000) FPS instead of 23.976 (24000/1001) FPS. Using makemkv fixed that issue.

Last edited by Salty01; 07-21-2019 at 01:55 PM.
Salty01 is offline  
post #1095 of 1179 Old 07-21-2019, 04:31 PM
AVS Forum Special Member
 
stef2's Avatar
 
Join Date: Mar 2005
Location: China
Posts: 1,275
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 483 Post(s)
Liked: 283
Thank you for this nice update! it has litterally eliminated any bothersome brightness jump on my display! (JVC RS500).


In the past, I used to scan my movies with your tool using display peak nits of 110 and target nits of 70.



From what I understand, I have rescanned my measurment files using the same display peak nits of 110. The target nits option has disappeared, though. Does the dynamic tuning option replace it?


If so, in order to obtain the same brightness as before, should I enter 70 under the dynamic tuning option, as I used to enter 70 in the former target nits option?



Thanks again for this extremely nice tool!
stef2 is offline  
post #1096 of 1179 Old 07-21-2019, 10:11 PM
Newbie
 
Join Date: Jul 2016
Posts: 2
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 2
Problems

Hello

I tried the last days to use this tool. but i can't get it running.
Don't get me wrong, i appreciate the work/tool but a little readme/howto in the start post would be nice for new people.

I had to search a lot just to find out i need LAV set to D3D11.
And also i was confused with the download. I thought the madVRhdrMeasure40.zip was it ,and wondered that nothing happened when clicking on it, until i found i had to go V4.0.0 post, and to that link.

So what i did now:
Extracted madVRhdrMeasure40.zip into my madvr v0.92.17 folder and overwrote files.
Extracted the madMeasureHDRDynamicOptimizer.exe into my madvr folder.
Reinstalled LAV Filter Installer 0.74.1 set Video to D3D11, (only works Windows 8 and above)

I load the tool, choose movie and madvr folder and click start.
A window pops up saying it can take up to 20 minutes, press skip/stop etc.

But there is no skip/stop button, and both windows are not responding.
I waited for 30 minutes, nothing happens, gpu load is @ 20% cpu 10%, movie is on SSD locally.

Used Windows 10 1903, i7-8700K, GTX 1070


What do i miss ? Are there any other requirements, like net framework, visual c run times etc ?

Thanks for your help.
Regards Sharonna
aerodynamics and Redhawk2000 like this.
sharonna is offline  
post #1097 of 1179 Old 07-21-2019, 11:01 PM
Member
 
Michael2907's Avatar
 
Join Date: Nov 2007
Posts: 63
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 0
Quote:
Originally Posted by Salty01 View Post
Thank's for the update Soulnight, it sounds great. I'm going to try it right now

I'm watching the latest Jurassic Park and the picture seem brighter than what im used to with the live algo

both pictures are dynamic tuning 100, target nits 100, dynamic clipping off.













I didnt have any issue with Alita & Robin Hood. I never Tried Alien but I had some titles in the past that wouldnt work. the problem was related to the fps and mkvtoolnix. They were in 23.976 (23976/1000) FPS instead of 23.976 (24000/1001) FPS. Using makemkv fixed that issue.
Take a Look on IT. Thx

Gesendet von meinem SM-N960F mit Tapatalk
Michael2907 is online now  
post #1098 of 1179 Old 07-22-2019, 09:32 AM
Member
 
Michael2907's Avatar
 
Join Date: Nov 2007
Posts: 63
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 0
Quote:
Originally Posted by Salty01 View Post
Thank's for the update Soulnight, it sounds great. I'm going to try it right now

I'm watching the latest Jurassic Park and the picture seem brighter than what im used to with the live algo

both pictures are dynamic tuning 100, target nits 100, dynamic clipping off.













I didnt have any issue with Alita & Robin Hood. I never Tried Alien but I had some titles in the past that wouldnt work. the problem was related to the fps and mkvtoolnix. They were in 23.976 (23976/1000) FPS instead of 23.976 (24000/1001) FPS. Using makemkv fixed that issue.
ITS Not that.
FPS are right

Gesendet von meinem SM-N960F mit Tapatalk
Michael2907 is online now  
post #1099 of 1179 Old 07-22-2019, 10:00 AM
Advanced Member
 
Neo-XP's Avatar
 
Join Date: Jun 2018
Location: Switzerland
Posts: 843
Mentioned: 135 Post(s)
Tagged: 0 Thread(s)
Quoted: 577 Post(s)
Liked: 699
Quote:
Originally Posted by Soulnight View Post
New Algo:
So the new algo is adressing the shortcoming of the old simple method while still using its knowledge of the future.
The new method is called "Smart Max Speed".
Goal is to give as much freedom to the target nits, while always controlling the speed of the target nits change (which should please @madshi ), while considering still considering the future and knowing what is the average target nits of each considered chapter.
I like this idea of a max speed.

The only problem I have found is with fade ins/out, because the speed is not fast enough (first scene of Samsung Chasing The Light Demo for instance).

But these special cases can be treated in a special way (immediately adapting the target on fade ins/outs, and only allowing the target going UP for fade ins and DOWN for fade outs), and for all other cases the max speed should be respected.

It also seems very easy to implement in the live algo, and safer than the brightness speeds

Thanks!
Manni01 and omarank like this.
Neo-XP is online now  
post #1100 of 1179 Old 07-22-2019, 10:19 AM - Thread Starter
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,312
Mentioned: 196 Post(s)
Tagged: 0 Thread(s)
Quoted: 968 Post(s)
Liked: 1435
Quote:
Originally Posted by Salty01 View Post
Thank's for the update Soulnight, it sounds great. I'm going to try it right now

I'm watching the latest Jurassic Park and the picture seem brighter than what im used to with the live algo

both pictures are dynamic tuning 100, target nits 100, dynamic clipping off.
Hi there.
Well the "brighter" is probably not due to the new smoothing algo.
Probably, it's due to the "detect and remove flat bright area" algo which is working at 100% in PRESET3.
As a result, FALL corrected value can get lower, which in turns drops the calculated target nits, which raises brightness.

You could turn the "flar bright area/sky algo" OFF to see if it's the reason.



Quote:
Originally Posted by Salty01 View Post

I didnt have any issue with Alita & Robin Hood. I never Tried Alien but I had some titles in the past that wouldnt work. the problem was related to the fps and mkvtoolnix. They were in 23.976 (23976/1000) FPS instead of 23.976 (24000/1001) FPS. Using makemkv fixed that issue.
@Manni01 and @Salty01 : is there any recommended way to "repair" a mkv giving INCOMPLETE measurement using mkvtoolnix, or the only way is to re-rip using maybe a different tool or audio track?
Soulnight is offline  
post #1101 of 1179 Old 07-22-2019, 10:24 AM - Thread Starter
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,312
Mentioned: 196 Post(s)
Tagged: 0 Thread(s)
Quoted: 968 Post(s)
Liked: 1435
Quote:
Originally Posted by stef2 View Post
Thank you for this nice update! it has litterally eliminated any bothersome brightness jump on my display! (JVC RS500).

In the past, I used to scan my movies with your tool using display peak nits of 110 and target nits of 70.

From what I understand, I have rescanned my measurment files using the same display peak nits of 110. The target nits option has disappeared, though. Does the dynamic tuning option replace it?

If so, in order to obtain the same brightness as before, should I enter 70 under the dynamic tuning option, as I used to enter 70 in the former target nits option?

Thanks again for this extremely nice tool!
Hi Stef. Glad you like it.
Can you tell us where dud you see "bothersome brightness jump" before?
With which settings did you issues before? With the tool or live algo?

Yes, you only need to enter your "real display nits" and the "dynamic tuning option".
(There was never a target nits option, maybe you are referencing to the "static target nits" field. which is still there)


Thx,
Flo
Soulnight is offline  
post #1102 of 1179 Old 07-22-2019, 10:36 AM
AVS Forum Special Member
 
Manni01's Avatar
 
Join Date: Sep 2008
Posts: 9,038
Mentioned: 314 Post(s)
Tagged: 0 Thread(s)
Quoted: 5384 Post(s)
Liked: 5465
Quote:
Originally Posted by Soulnight View Post
@Manni01 and @Salty01 : is there any recommended way to "repair" a mkv giving INCOMPLETE measurement using mkvtoolnix, or the only way is to re-rip using maybe a different tool or audio track?
It depends on the reason. With some titles, there is a bug or an undesired behavior with makemkv when the sound track is longer than the picture track. In that case, makemkv reports a length of the longest track (audio) and madmeasureHDR reports .incomplete because it can't find the video matching the audio. There is a very simple way to check if this is the reason for the .incomplete file: play the mkv, and if at the end you have a black screen but can't parse it forward or backward using the player's controls, that's the reason. Titles having this issue include Fantastic Four (US) and a few others. In that case, remuxing with mkvtoolnix will solve the issue, because mkvtoolnix reports the length of the shortest track, so madVR stops before it runs out of video. Mike has been made aware of this issue, but it's very unlikely that it will be solved any time soon.

There are some titles that do not measure, for a different reason, irrespective of the tool used. Robin Hood UK is one of them, and so is Despicable Me UK or Pacific Rim UK. I have no idea what the problem is with these titles, because they have issues even when you measure the .mpls directly (I use BD Folders now, not .mkvs anymore), so clearly not a remuxing issue with these. So I'm afraid that I have no solution for these. Not a bad rip either, I ripped them myself from the original discs and they play fine, just can't be measured.
Soulnight likes this.

JVC Autocal Software V11 Calibration for 2019 Models (Google)
Batch Utility V4.02 May 16 2019 to automate measurements files for madVR with support for BD Folders

Last edited by Manni01; 07-22-2019 at 12:15 PM.
Manni01 is offline  
post #1103 of 1179 Old 07-22-2019, 11:10 AM - Thread Starter
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,312
Mentioned: 196 Post(s)
Tagged: 0 Thread(s)
Quoted: 968 Post(s)
Liked: 1435
Quote:
Originally Posted by Neo-XP View Post
I like this idea of a max speed.

The only problem I have found is with fade ins/out, because the speed is not fast enough (first scene of Samsung Chasing The Light Demo for instance).

But these special cases can be treated in a special way (immediately adapting the target on fade ins/outs, and only allowing the target going UP for fade ins and DOWN for fade outs), and for all other cases the max speed should be respected.

It also seems very easy to implement in the live algo, and safer than the brightness speeds

Thanks!
Hi Neo-XP Thanks a for the feedback.

Can you explain a bit more:
You used PRESET 3 and saw something distracting with fade in/out? Can you describe it /document it?
How is "slow" being distracting?

------------------------------


The "max speed" idea could very easily be implemented in madVR LIVE algo, and it would look probably very bad most of the time.
The reason why is that the LIVE algo does not know the future, and even when it will it will be only for a few frames.


Knowing ALL the future, the "max speed" can turn to a a "SMART max speed" instead and will then look good.
I already explained it in the bigger post but let's try here a diffrent version of the explanation

Let's say we have identified a 10min long "Chapter".
- The new "Smart max Speed algo", will first calculate a LARGE rolling average of 1min for each single frames.
- Then it will look for all the "crossing points" where the ideal target value for each single frames is crossing the rolling average values.

To make it even more simple, let's imagine one single average Target Nits value of 300nits for the whole 10min chapter (That's how we had started anyway ).
Then we scan the whole chapters and create sub-section of the chapters:
- Consecutive Frames with ideal target nits ABOVE the chapter average target nits
- Consecutive Frames with ideal target nits BELOWthe chapter average target nits

Each time, the ideal target cross the "Chapter average", we set a new crossing point.

Then we give to EACH crossing point frame, a target nits equal to the average of the whole chapter.
And in between 2 consecutive crossing point, we can now calculate the MAX SPEED.
But you need to meet the other crossing point set at the average target nits.
So with start from the first crossing point, use the max speed and go forward: we get a "LEFT MAX SPEED Target nits for every frames between the 2 corner points.
But of course the last frame will not be back "by it self" at the average target Nits.
So we do the same thing, but backwards in time starting from the next crossing point: we get a "RIGHT MAX SPEED Target nits for every frames between the 2 corner points.

And then we look for where the LEFT and RIGHT Max Speed Target Nits intersect, and keep the left part of the LEFT target nits lists, and the right part of the RIGHT target nits LIST.

Like this, no matter what, you are always close to the "Chapter Average" at the crossing points.
And actually, we even use a more local "LARGE rolling average of 1minute" instead of the chapter average to make it more local than the full 10min chapter length.

-------------------

-->In any case, the LIVE algo does not know the chapter average, or the local 1min rolling average, of the chapter length even.
If you fix the speed at a certain rate in the LIVE algo (which is somehow already what is beiing done even if the speed can vary), you have to start from somewhere and from a certain target nits.
And this target nits will probably NOT be (except if your lucky) the chapter average.
Now let's say the target nits goes down for 5s, and then go back up VERY quickly and cross the average again while always going higher.
What will happen is that the LIVE algo will be surprised, will now try to go up again, but will be WAY too low (not at the average of the chapter) when it crosses the average, and the next bright scenes requireíng higher target nits will be overblown.
That's actually already the case to some extend with the LIVE algo. But it#s mitigated with target nits RESET at each camera cut.
And as we know, resetting at each camera cuts leads to sometimes strange brightness jump like in the "little girl intro scene of the MEG" or the "Interrogation scene" in Venom.


--------------

What we have implemented with the new "SMART Max Speed" is actually very similar to what Madshi implemented with together with the measurement file for the frame peak nits. (We got the idea from what he did there, the difficulty was to translate it to the target nits world).
The measurement file was at the begining created for the sole purpose of knowing the future so that madshi could exactly always choose a frame peak nits which would NEVER CLIP while always moving at a FIXED invisible Speed.
Only, for him the "local scene peak maximums" during a identified scene were his crossing points.
Where for us, those crossing point relate to the "Chapter average, or local 1min rolling average".

So as good and impressive as the LIVE algo is, it has its limitations because it can and will get surprised and CLIP (peak nits) or be a bit late to adapt brightness (too dark, too bright) with dynamic target nits.


Hopes this clarify what we are doing.
Flo
Manni01, omarank, Salty01 and 1 others like this.

Last edited by Soulnight; 07-22-2019 at 12:01 PM.
Soulnight is offline  
post #1104 of 1179 Old 07-22-2019, 12:03 PM
Advanced Member
 
Neo-XP's Avatar
 
Join Date: Jun 2018
Location: Switzerland
Posts: 843
Mentioned: 135 Post(s)
Tagged: 0 Thread(s)
Quoted: 577 Post(s)
Liked: 699
Quote:
Originally Posted by Soulnight View Post
Hi Neo-XP Thanks a for the feedback.

Can you explain a bit more:
You used PRESET 3 and saw something distracting with fade in/out? Can you describe it /document it?
How is "slow" being distracting?
Yes, PRESET 3 with FALL algo, 150 real nits and 50 dynamic tuning.
On the first fade in from "Samsung Chasing The Light Demo", the brightness goes UP and then, instead of stabilizing, it goes visibly DOWN.
On passthrough HDR mode it goes UP and then doesn't move at all. I guess it is because the target is late and has to catch up when the fade in is done, but I'm not sure...

Quote:
Originally Posted by Soulnight View Post
Hopes this clarify what we are doing.
Thank you for the explanation, I did not understand the "Smart" part before, now it is clear. It was too good to be true for the LIVE algo, because it is very smooth now with your tool...
I still think we can achieve a very professional result by only seeing a few frames ahead with the LIVE algo, to avoid target reset on flashes on part of the image, and to smooth out the target adaptation as much as we can.
Neo-XP is online now  
post #1105 of 1179 Old 07-22-2019, 04:08 PM
AVS Forum Special Member
 
stef2's Avatar
 
Join Date: Mar 2005
Location: China
Posts: 1,275
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 483 Post(s)
Liked: 283
Quote:
Originally Posted by Soulnight View Post
Hi Stef. Glad you like it.
Can you tell us where dud you see "bothersome brightness jump" before?
With which settings did you issues before? With the tool or live algo?

Yes, you only need to enter your "real display nits" and the "dynamic tuning option".
(There was never a target nits option, maybe you are referencing to the "static target nits" field. which is still there)


Thx,
Flo
Hi! there was no target nits option, it was called dynamic target nits, and there I used to enter 70.
The brightness jumps I used to see were with the tool. Though, I was never really able to tell if the live algo or the tool was being used.
stef2 is offline  
post #1106 of 1179 Old 07-22-2019, 09:11 PM
Senior Member
 
Join Date: Dec 2018
Location: Seattle, WA, USA
Posts: 253
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 290 Post(s)
Liked: 174
Some people seem confused about how to install this, so here is a quick set of steps:
  1. If you haven't already built a HTPC and installed MadVR with a compatible media player, follow this guide to get the base MadVR configuration: https://www.avsforum.com/forum/26-ho...tpc-madvr.html
  2. After you have the basic config, you can enable the Dynamic Tone mapping and other fancy features from the MadVR Beta builds. Download one of the ZIP files for a beta build and extract it over your MadVR installation. Replace all existing files. http://madshi.net/madVRhdrMeasure40.zip is the version recommended in this thread, but http://madshi.net/madVRhdrMeasure86.zip is what I use and it seems to work ok for me. Newer versions might also be available some time after this post.
  3. Now, download the Dynamic Optimizer, but don't extract it to your MadVR install directory. The dynamic optimizer is just a separate tool, not a replacement for anything in your MadVR install directory. You can run the .exe or run the setup. Either is fine. http://projectiondream.com/download/...amicoptimizer/
  4. That's it. Run the tool, point it at your MadVR install directory, tell it what videos to measure, and you're all set.


Hopefully that helps. Let me know if I missed anything. That's the process that worked for me. I made the mistake of replacing the madMeasureHDR.exe file with the madMeasureHDRDynamicOptimizer.exe file thinking that it was just a new version of that (from the rename) but that was not correct. The tool threw an error if I did that, for now obvious reasons.
Soulnight and Redhawk2000 like this.
wombats is offline  
post #1107 of 1179 Old 07-22-2019, 09:22 PM
Senior Member
 
Join Date: Dec 2018
Location: Seattle, WA, USA
Posts: 253
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 290 Post(s)
Liked: 174
So I'm a bit confused as to what the purpose of this DynamicOptimizer is. I just did a few tests and I don't notice any major GPU or CPU usage differences between the following three configurations:
  1. The base MadVR install with no beta builds; default HDR tone mapping setup. I get 16-20% CPU usage and an oscillation between 20% and 60% GPU usage. All default settings. The stats/debug doesn't indicate any frame measurements.
  2. MadVR beta build 86 with all default settings. The stats/debug shows each frame measurement and a tone map value. Cool. This is what I've been using for the past few months. The CPU and GPU usage is about the same as default base MadVR though.
  3. I download the DynamicOptimizer tool, run setup, and then run the tool. I measured the movie, it ran perfectly, the measurement files are where I expect them. When I run the movie the stats/debug I see additional measurement info, frame/avg/scene/movie and the tone map value. Cool, something is working, but what? The CPU and GPU usage is still the same as well.

So what exactly is this tool doing? I figured it was a way to pre-process the frame measurements and save a little bit of compute power, but I'm not seeing it. Is there any image quality benefit to this tool? Maybe the difference would be more apparent on a slower GPU? What exactly am I gaining my pre-measuring my videos?
wombats is offline  
post #1108 of 1179 Old 07-23-2019, 12:22 AM
Member
 
Michael2907's Avatar
 
Join Date: Nov 2007
Posts: 63
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 30 Post(s)
Liked: 0
have again run mkvtoolnix over it, you can see synonymous with the size of what has changed, but file is still not recognized without error. A comparison of different files with mkv Info has shown no relevant differences.

Gesendet von meinem SM-N960F mit Tapatalk
Michael2907 is online now  
post #1109 of 1179 Old 07-23-2019, 10:30 AM - Thread Starter
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,312
Mentioned: 196 Post(s)
Tagged: 0 Thread(s)
Quoted: 968 Post(s)
Liked: 1435
Quote:
Originally Posted by Neo-XP View Post
Yes, PRESET 3 with FALL algo, 150 real nits and 50 dynamic tuning.
On the first fade in from "Samsung Chasing The Light Demo", the brightness goes UP and then, instead of stabilizing, it goes visibly DOWN.
On passthrough HDR mode it goes UP and then doesn't move at all. I guess it is because the target is late and has to catch up when the fade in is done, but I'm not sure...


Thank you for the explanation, I did not understand the "Smart" part before, now it is clear. It was too good to be true for the LIVE algo, because it is very smooth now with your tool...
I still think we can achieve a very professional result by only seeing a few frames ahead with the LIVE algo, to avoid target reset on flashes on part of the image, and to smooth out the target adaptation as much as we can.
So I downloaded this demo to test myself if I could see the visible brightness change you have mentionned.
And to my surprise, I did see it.

But the surprise was only short this demo is in 60 frames per second.
And the 0.5% allowed target nits change PER frame we had defined was for 24p content.

So, to look at this demo with the same brightness max speed / seconds as we had in mind for 24p content, you have to mutiply the 5 by 24/60=0.4.
So for 60p content the "maximum target adaptation speed" you/we/everybody should use should be 5*0,4=2 instead.

I know that we have to change this logic "PER frame" to a "PER second" in the future.
It's on our TO DO list But we have to somehow grab the framerate information with something that everybody has already installed on their computer (it's not saved in the measurement file itself but maybe that's something @madshi could add?). Or we could require everybody using the tool to install "mediainfo" and get the framerate from there...
But for the moment, for us, it was not that important since most everything we are watching is in 24p.
Mostly only Demos and Billy Lynn is in 60p.

And I don't think the adaptation you saw had something to do with the fade in. I believe it had to do with next camera cut coming afterwards which has a higher target nits requirement. So if the max speed it set too high, then you will perceive the change to those coming higher target nits. PRESET 3 is handling the whole demo as 1 single chapter.
But with the same speed per second as we defined for 24p (5 for 24p, 2 for 60p), I don't perceive the issue anymore despite going to higher target nits because it's a much smoother change.

Everyone sensibility can be however different, so if 5 for 24p, or 2 for 60p is still too fast, then you have the option to lower the value to render it invisible to your eyes as well.
Right now, 5 for 24p was "calibrated" for Anna and my sensibility. Hopes this fit most of the population.

Flo
Manni01 likes this.

Last edited by Soulnight; 07-23-2019 at 10:47 AM.
Soulnight is offline  
post #1110 of 1179 Old 07-23-2019, 11:19 AM - Thread Starter
AVS Forum Special Member
 
Soulnight's Avatar
 
Join Date: Dec 2011
Location: Germany
Posts: 1,312
Mentioned: 196 Post(s)
Tagged: 0 Thread(s)
Quoted: 968 Post(s)
Liked: 1435
Quote:
Originally Posted by wombats View Post
So what exactly is this tool doing? I figured it was a way to pre-process the frame measurements and save a little bit of compute power, but I'm not seeing it. Is there any image quality benefit to this tool? Maybe the difference would be more apparent on a slower GPU? What exactly am I gaining my pre-measuring my videos?
From my previous personal testing and some other reporting from other madVR user, running with measurement files DID save rendering time. Maybe something was changed in the latest beta version.

But the main focus of the measurement file and its optimization through our tool is to improve quality of madVR HDR tone mapping.

Here a bit of explanation /history:

1) The measurement process was introcuded by Madshi to know in advance all the scenes cuts and all the frames peak nits.
Initial goal was there to get rid of ANY unwanted clipping (through knowledge of the future) while changing the peak value used in the tone mapping with a fixed speed to stay invisible.
-->The LIVE algo does not know what are the next frame which are coming. Therefore if it's currently tone mapping up to 500nits, but the next frame peak is 2000nits, it will NOT tone map to 2000nits because it would create visible flicker. Instead, it will maybe tone map slightly higher than 500nits, maybe 510nits, and everything between 510nits and 2000 will be clipped.

2) Then, madshi explained us the structure of the measurement file and we created in the tool something called "dynamic clipping".
For every frame, we look for a highlight knee to overwrite the real frame peak with an "effective peak" to gain back brightness and contrast/HDR in the picture at the cost of clipping only a few pixels above the said detected "highlight knee".
The algo was partially implemented in the LIVE algo, but because of 1), it can lead to visible clipping since it then combines the uncontrolled clipping of the LIVE algo + the controlled clipping from the dynamic clipping. That's why most people keep "dynamic clipping" deactivated in the LIVE algo but keep it activated at 100% with the optimizer tool to get an extra "punch in the picture".

3) Then we wanted for a long time to try out a "dynamic Target nits" per frame. For Christmas , madshi allowed the measurement file to hold such "Target Nits per frame" so we could play around but there is dynamic target nits present in the measurement file to begin with.
We then started a small journey with the madVR testing community here to develop a nice method and algo for a target nits per frame.
We came up with the FALL algo, chapter detection (containing mutiple camera cut) and centered rolling avg.

Since the the dynamic target nits idea and result was a success, it was implemented by Madshi in the LIVE algo. Only, once again, the LIVE algo does not know the future which means it can get surprised if the FALL gets higher or lower which can result in a too bright or too dark picture when this happen.
Also, to mitigate this, the LIVE algo logic is to reset the target nits at every camera cut, since it is mostly invisible when you do it there.
But, resetting at every camera cut can lead to strong brightness change between 2 following camera cut for 2 persons speaking in the same room or outside (Little girl intro scene in the MEG, Interrogation scene in Venom, are both good examples).

The optimizer tool however has access to the whole movie histogramm per frame saved in the measurement. So it knows in advance what's coming in the future.
Therefore, the tool can completely smooth out all the camera cut and just roll smartly over everything because it knows in advance what's coming and can prepare for it (same logic that was madshi implemented for the peak nits with the measurement).


Conclusion: the LIVE algo looks great in itself but since it cannot know the far future it has limitation that the optimizer tool together with the measurement file does not have.
So best quality, per frame but also smoothness wise will be adchieved with the measurement process introduced by madshi.
In itself, the LIVE algo still looks pretty awesome and does not need any measurement and can/will be integrated in the ENVY to get madVR awesome tone mapping for ANY source!

Like always, the last few percent of quality are always the hardest to achieve, and therefore that's not always for "everybody".
Here it means you have to perform a full measurement of the movie before hand and optimize it. So you loose the convenience of just starting a movie at anytime IF you did not do those steps yet.
But the tool makes those steps simple, even it measurement stays time consuming (~15min per movie with a GTX1080ti).


Hopes this clarify what our tool is for.
And my thanks to Madshi for making all this possible!

Flo

Last edited by Soulnight; 07-23-2019 at 11:29 AM.
Soulnight is offline  
Sponsored Links
Advertisement
 
Reply Home Theater Computers

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