Guide: Building a 4K HTPC for madVR - Page 133 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 998Likes
Reply
 
Thread Tools
post #3961 of 4010 Old 01-13-2020, 01:46 PM
AVS Forum Special Member
 
SirMaster's Avatar
 
Join Date: Aug 2009
Location: Wisconsin
Posts: 1,087
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 922 Post(s)
Liked: 598
Quote:
Originally Posted by noob00224 View Post
To use madvr's DTM where the real nits is required. Screen is Cinegrey 3D which the manufacturer claims is 1.2, but the 5D was supposed to be 1.5 and in testing was 1.0. I don't know if the the 3D is also 33% as well.
I can understand that you want to know your screen's nits, but you don't need to know the gain to know the nits.

If you just place your i1 Display Pro back from the screen a bit and point it at your screen from a similar viewing angle as you sit at, it will read and tell you the nits of your screen including the gain since it's reading from the screen.


I can read my screen with a normal light meter, held up against the screen and reading from the lens and I get 20 foot candles.

When I then use my i1 display Pro from about 7ft back reading my screen from the same angle that I view it from with my eyes, I read 75 nits. This would be 22 foot lamberts which is a 10% higher reading than the light meter got from the lens.

And my screen claims that it's 1.1 gain so it seems to check out that the i1 Display Pro is reading the nits close enough.

Unless you have a really high end light meter, I would say the i1 Display Pro should be reading the nits accurately enough. Or at least just as accurately as you trying to find the gain of your screen vs a piece of white paper and then applying that to your light meter reading to find your nits.

There are too many variables with your method that I do not believe you would get a more accurate result than just reading your screen with the i1 Display Pro and measuring the nits directly.
noob00224 likes this.

Last edited by SirMaster; 01-13-2020 at 01:54 PM.
SirMaster is online now  
Sponsored Links
Advertisement
 
post #3962 of 4010 Old 01-13-2020, 01:48 PM
Member
 
Join Date: Jan 2020
Posts: 15
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Quote:
Originally Posted by noob00224 View Post
The rendering value is still missing.

Have you tried the restore .bat?
Yes I have. Same info. I have never seen these rendering values, not even after fresh install windows. Does anyone have a solution?
SuurbieR is offline  
post #3963 of 4010 Old 01-13-2020, 02:02 PM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 1,772
Mentioned: 49 Post(s)
Tagged: 0 Thread(s)
Quoted: 952 Post(s)
Liked: 270
Quote:
Originally Posted by SirMaster View Post
I can understand that you want to know your screen's nits, but you don't need to know the gain to know the nits.

If you just place your i1 Display Pro back from the screen a bit and point it at your screen from a similar viewing angle as you sit at, it will read and tell you the nits of your screen including the gain since it's reading from the screen.


I can read my screen with a normal light meter, held up against the screen and reading from the lens and I get 20 foot candles.

When I then use my i1 display Pro from about 7ft back reading my screen from the same angle that I view it from with my eyes, I read 75 nits. This would be 22 foot lamberts which is a 10% higher reading than the light meter got from the lens.

And my screen claims that it's 1.1 gain so it seems to check out that the i1 Display Pro is reading the nits close enough.

Unless you have a really high end light meter, I would say the i1 Display Pro should be reading the nits accurately enough. Or at least just as accurately as you trying to find the gain of your screen vs a piece of white paper and then applying that to your light meter reading to find your nits.

There are too many variables with your method that I do not believe you would get a more accurate result than just reading your screen with the i1 Display Pro and measuring the nits directly.
Right, I don't need to know the gain of the screen, just the nits.

Does the i1Display Pro's wider viewing cone help here? If not I'd just get the ColorMunki.
noob00224 is offline  
Sponsored Links
Advertisement
 
post #3964 of 4010 Old 01-13-2020, 02:06 PM
AVS Forum Special Member
 
SirMaster's Avatar
 
Join Date: Aug 2009
Location: Wisconsin
Posts: 1,087
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 922 Post(s)
Liked: 598
Quote:
Originally Posted by noob00224 View Post
Right, I don't need to know the gain of the screen, just the nits.

Does the i1Display Pro's wider viewing cone help here? If not I'd just get the ColorMunki.
I don't think the viewing cone on the i1 Display Pro is any wider.

And even if it was, I would think you would prefer a smaller viewing cone so you can move the meter farther away from the screen without risking the viewing cone going off the top and bottom of the screen.

There is no more accurate place to calibrate your screen from than placing the meter actually in the MLP so it sees exactly what your eyes see. But with the wide view cone on the i1 Display Pro, you have to move it closer to the screen otherwise it will read beyond the screen frame on the top and bottom.
noob00224 likes this.

Last edited by SirMaster; 01-13-2020 at 02:19 PM.
SirMaster is online now  
post #3965 of 4010 Old 01-13-2020, 02:18 PM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 1,772
Mentioned: 49 Post(s)
Tagged: 0 Thread(s)
Quoted: 952 Post(s)
Liked: 270
Quote:
Originally Posted by SirMaster View Post
I don't think the viewing cone on the i1 Display Pro is any wider.

And even if it was, I would think you would prefer a smaller viewing cone so you can move the meter farther away from the screen without risking the viewing cone going off the top and bottom of the screen.

There is no more accurate place to calibrate your screen from but placing the meter actually in the MLP so it sees exactly what your eyes see. But with the wide view cone on the i1 Display Pro, you have to move it closer to the screen otherwise it will read beyond the screen frame on the top and bottom.
The screen has a hotspot so further away would be better.
What's MLP? Google is not helpful.

How can you tell if the field of view of the colorimeter exceeds the screen edges?
noob00224 is offline  
post #3966 of 4010 Old 01-13-2020, 02:30 PM
AVS Forum Special Member
 
SirMaster's Avatar
 
Join Date: Aug 2009
Location: Wisconsin
Posts: 1,087
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 922 Post(s)
Liked: 598
Quote:
Originally Posted by noob00224 View Post
The screen has a hotspot so further away would be better.
What's MLP? Google is not helpful.

How can you tell if the filed of view of the colorimeter exceeds the screen?
MLP is just a HT term people seem to use for main listening position. I guess it applies for audio but why not video too since it's the same position in a home theater. It just means the main seat.

According to the specifications for the i1 Display Pro, the viewing angle is 160 degrees. So you would need to calculate that from the distance.

https://keisan.casio.com/exec/system/1273850202

Use the height and vertex angle option and the vertex angle is 80 and the height is the meter distance.
neo_2009 and noob00224 like this.
SirMaster is online now  
post #3967 of 4010 Old 01-13-2020, 04:02 PM
Member
 
Join Date: Nov 2019
Posts: 123
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 64 Post(s)
Liked: 19
So I have everything hooked up.
At first when playing uhd hdr content and screen was purple but did some changes in nvidia control panel and got rid of it.

Question is how do I start this dtm
Right now I have this. Ignore dropped frames they haven’t moved since I made changes
Attached Thumbnails
Click image for larger version

Name:	12873164-8618-4FC9-A4FF-0F78A142C11B_1578956551534.jpg
Views:	34
Size:	103.2 KB
ID:	2669484  
Bellaseni is offline  
post #3968 of 4010 Old 01-13-2020, 04:07 PM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 2,374
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 1319 Post(s)
Liked: 475
devices -> "your device name" -> hdr -> tone map using pixelshader. after that you can switch bad to RGb output because you are not using HDR anymore.

BTW-don't use DXVA2 decoding and don't use NGU sharp very high on chroma.
mightyhuhn is offline  
post #3969 of 4010 Old 01-13-2020, 04:19 PM
Member
 
Join Date: Nov 2019
Posts: 123
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 64 Post(s)
Liked: 19
Having switch otherwise would that cause settings issue or quality?
Bellaseni is offline  
post #3970 of 4010 Old 01-13-2020, 04:33 PM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 2,374
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 1319 Post(s)
Liked: 475
DXVA2 decoding degrades the image NGU very high is just to costly if you want to tone map.

i don't know why sharp is selected intesd of AA for chroma but each to there own.
mightyhuhn is offline  
post #3971 of 4010 Old 01-13-2020, 04:37 PM
Member
 
Join Date: Nov 2019
Posts: 123
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 64 Post(s)
Liked: 19
The thing is this is my first time
So I don’t know where to even start. Now I have checked so
Many settings I am
Hoping I did not
Mess something up inside of nvidia control
Panel
Or elsewhere
Bellaseni is offline  
post #3972 of 4010 Old 01-13-2020, 04:47 PM
AVS Forum Special Member
 
markmon1's Avatar
 
Join Date: Dec 2006
Posts: 6,790
Mentioned: 119 Post(s)
Tagged: 0 Thread(s)
Quoted: 5679 Post(s)
Liked: 3734
Quote:
Originally Posted by noob00224 View Post
To use madvr's DTM where the real nits is required. Screen is Cinegrey 3D which the manufacturer claims is 1.2, but the 5D was supposed to be 1.5 and in testing was 1.0. I don't know if the the 3D is also 33% less as well.

LE: In the near future going to change to a Silver Fire mix which can be customized, and I'll need to know the gain on that one as well.
Why would you assume a piece of paper is exactly a 0 gain/loss baseline? This sounds like an incorrect assumption.

Video: JVC RS4500 135" screen in pure black room no light, htpc nvidia 1080ti.
Audio: Anthem mrx720 running 7.1.4, McIntosh MC-303, MC-152, B&W 802d3 LR, B&W HTM1D3 center, B&W 805d3 surround, B&W 702S2 rear, B&W 706s2 x 4 shelf mounted for atmos, 2 sub arrays both infinite baffle: 4x15 fi audio running on behringer ep4000 + 4x12 fi audio running on 2nd ep4000.
markmon1 is offline  
post #3973 of 4010 Old 01-13-2020, 04:48 PM
AVS Forum Special Member
 
markmon1's Avatar
 
Join Date: Dec 2006
Posts: 6,790
Mentioned: 119 Post(s)
Tagged: 0 Thread(s)
Quoted: 5679 Post(s)
Liked: 3734
Quote:
Originally Posted by noob00224 View Post
Madvr should be installed in the same partition as the OS, preferably in the Program Files folder.
You could add an empty folder ShowRenderSteps in madvr's folder.
There is a restore default settings.bat in madvr's folder as well.
It makes no difference which drive/folder in which madVR is installed unless it's on a removable drive.

Video: JVC RS4500 135" screen in pure black room no light, htpc nvidia 1080ti.
Audio: Anthem mrx720 running 7.1.4, McIntosh MC-303, MC-152, B&W 802d3 LR, B&W HTM1D3 center, B&W 805d3 surround, B&W 702S2 rear, B&W 706s2 x 4 shelf mounted for atmos, 2 sub arrays both infinite baffle: 4x15 fi audio running on behringer ep4000 + 4x12 fi audio running on 2nd ep4000.
markmon1 is offline  
post #3974 of 4010 Old 01-13-2020, 04:52 PM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 2,374
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 1319 Post(s)
Liked: 475
avoid program files and such folder there is a small potential for some small issues because windows handles them differently so read write access can get a bit messy.
mightyhuhn is offline  
post #3975 of 4010 Old 01-13-2020, 05:02 PM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 1,772
Mentioned: 49 Post(s)
Tagged: 0 Thread(s)
Quoted: 952 Post(s)
Liked: 270
Quote:
Originally Posted by markmon1 View Post
Why would you assume a piece of paper is exactly a 0 gain/loss baseline? This sounds like an incorrect assumption.
I assume a white A4 (or larger) paper has a 1.0 gain. Is that incorrect?

Saw it done here:

Quote:
Originally Posted by Dominic Chan View Post
I received the sample kit and measured the gains of each material. The gain measurements are correct in relative terms, but the values are assuming 1.0 for the white paper. The R,G,B percentages are relative to the 1.3 gain raw material, as that's what I calibrated my projector to.

Material; Gain; R,G,B percentages
Walmart Pen Gear 92 White Paper: 1.00 (reference) 101, 100, 101
1.3 Raw Material: 1.11 100, 100, 100 (reference)
1.1 Matte white: 0.99 98, 100, 101
1.0 Matte grey: 0.59 95, 100, 106
0.95 High Contrast Grey: 0.67 101, 100, 98
1.5 Silver Glossy Side 1.03 99, 100, 102
1.5 Silver Matte Side: 0.66 99, 100, 102
1.1 Woven Acoustic WAB: 0.74 98, 100, 101

Not surprisingly, the measured gains are less than the rated values, some significantly so. The worst one is the Silver material - 0.66 vs 1.5 for the matte side which supposedly is the front.

Note also that the matte materials have the same gains over a wide viewing angle, whereas the gains for other materials fall off rapidly off centre (hotspotting):



I have also posted a picture comparing the smoothness of the 1.3 with the 1.1:
https://www.avsforum.com/forum/23-sc...l#post59004682

Quote:
Originally Posted by markmon1 View Post
It makes no difference which drive/folder in which madVR is installed unless it's on a removable drive.
Sorry, it was the other way around, the OSD was having issues when adding empty folders (in the madvr folder) when it was in the Program Files.
Worked when it moved out.
noob00224 is offline  
post #3976 of 4010 Old 01-13-2020, 05:08 PM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 2,374
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 1319 Post(s)
Liked: 475
looks like i guess correctly again.
mightyhuhn is offline  
post #3977 of 4010 Old 01-13-2020, 05:09 PM
AVS Forum Special Member
 
markmon1's Avatar
 
Join Date: Dec 2006
Posts: 6,790
Mentioned: 119 Post(s)
Tagged: 0 Thread(s)
Quoted: 5679 Post(s)
Liked: 3734
Quote:
Originally Posted by SuurbieR View Post
So I have reinstalled MPC-HC and uninstalled madVR, installed again in another location. Still no shown rendering times. Is there any other way to reset madVR to default? When I looked into settings I saw same settings the way I left them.


Quote:
Originally Posted by noob00224 View Post
If the render time is under 16.68ms then it's another problem, refresh rate match, or maybe too slow hdd ( or other).
Have you looked at Custom Modes
https://forum.kodi.tv/showthread.php?tid=259188



@noob00224 : Unless I configure my AVR receiver (Denon X4500H) as a digital monitor under madVR devices, I am not able to insert custom modes for my TV. Also, it can't be my HDD, as I tried to play the files from my SSD as well, same problem..
Run the reset.bat then delete settings.bin. Run something that uses madVR and settings.bin should get recreated.

Video: JVC RS4500 135" screen in pure black room no light, htpc nvidia 1080ti.
Audio: Anthem mrx720 running 7.1.4, McIntosh MC-303, MC-152, B&W 802d3 LR, B&W HTM1D3 center, B&W 805d3 surround, B&W 702S2 rear, B&W 706s2 x 4 shelf mounted for atmos, 2 sub arrays both infinite baffle: 4x15 fi audio running on behringer ep4000 + 4x12 fi audio running on 2nd ep4000.
markmon1 is offline  
post #3978 of 4010 Old 01-13-2020, 05:13 PM
AVS Forum Special Member
 
markmon1's Avatar
 
Join Date: Dec 2006
Posts: 6,790
Mentioned: 119 Post(s)
Tagged: 0 Thread(s)
Quoted: 5679 Post(s)
Liked: 3734
Quote:
Originally Posted by noob00224 View Post
I assume a white A4 (or larger) paper has a 1.0 gain. Is that incorrect?
I wouldn't make any assumptions about this at all. Papers are all different, come in different brightnesses and even printer paper has some level of reflectivity.

Video: JVC RS4500 135" screen in pure black room no light, htpc nvidia 1080ti.
Audio: Anthem mrx720 running 7.1.4, McIntosh MC-303, MC-152, B&W 802d3 LR, B&W HTM1D3 center, B&W 805d3 surround, B&W 702S2 rear, B&W 706s2 x 4 shelf mounted for atmos, 2 sub arrays both infinite baffle: 4x15 fi audio running on behringer ep4000 + 4x12 fi audio running on 2nd ep4000.
markmon1 is offline  
post #3979 of 4010 Old 01-13-2020, 05:37 PM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 1,772
Mentioned: 49 Post(s)
Tagged: 0 Thread(s)
Quoted: 952 Post(s)
Liked: 270
Quote:
Originally Posted by markmon1 View Post
I wouldn't make any assumptions about this at all. Papers are all different, come in different brightnesses and even printer paper has some level of reflectivity.
What would a known/certified/measured 1.0 surface be?
A sample of a measured screen fabric?
noob00224 is offline  
post #3980 of 4010 Old 01-13-2020, 05:46 PM
AVS Forum Special Member
 
Join Date: Jul 2017
Posts: 1,772
Mentioned: 49 Post(s)
Tagged: 0 Thread(s)
Quoted: 952 Post(s)
Liked: 270
Quote:
Originally Posted by SirMaster View Post
MLP is just a HT term people seem to use for main listening position. I guess it applies for audio but why not video too since it's the same position in a home theater. It just means the main seat.

According to the specifications for the i1 Display Pro, the viewing angle is 160 degrees. So you would need to calculate that from the distance.

https://keisan.casio.com/exec/system/1273850202

Use the height and vertex angle option and the vertex angle is 80 and the height is the meter distance.
I can only find this info regarding angle:
Ambient measurement angle: 160 degrees cosine response
http://www.vpixx.com/datasheets/ds_i1display_pro.pdf

Is the 160 degrees vertical or horizontal?

LE: for an 106" (16:9) screen in order for the field of view not to exceed the screen it needs to be placed at 20.7cm/8.14" on the center axis. Isn't this too close?
I'm not sure how the meter will react if it's placed at the MLP (320cm/10.5').

LE2:
Would a velvet tunnel work?

Last edited by noob00224; 01-13-2020 at 10:08 PM.
noob00224 is offline  
post #3981 of 4010 Old 01-13-2020, 05:52 PM
AVS Forum Special Member
 
markmon1's Avatar
 
Join Date: Dec 2006
Posts: 6,790
Mentioned: 119 Post(s)
Tagged: 0 Thread(s)
Quoted: 5679 Post(s)
Liked: 3734
Quote:
Originally Posted by noob00224 View Post
What would a known/certified/measured 1.0 surface be?
A sample of a measured screen fabric?
I'd say the closest reference material would be stewart ST100 studiotek - you can get a free sample from stewart's website in contact section.
noob00224 likes this.

Video: JVC RS4500 135" screen in pure black room no light, htpc nvidia 1080ti.
Audio: Anthem mrx720 running 7.1.4, McIntosh MC-303, MC-152, B&W 802d3 LR, B&W HTM1D3 center, B&W 805d3 surround, B&W 702S2 rear, B&W 706s2 x 4 shelf mounted for atmos, 2 sub arrays both infinite baffle: 4x15 fi audio running on behringer ep4000 + 4x12 fi audio running on 2nd ep4000.
markmon1 is offline  
post #3982 of 4010 Old 01-13-2020, 06:50 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 7,590
Mentioned: 151 Post(s)
Tagged: 0 Thread(s)
Quoted: 5394 Post(s)
Liked: 2135
Quote:
Originally Posted by noob00224 View Post
What would a known/certified/measured 1.0 surface be?
A sample of a measured screen fabric?
The "standard" for 1.0 gain is a magnesium oxide white board, which I do not have. Hence I just use a sheet of white paper, and explicitly stated my assumption so the reader can interpret the posted gains.

Note that the validity of this assumption does not affect the relative gains between different screen materials, as all quoted gains are relative to the same "reference". IOW, if Screen A is 20% brighter than Screen B in the comparison, it is still 20% brighter even if the white paper has a gain of 0.9 instead of 1.0. Nonetheless, I'm quite sure the assumed gain of 1.0 of the white paper is quite close, based on the Silver Ticket Matte White which others have also measured to be about 0.9.

Even though white papers have different "rated brightness", they differ from each other mostly in the presence of UV light, which is converted to visible blue light by the optical brightener in some of the papers. Those papers do not look any brighter under incandescent light (as an example), as there's negligible amount of UV light there.

I replied to the question in this thread as I was quoted, but measuring screen gains is really off-topic here.
noob00224 and SuurbieR like this.

Last edited by Dominic Chan; 01-13-2020 at 07:21 PM.
Dominic Chan is online now  
post #3983 of 4010 Old 01-14-2020, 03:03 AM
Advanced Member
 
Join Date: Feb 2015
Posts: 888
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 537 Post(s)
Liked: 117
is there a way to set different color profiles (gamma, brightness, contrast, saturation, etc.) for hdr and non hdr within madvr? fwiw i'm using madvr with potplayer.

thanks!
aeneas01 is offline  
post #3984 of 4010 Old 01-14-2020, 01:41 PM
Member
 
Join Date: Nov 2019
Posts: 123
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 64 Post(s)
Liked: 19
Any madvr experts willing to remote in to go through some of my settings ? Would greatly appreciate it shoot me a pm?
Bellaseni is offline  
post #3985 of 4010 Old 01-14-2020, 03:52 PM
Member
 
Join Date: Jan 2020
Posts: 15
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 7 Post(s)
Liked: 0
Quote:
Originally Posted by markmon1 View Post
Run the reset.bat then delete settings.bin. Run something that uses madVR and settings.bin should get recreated.
Thnx! This did the trick, deleting settings.bin. With current settings I'm having rendertimes of 20-25 ms for 23fps movies with my current settings.
SuurbieR is offline  
post #3986 of 4010 Old 01-15-2020, 07:04 PM
Member
 
viper008's Avatar
 
Join Date: Nov 2009
Posts: 39
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 11
Quote:
Originally Posted by Postmoderndesign View Post
GTX 1660 SUPER is reported good for 54 fps. So may not be able to handle 60.

Can you provide settings that caused those issues. Also FPS is a strange number to provide. You are supposed to look at render times. At least that is my understanding from reading the set up. Also it makes sense to monitor render times. FPS is not factored in to that.
Not sure who was running those tests and what settings. Using NNEDI 256 perhaps, but can’t comment just yet as I haven’t done those tests.
Might be worth setting up a table of render time vs settings for mid-high and high end GPUs.
viper008 is offline  
post #3987 of 4010 Old 01-15-2020, 07:42 PM
AVS Forum Special Member
 
mightyhuhn's Avatar
 
Join Date: Nov 2013
Posts: 2,374
Mentioned: 12 Post(s)
Tagged: 0 Thread(s)
Quoted: 1319 Post(s)
Liked: 475
nnedi3 is not even part of madVR anymore.

and you can't do such a test GPU are terrible unreliable if it comes to clock speeds and some system go crazy when they are using copyback other don't even care.
you can easily end up with higher rendertimes with lower setting because the GPU just starts to not boost and such.
mightyhuhn is offline  
post #3988 of 4010 Old 01-16-2020, 12:25 AM
AVS Forum Special Member
 
SirMaster's Avatar
 
Join Date: Aug 2009
Location: Wisconsin
Posts: 1,087
Mentioned: 44 Post(s)
Tagged: 0 Thread(s)
Quoted: 922 Post(s)
Liked: 598
Quote:
Originally Posted by viper008 View Post
Can you provide settings that caused those issues. Also FPS is a strange number to provide. You are supposed to look at render times. At least that is my understanding from reading the set up. Also it makes sense to monitor render times. FPS is not factored in to that.
Not sure who was running those tests and what settings. Using NNEDI 256 perhaps, but can’t comment just yet as I haven’t done those tests.
Might be worth setting up a table of render time vs settings for mid-high and high end GPUs.


Technically you can use FPS. It’s the inverse of render time.

If you were getting 23ms render time, that would be equivalent to 1000/23 = 43.48fps.
SirMaster is online now  
post #3989 of 4010 Old 01-16-2020, 03:19 AM
Member
 
viper008's Avatar
 
Join Date: Nov 2009
Posts: 39
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 11
Quote:
Originally Posted by mightyhuhn View Post
nnedi3 is not even part of madVR anymore.

and you can't do such a test GPU are terrible unreliable if it comes to clock speeds and some system go crazy when they are using copyback other don't even care.
you can easily end up with higher rendertimes with lower setting because the GPU just starts to not boost and such.
your response doesn't make sense. you can monitor impact of the settings by looking at the render times

So what settings were you doing to get that impact on FPS?
Was it a 1080p -> 4K upscaling
or
4k 60fps source -> 4k 60fps with chroma upscaling + ? ?
or
something else?
viper008 is offline  
post #3990 of 4010 Old 01-16-2020, 03:20 AM
Member
 
viper008's Avatar
 
Join Date: Nov 2009
Posts: 39
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 11
Quote:
Originally Posted by SirMaster View Post
Technically you can use FPS. It’s the inverse of render time.

If you were getting 23ms render time, that would be equivalent to 1000/23 = 43.48fps.
yes, but we have seen people having issues with source FPS and their display FPS so I would rather keep the focus on the render times that are displayed when using MadVR. That way we know it's not due to a setting issue causing the display to be different to the source
viper008 is offline  
Sponsored Links
Advertisement
 
Reply Home Theater Computers

Tags
htpc build , madvr

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