Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
22nd February 2013, 10:40 | #17641 | Link |
Registered User
Join Date: Aug 2008
Location: the Netherlands
Posts: 851
|
Can you explain a little more about this? Don t understand what you mean exactly. Will I benefit a lot from a high quality chroma upscaling setting when playing 1080p blu-ray on a 1080p display device?
|
22nd February 2013, 10:50 | #17642 | Link | |||
Registered User
Join Date: Dec 2008
Posts: 496
|
Quote:
Quote:
Quote:
Luma upscaling settings should not make a difference if your source resolution = target resolution. Luma downscaling only is active if you´re target resolution is lower than your source resolution, which is not the case very often (e.x. only if you still have a 1680x1050 screen and you have 1080p content). Last edited by iSunrise; 22nd February 2013 at 10:57. |
|||
22nd February 2013, 10:53 | #17643 | Link |
Registered User
Join Date: Jul 2008
Posts: 157
|
Doesn't seem to work very well with interlaced 1080i 25fps TV recordings I have, it introduces a lot of judder. 1080p@ 25fps works great though as does 576i 25fps material.
On a second computer, I can't get it to work without bad judder with a 24fps movie either, but I am not sure if thats because the video card is so low end (Radeon HD5570) or not. What kind of requirements are needed? Oh settings are all default on both computers btw. |
22nd February 2013, 10:57 | #17644 | Link |
Registered User
Join Date: Aug 2008
Location: the Netherlands
Posts: 851
|
thxz for your reply iSunrise. Let me ask the question more simple:
Right now I have a GTX560ti. Would switching to a GTX650ti be an upgrade or downgrade when madVR is the only purpose of the card (and keeping in mind that I ONLY play 1080p on a 1080p display device)? And with bandwidth you mean Bandwidth in GB/s? Because the GTX650ti has 86.4 and the 560ti has 128. But the GFLOPs and Transistors of the 650ti are higher than the GFLOPs and Transistors of the 560ti. Last edited by THX-UltraII; 22nd February 2013 at 11:01. |
22nd February 2013, 11:05 | #17646 | Link | |
Registered User
Join Date: Dec 2008
Posts: 496
|
Quote:
Yes, exactly. The tiny bit higher throughput in GFLOPS is not that important if you only need to upscale chroma and not luma. Last edited by iSunrise; 22nd February 2013 at 11:10. |
|
22nd February 2013, 11:11 | #17648 | Link | |
Registered User
Join Date: Aug 2008
Location: the Netherlands
Posts: 851
|
Quote:
What is you recommendation for the Chroma Upscaling setting when only playing 1080p on a 1080p display? |
|
22nd February 2013, 11:17 | #17650 | Link | |
Registered User
Join Date: Dec 2008
Posts: 496
|
Quote:
There´s no real recommendation, because every algorithm comes with positives/negatives, you need to judge for yourself by testing them. The defaults should be fine. Last edited by iSunrise; 22nd February 2013 at 11:21. |
|
22nd February 2013, 11:23 | #17651 | Link | |
Registered User
Join Date: Aug 2008
Location: the Netherlands
Posts: 851
|
Quote:
The biggest problem is that the lip sync issue can only be seen after a very long playback time without pausing the movie (pausing the movie and play it again fixes the lip sync issue). So testing takes a lot of patience and time. I can only see it very clear after 2 hour movie playback or longer. |
|
22nd February 2013, 12:12 | #17652 | Link | |
Registered User
Join Date: Jul 2008
Posts: 157
|
Quote:
|
|
22nd February 2013, 13:32 | #17653 | Link | |
Registered User
Join Date: Oct 2012
Posts: 7,926
|
Quote:
madvr drops/repeats frames so thinks like that should never happen. what did the madvr osd say how high is the clock deviation. when this only happens with smooth motion make a bug report but i don't see the problem at madvr. |
|
22nd February 2013, 13:47 | #17654 | Link |
Registered User
Join Date: Sep 2012
Location: Turin
Posts: 104
|
Frame interpolation does generate artifacts, the highest the settings, the highest the fluidity, the highest the artifacts are. I do not mind the artifacts, on the other hand I find content with less than 50 or 60 (real or interpolated) fps almost unwatchable... It's really a matter of taste. Give SVP a try. I cannot live without it
__________________
Raven RVZ01 * i7-4790k * 16GB RAM * Zotac GTX 970 4G * SSD 850Evo 500GB * Blu-Ray Burner Slot-In * PSU SFX 80+ Gold 450Watt * Windows 10 64bit * MPCHC+MadVR+SVP * Panasonic 50" VT30 ^^ Last edited by AndreaMG; 22nd February 2013 at 13:57. |
22nd February 2013, 14:13 | #17655 | Link |
Registered User
Join Date: Jan 2013
Posts: 41
|
Hi,
I would avoid any misunderstanding about my knowledge on how the smooth motion functionality works. Can someone please summarize the advantages of using this feature ? I currently use JRMC with ROHQ (MadVR / LAV). On JRMC, there is a similar functionality called "Videoclock". Is that the same thing ? With Videoclock, I can set the output frequence compared with the native video track frequency. Thanks for the clarification. |
22nd February 2013, 15:01 | #17656 | Link |
Registered User
Join Date: Jan 2007
Posts: 530
|
@adhara:
Read here about VideoClock in JRiver. Obviously not the same thing as madVR FRC, though potentially complimentary as is described above regarding Reclock. Simply put, madVR FRC "eliminates" judder caused by framerate/refresh rate mismatch. So, for example, on TVs/monitors without native 24p (?interpolate to 60Hz?), FRC smooths playback at 60Hz. |
22nd February 2013, 15:30 | #17657 | Link | ||
Registered User
Join Date: Jan 2013
Posts: 41
|
Quote:
Thanks. So you suggest to use both videoclock and madVR FRC ? Any contraindication in doing that ? Here in France, the TV rate is 50Hz and mine (Pioneer Kuro) is fully 24 Hz compatible. Is madVR FRC usefull in my own case ? regards. Other point: About the table below: I heard some guys saying that softcubic 80 gives a better result than Jinc 3 or Bicubic 75 or Lanczos for both chroma and image upscaling. Is that correct ? Quote:
To be honest, I don't see much differences between them except for GPU / CPU load For now, I've set chroma and image up to softcubic80 (+AR) and image down to CatmullRom (AR+LL). I only have a HD4000 (1350 MHz) as GPU. Not sure Jinc 3 will work fine w/o a dedicated graphic card. Last edited by adhara; 22nd February 2013 at 15:39. |
||
22nd February 2013, 16:05 | #17658 | Link |
Registered User
Join Date: Jan 2007
Posts: 530
|
I'm currently testing scenarios with VideoClock on and off with FRC on, playing back 24p SD/HD material at 60Hz (my TV doesn't support native 24p). Right now, I don't see a difference with VideoClock on or off, it's working great and beats the heck out of my TV's "conversion" of 24Hz to 60Hz.
Fwiw, I prefer Jinc3 Chroma. For SD playback (on 1080 TV), I prefer Jinc3 w/AR image upscale. YMMV, trust your eyes, but of course, your hardware is likely a performance limitation. I'm currently testing with AMD HD6570. |
22nd February 2013, 18:51 | #17659 | Link | |
Registered User
Join Date: Feb 2002
Location: San Jose, California
Posts: 4,407
|
Quote:
I really like the look of Jinc 3 but, yes, an HD4000 doesn't like Jinc. |
|
22nd February 2013, 19:21 | #17660 | Link | |
Registered User
Join Date: Jan 2013
Posts: 41
|
Quote:
With my HD4000, I can use the "high" preset (bicubic / Lanczos) w/o any problem. Is bicubic better than softcubic ? There is no much topic / forum where guys talk about the differences between MadVR algorithms. I've got a low profile graphic card not yet installed on my passive HTPC (Radeon HD6450). Do you think this card will able me to run Jinc 3 ? thx |
|
Tags |
direct compute, dithering, error diffusion, madvr, ngu, nnedi3, quality, renderer, scaling, uhd upscaling, upsampling |
|
|