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. |
14th April 2013, 20:49 | #18384 | Link | ||||
Kid for Today
Join Date: Aug 2004
Posts: 3,477
|
Quote:
All I know is that the slow response time Hitachi 46" CCFL LCD TV I tried a few months ago appeared smooth in 48Hz(with mVR 0.84/296.10 drivers). I'll roll back to this software combination and see what happens. That OLED pro SONY monitor I mentioned a few weeks ago displays 24Hz@72Hz and 25Hz@75Hz, I'm not sure if that Hitachi TV could have somehow "blurred" frames more efficiently because it was fed 48Hz instead of 24Hz? And I presume that mVR's dithering is done on the actual frames so that shouldn't matter? Ah well I also remember that I really enjoyed LSF on 48Hz DLP because it would sharpen up the motion-blur I'm gonna roll drivers and see what happens.....basically I don't get any tearing and 24p doesn't appear to be internally converted by that Sammy TV, it's just that once in a while judder tests "hiccup"(for lack of a better word)...such as the video file attached to this message(remuxed to MKV with an audio file and with Reclock's judder test on top of it in 24Hz). Quote:
Quote:
Quote:
Yep, I vaguely remember your reply being that it was "planned" and that you would completely change the way the config panel is set....but I don't recall any <1.0 or >1.0 time frame being specified and I believe it to be a feature many people crave at this point Last edited by leeperry; 14th April 2013 at 21:18. |
||||
14th April 2013, 21:21 | #18385 | Link | |
Registered User
Join Date: Dec 2012
Location: Neverland, Brazil
Posts: 169
|
Quote:
__________________
madVR scaling algorithms chart - based on performance x quality | KCP - A (cute) quality-oriented codec pack |
|
14th April 2013, 21:38 | #18386 | Link | |||
Registered User
Join Date: Apr 2009
Posts: 1,019
|
Quote:
Quote:
Typically the sRGB primaries (which are the same as the BT.709 primaries) are used with a 2.2 power curve gamma. Quote:
But CRT has a very low ANSI contrast ratio - about 150:1 with the best models, and 100:1 or less with typical models. This is a result of the "blooming" that you describe. This is why it has always surprised me that people that like CRTs wouldn't give Full Array LED backlit sets a chance. While they also exhibit blooming, it is much more contained (on sets with 100-300 zones) and the zones can be turned off completely, or to an extremely low level (>80,000:1) without crushing shadow detail or requiring external LUT trickery (depends on the implementation, but Sony's is good) and they have in excess of 10,000:1 ANSI in most cases - some sets, in excess of 15,000:1. So they have both higher on-off contrast (dynamic range) and orders of magnitude higher ANSI contrast. The end result is actually a very CRT-like image on my HX900. Traditional LCDs have essentially a fixed contrast ratio. The best sets have a contrast of about 3,000-4,000:1, but this is for both on-off and ANSI contrast. There are other dimming techniques for these LCDs (including "local dimming" with 8-32 zones for edge LED sets) but these are ineffective in my opinion. Plasmas usually have very little difference between their on-off contrast and their ANSI contrast values. The latest Panasonics are around 20,000:1 on-off when calibrated, and 12,500:1 ANSI. This is very good performance, but because the on-off contrast is 20,000:1 the set never truly looks black in a dark room. (Kuros were around 30,000 on-off & 15,000 ANSI) I would much rather take the "infinite" dynamic range of a local dimming LCD and high ANSI vs the lower dynamic range and potentially slightly higher ANSI contrast of a Plasma display. (but they have a number of other issues too) Fortunately OLED is on the way (it should be here in about two or three year, realistically) which should combine both extremely high dynamic range with extremely high ANSI contrast. But this discussion doesn't really belong in the madVR topic. |
|||
14th April 2013, 21:38 | #18387 | Link | ||||
Registered Developer
Join Date: Sep 2006
Posts: 9,140
|
Quote:
Quote:
Quote:
Quote:
Then why don't you use search. I'm really tired of discussing feature requests, or replying to questions about what is on my to do list and what isn't. |
||||
14th April 2013, 21:48 | #18388 | Link |
Registered User
Join Date: Feb 2008
Posts: 45
|
What's up with 1440x1080 HDTV files like this? With Jinc3+AR image upscaling algorithm (and Bicubic75+AR chroma upscaling), the rendering time is
- 1440x1080 --> 1920x1080: ~17.3ms - 1440x1080 --> 1918x1079 (or any similar resolution such as 1900x1069 [1% zoom out]): ~10.4ms I haven't seen such a drastic reduction of the rendering time by a slight change of resolution with any other file. A10-5800K+DDR3-2400+1080MHz iGPU. Last edited by renethx; 14th April 2013 at 21:50. |
14th April 2013, 22:03 | #18389 | Link | |
Registered User
Join Date: Jan 2008
Posts: 589
|
Quote:
You said that when given the choice between smooth motion and native refresh rate, one should choose native refresh rate. I wouldn't be so sure about that. Maybe a good idea would be for us to put feature requests in the bug tracker, so that they can just sit there for some (possibly long) amount of time until you decide to consider them? Last edited by e-t172; 14th April 2013 at 22:25. |
|
14th April 2013, 22:46 | #18391 | Link | ||
Registered User
Join Date: May 2008
Posts: 1,840
|
Quote:
Quote:
__________________
PC: FX-8320 GTS250 HTPC: G1610 GTX650 PotPlayer/MPC-BE LAVFilters MadVR-Bicubic75AR/Lanczos4AR/Lanczos4AR LumaSharpen -Strength0.9-Pattern3-Clamp0.1-OffsetBias2.0 |
||
14th April 2013, 22:48 | #18392 | Link | |
Registered User
Join Date: Dec 2008
Posts: 496
|
Quote:
What would our PC monitors be good for if they didnīt adhere to sRGB as close as possible. Why would I invest money if they didnīt offer me a close resemblance of standards. Certainly not because of their 850:1 contrast, which pales in comparison to dedicated TVs or insanely expensive (broadcast) monitors. I agree that sRGB is certainly not a typical target for film or video use, but thatīs not the point. PS: I bought an iPad 3 simply because of the reason that Apple factory calibrated the first batch of iPad 3 devices to sRGB and they are almost perfectly matching it (a bit too blue, ~7.000 Kelvin). I was also confused by that a bit. I also asked if thereīs any reason to go for a multiple if you also had the choice of native refresh rates that donīt flicker or judder, but unfortunately I didnīt get any answers to that. I guess that madshi meant to say that "in itīs currently implementated form, smooth motion should only be considered if you donīt have a native refresh rate available". I would say that smooth motion works really well, though. Thereīs scenes that look simply incredible with it. Itīs a shame that on very sharp content (like PC encodes) it looks awful, because thereīs a mixture of sharp/unsharp frames that are mixed together. Last edited by iSunrise; 14th April 2013 at 23:21. |
|
14th April 2013, 23:14 | #18393 | Link | ||||||||
Kid for Today
Join Date: Aug 2004
Posts: 3,477
|
Quote:
But you are entirely right that I've got 200Hz BFI enabled on that Sammy TV and that tends to gives a very sharp picture that's as unforgiving as it gets. Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Last edited by leeperry; 14th April 2013 at 23:30. |
||||||||
14th April 2013, 23:44 | #18394 | Link | ||
Registered User
Join Date: Jan 2008
Posts: 589
|
Quote:
Regarding ghosting: maybe that's caused by the LCD pixel response time, which would be more visible when the image changes 60 times per second instead of 24. I'm not sure. Quote:
"I don't think it means what you think it means." An analog projector at 48 Hz (which is basically the same thing as a 48 Hz CRT in this regard) will display 24p much, much better than a 24Hz (or 48Hz) LCD because the former flickers and relies on human persistence of vision to fill in the gaps, which works very well. On the other hand, LCDs do not flicker, the image stays on the screen between refreshes, which defeats persistence of vision and results in non-smooth motion. I'm starting to believe that this unfortunate property of LCDs can be counterbalanced by madVR's smooth motion feature. (but then the question becomes: why didn't the manufacturers think of it?) Last edited by e-t172; 14th April 2013 at 23:48. |
||
14th April 2013, 23:56 | #18395 | Link |
Kid for Today
Join Date: Aug 2004
Posts: 3,477
|
Well, I was using this movie for test purposes this afternoon. It's got a lot of both fast and slow outdoor panning around Olga Kurylenko and it's judderland on the Sammy TV in 24Hz and dead smooth in 96Hz on my CRT.
I do need to find a way to get blending but not as ghosty as FRC...but then again madshi blames my OS when I have a hard time believing that W7 will make FRC bearable to my eyes. Oh well, I remember needing some time before I could find 48Hz smooth on my DLP so I'll see if my brain will ever get used to 24Hz LCD Last edited by leeperry; 14th April 2013 at 23:59. |
15th April 2013, 00:03 | #18396 | Link | |
Registered User
Join Date: Jan 2008
Posts: 589
|
Quote:
As I said, maybe that has to do with LCD pixel response time. That would explain why it looks good to some people but not to others. Maybe it will always look great on a gaming monitor which is optimized for fast response, and it will look bad on some TVs at the other end of the spectrum. But I'm just speculating here. Maybe what I'm saying is bollocks and it's just that some people are more sensitive to it. |
|
15th April 2013, 01:32 | #18397 | Link |
Registered User
Join Date: Jan 2007
Posts: 130
|
Yes I am using overlay. Looks like the backbuffer queue is empty. It alternates between 0-8/8 and 7-8/8. The other queues never go to zero. On videos with no problem, backbuffer q is steady at 7-8/8.
__________________
Win 7 x64, AMD Phenom II X6 1075T, nvidia 430GT, mpc-be x86 |
15th April 2013, 02:16 | #18398 | Link | |
Registered User
Join Date: Dec 2008
Posts: 496
|
Itīs probably a mix of response time and panel type. The Eizo I use also has an H-IPS with very good response and my experiences with smooth motion seem to match yours. I would say that I am fairly happy with itīs current implementation. It looks very close to the native refresh that I am able to achieve with the Eizo. For scenes where thereīs only one or only a few solid objects against a more or less solid background (which arenīt that common when watching real-world content) it looks incredibly good, so good that I canīt see any differences at all to native playback.
Hereīs two samples Iīve found that look perfect with smooth motion: http://www.mediafire.com/?5xccccrs7109mjn http://www.mediafire.com/?1zos93buokb981r Quote:
|
|
15th April 2013, 02:51 | #18399 | Link | |
Registered User
Join Date: Dec 2012
Location: Neverland, Brazil
Posts: 169
|
Quote:
__________________
madVR scaling algorithms chart - based on performance x quality | KCP - A (cute) quality-oriented codec pack |
|
15th April 2013, 03:11 | #18400 | Link | |
Registered User
Join Date: Dec 2008
Posts: 496
|
Quote:
After that, you can also check with test patterns from e.x. AVS HD within madVR if your black levels are as they should be. If you need an adjustment you currently can either adjust gamma ("enable gamma processing" and set the target gamma" a bit (lower value = brighter, higher value = darker) or use the brightness slider under the "color & gamma" tab. You should stay away from the "saturation", "contrast" and "hue" slider if possible. Thatīs as far as youīre able to go without actually doing any calibration yourself. Depending on the quality and native gamma of your panel and the careness of the implementation of the sRGB mode on your monitor, the results can vary though. It should however look a lot better than non-managed. Note that you wonīt see any difference when just playing with the calibration values themselves. They are however important so that madVR "knows" your settings for appropriate conversions to take place. Last edited by iSunrise; 15th April 2013 at 03:26. |
|
Tags |
direct compute, dithering, error diffusion, madvr, ngu, nnedi3, quality, renderer, scaling, uhd upscaling, upsampling |
|
|