madVR - high quality video renderer (GPU assisted)

[xabregas]

I dont really care about sharpness loss as i can do sharp with ffsdhow/ ffsdshow raw if i wish or with driver settings, the problem was the flickering and knowing if i need to use reclock or not. So reclock isnt needed at all when smooth motion is on and ive just unchecked the trade quality for performance option. Anyway just the smooth in quickly action scenes and fight scenes is awesome. There are movies i saw without smooth motion and when i see them now for me its completly new. Some figh scenes i now can clearly see what happens, AMAZING :eek:

But still 1 option in trade quality for performance is checked (store custom pixel shader results in 16bit buffer. instead of 32bit Wonder what it does :sly:

I have an ati HD7770 and im in windows 8 x64

TIA

[DragonQ]

madshi, if I play a 1080i/30 file which is really 1080p/24 and then turn on IVTC, it plays back properly at 24 fps. However, smooth motion doesn't enable (60 Hz monitor) and "movie" is still set to 29.970 fps. A bug?

[e-t172]

madshi, if I play a 1080i/30 file which is really 1080p/24 and then turn on IVTC, it plays back properly at 24 fps. However, smooth motion doesn't enable (60 Hz monitor) and "movie" is still set to 29.970 fps. A bug?

Smooth motion with 1080i30 IVTC looks perfectly fine on my system.

[dansrfe]

Is BFI planned for any future madVR build? Just curious...

[DragonQ]

Smooth motion with 1080i30 IVTC looks perfectly fine on my system.
With a 60 Hz monitor and "only if there would be motion judder without it" ticked?

[leeperry]

Talking about BFI, this link (http://hardforum.com/showthread.php?t=1742666) is rather interesting :)

Of course it dims the display and will make light bleeding/clouding even more noticeable as you'll be forced to increase the backlight brightness in order to compensate, but on Sammy TV's BFI really does wonders to my eyes(that just can't bear PWM dimming (https://www.youtube.com/watch?v=bKhDS7CHDPk) and have been watching CRT's since forever).

I've spent a few hours playing around with several 2013 Sammy TV's and as much as the new 5 serie provides BFI & DNIE(serie 4 only provides BFI), in the 6 serie you can only enable BFI if you have FI enabled, duh! That sure wasn't the case in the 2012 models, impressive failtastic choice here...as I had my eyes on the 32F6800, in order to increase my chances of winning the panel lottery :mad:

Anyway UHD is coming and as much as the current HDMI specs would only allow 2160p up to 30Hz, several of them seem to allow 1080p@120Hz (http://www.blurbusters.com/seiki-39-4k-hdtv-699-also-supports-1080p120hz/) so I believe BFI from the PC would make sense.

Either way, even at 100% brightness all the PC monitors I've seen hurt my eyes so I dunno wth is up.....OTOH even the cheapest TV's don't, so it might indeed be a problem with PWM after all :confused:

PS: ouuuuh 500Hz Native Refresh Rate — No Interpolation! (http://www.blurbusters.com/500hz-native-refresh-rate-no-interpolation/)

[mdrejhon]

Here's a web-based demonstration of black frame insertion:

www.testufo.com/#test=blackframes (http://www.testufo.com/#test=blackframes)

You need a very recent web browser for these precision animations. I recommend Chrome when viewing the above link. (FireFox won't play smoothly until Version 24+ pre-beta)
Try out the other UFO motion tests selectable at top-right corner, including the Optical Illusion Demo as well as 30fps-vs-60fps Demo.

Note: I'm the one who runs the Blur Busters Blog.

[dansrfe]

Here's a web-based demonstration of black frame insertion:

www.testufo.com/#test=blackframes (http://www.testufo.com/#test=blackframes)

I'm running chrome 30 and I see "poor performance" when running the BFI test above. ~32 fps, ~32Hz, ~30 PPF, ~960 PPS. I'm running no other tabs, browsers or other applications other than a code editor. i7, GTX 670, 16GB RAM.

Maybe you need to use requestAnimationFrame? Or are you already? https://developer.mozilla.org/en-US/docs/Web/API/window.requestAnimationFrame

[Mark Rejhon]

(EDIT: It appears I somehow have an old duplicate account created many years ago, I'll need to ask a mod to remove 'mdrejhon')

-- It already is. Launch chrome://gpu and make sure Chrome is GPU accelerated in all categories. If you've made modifications to your high-precision timer, this can interfere. If you are using a SLI/multimonitor system, there's a Chrome issue regarding that (I've forgotten the Chromium bug #). Alternatively, try FireFox 24+ Aurora (http://www.mozilla.org/en-US/firefox/aurora/).
-- You can try IE10, if you're only using a 60Hz monitor. It also synchronizes to VSYNC too.
-- I just prefer Chrome since it supports 120fps@120Hz on 120Hz monitors. My weblogs shows several 120Hz and 144Hz users playing these animations at full framerates (e.g. 120fps@120Hz).
-- Supported browsers (if GPU accelerated): FF24+, Opera15+, Chrome18+, IE10+, and Mac's Safari6+. The TestUFO site performs at 60fps on iPads too (iPad 2 and newer)

[turbojet]

Thanks for the tests, they work fine here with Chrome 28 and many tabs open.

I'm not sure I prefer 30+BFI over 30fps, bfi is higher clarity but the pulsating gave me a headache within a minute. Maybe it takes some getting used to but currently prefer the occasional blended frame with frc to it.

Users that are asking for bfi, is it to replace or complement frc?

[Mark Rejhon]

Ah yes -- www.testufo.com/#test=blackframes (http://www.testufo.com/#test=blackframes) is just a demo of software-based BFI, which flickers like a 30Hz CRT to show how flicker can reduce motion blur. I agree, 30Hz flicker is annoying! But that web animation is actually quite useful to help people understand how and why BFI reduces motion blur. If you have a 120Hz monitor, you can also do software-based BFI to have blur-free 60fps@120Hz, and the 60Hz flicker is far less annoying.

Ideally, hardware-based BFI at 120Hz or 240Hz is even better. That eliminates the flicker annoyance for most. Some 120Hz monitors (http://www.blurbusters.com/faq/120hz-monitors) with LightBoost actually already do this, by turning off the backlight while waiting for LCD pixels to finish refreshing (high speed video (http://www.youtube.com/watch?v=hD5gjAs1A2s)), and strobing between the refreshes. They flicker like a 120Hz CRT, but that allows the CRT motion clarity effect during 120fps@120Hz videogames, as well as zero-motion-blur playback of 120fps videos (e.g. pre-interpolated first with SmoothVideo, etc).

P.S. Another fun animation is the optical illusion at www.testufo.com/#test=eyetracking (http://www.testufo.com/#test=eyetracking)

[dansrfe]

I just tested my browser's fps capability on this page: http://ie.microsoft.com/testdrive/performance/fishIETank/default.html (with 1000 fish) and it renders at 60fps. Not sure why the BFI test says poor performance and stops at ~30fps...

http://imageshack.us/a/img35/617/dgm.png

[sneaker_ger]

So, the first official/public XySubFilter beta has been released and since it is - to my knowledge - the only filter making use of a new (open) custom subtitle interface supported by madVR, it might be of interest to some:
http://code.google.com/p/xy-vsfilter/wiki/ReleaseNotes?tm=6 (changelog/description)
http://code.google.com/p/xy-vsfilter/wiki/Downloads?tm=2 (download, scroll down to XySubFilter)

Unlike (xy-)vsfilter, it renders subtitles at the target resolution instead of the video resolution, so subtitle quality can get enhanced greatly. Since most video content is 4:2:0 this even goes for 1080p videos on a 1080p screen, although the difference is not that big as for in this example SD videos. The difference to the internal MPC-HC subtitle filter is the new interface and huge (performance) improvements, bug fixes and maybe some features. Since it does not sit between the video decoder and madVR it is compatible with DXVA2 (native).

Any questions that are not directly related to the madVR<>XySubFilter interaction should be posed in the following thread to not stray off the topic of this thread too far:
xy-VSFilter Project (High Performance VSFilter Compatible Subtitle Filters) (http://forum.doom9.org/showthread.php?t=168282)

[Soukyuu]

Well, since madvr is currently the only renderer supporting XySubFilter, I imagine most questions will be posted here anyway.
Here's the first one: I have the latest mpc-hc+madvr+xysubfilter beta. The .ass subtitles are rendered in an opaque black box instead of blending it unto the video, does anyone else have this problem?

[xabregas]

So i use MPC internal subtitle filter. Why should i change to xyvsfilter?

My problem with vsfilter is that i always need to extend to 16:9 to put subtitles in the black bars when watching 2.35 movies and i always feel the filter mess up the PQ a bit :sly:

[kasper93]

@sneaker_ger, please make new thread for it. No need to spam in madVR thread.

[ryrynz]

@sneaker_ger, please make new thread for it. No need to spam in madVR thread.

It's most certainly a useful post and is somewhat related. He was aware it was not the place for discussion and made the appropriate remark.
Personally I think Cyberbeing should make a new thread for it here but there is one over at the CCCP forum (http://www.cccp-project.net/forums/index.php?PHPSESSID=9dg2q7k7jq2272c7gb7msvqdg3&topic=5976.225) where I suggest people direct their questions and issues right now.

[madshi]

Now my problem is i have 60HZ screen with no 24p option and i have 23,976 and 25 FPS material. Do i need to use reclock pal speed down when watching 25 FPS with madvr smooth motion on??

I tested both and it gave me flickering with reclock and without reclock minor ghosting, but more than i have with 23,976 material :(
Some people are reporting ghosting. I don't know why, I don't see it myself. Either I'm just not as perceptive to this kind of artifacts, or maybe your display uses a different refresh rate than 60Hz. E.g. if your display internally uses 70Hz, then it will probably repeat some frames to get from 60Hz to 70Hz. If those repeated frames happen to be madVR blended frames, you will see obvious ghosting. But I don't really know if that's what's happening on your PC. Can your display handle 72Hz in native display resolution? If so, try that.

Reclock is not needed for smooth motion to work correctly.

But still 1 option in trade quality for performance is checked (store custom pixel shader results in 16bit buffer. instead of 32bit Wonder what it does
If your PC is fast enough, uncheck all those "trade quality for performance" options.

madshi, if I play a 1080i/30 file which is really 1080p/24 and then turn on IVTC, it plays back properly at 24 fps. However, smooth motion doesn't enable (60 Hz monitor) and "movie" is still set to 29.970 fps. A bug?
Do you happen to use native DXVA decoding? IVTC doesn't work in that case.

Is BFI planned for any future madVR build? Just curious...
Was discussed before. You should be able to find it via search.

So, the first official/public XySubFilter beta has been released and since it is - to my knowledge - the only filter making use of madVR's new subtitle interface, it might be of interest to some
There will be an official thread for it here on doom9 soon. FWIW, the new subtitle interface is not "mine", although I played a role defining it and the header is hosted on my server. Still, it's a BSD licensed open specification and can be used by anyone who's interested. An MPC-HC dev already hinted that maybe they could use it for the MPC-HC internal renderers, too, but that's far from a "done deal". In any case, it's not "madVR's new subtitle interface", but just the "new subtitle interface" and madVR is one of the first who supports it. Just to clarify the situation...

Here's the first one: I have the latest mpc-hc+madvr+xysubfilter beta. The .ass subtitles are rendered in an opaque black box instead of blending it unto the video, does anyone else have this problem?
The way the new subtitle interface works, madVR needs the GPU to perform subtitle alpha blending - and in 16bit. Unfortunately some older GPUs only support alpha blending in 8bit. For such GPUs you're likely to see an opaque black box instead of smoothly blended subtitles. I'm sorry, but at this point I don't see how I could fix it. It's a hardware limitation of your GPU and the only solution will probably be to upgrade your GPU to a newer model. I believe all newer GPUs can do alpha blending in 16bit without any problems. At least my HD4000 can.

Maybe I'll find a workaround in the long run. But in the short run I don't think I can solve this for old GPUs like yours... :(

So i use MPC internal subtitle filter. Why should i change to xyvsfilter?
You may want to change to XySubFilter, not to xy-vsfilter. XySubFilter allows you to render subtitles in the same quality as MPC's internal subtitle filter - but with less CPU consumption, with correct colors and with correctly scaled 3D effects. The MPC internal subtitle filter is slower, shows incorrect colors and has problems scaling certain ASS effects correctly.

[Mark Rejhon]

I realized I neglected to *directly* answer a very important question about video renderers that add BOTH interpolation and BFI:Users that are asking for bfi, is it to replace or complement frc?Either or both.
Some existing HDTV's (http://www.blurbusters.com/faq/existing-scanning-backlights/) do this already by combining scanning backlight (equivalent of hardware BFI) and interpolation.

The problem is that to do both in software, you really need a very high native refresh rate:
(1) 60fps -> 120fps (via interpolation) -> 240fps (via black frames).
(2) 60fps -> 240fps (via interpolation) -> 960fps (via black frames). This is how some expensive Sony "960Hz TVs" work.
This can only be done in hardware, and cannot be done over a video cable. To do this sort of stuff in software, you need high native refresh rate. Usually combining both BFI and interpolation is something mainly done in hardware inside a 240Hz HDTV.

The modern fast-refresh monitors are the 120Hz and 144Hz computer monitors. The 120Hz and 144Hz headroom gives you enough headroom to do software-based BFI, and gives you an opportunity to combine interpolation and BFI, or do some very interesting tricks:
(1) For example, you can interpolate a movie 24fps -> 72fps (interpolation) -> 144fps (BFI) -> 144Hz monitor
(2) Or you can do 30fps -> 60fps (interpolation) -> 120fps (BFI) -> 120Hz monitor

BFI can also be used by itself, too.
(1) You can do 60fps video -> 120fps (BFI) -> 120Hz monitor
(2) Or simulate an old double-strobe flicker movie projector via 24fps -> 48fps (duplicate frame) -> 144fps (BFI at 2:1 ratio) -> 144Hz monitor

There are a lot of great uses for BFI, but software-based BFI is mainly useful with 120Hz monitors because otherwise it flickers too much. I do, however, recommend adding this feature at the video renderer levels (e.g. ffdshow, madVR, etc) and adding a warning that BFI is mainly useful for 120Hz computer monitors, that it requires very accurate synchronization with the monitor (fast system, good performance), and that it can severely darken the image. But that it can be useful when motion blur reduction is critical. It definitely noticeably reduces motion blur and BFI looks much better on 120Hz monitors, but terrible on 60Hz monitors.

[DragonQ]

Do you happen to use native DXVA decoding? IVTC doesn't work in that case.
No, software. Stepping through frames shows IVTC to 24p is working properly (no repeated frames or combing) but smooth motion doesn't turn on unless I force it in the settings.

[madshi]

If you have a 120Hz monitor, you can also do software-based BFI to have blur-free 60fps@120Hz, and the 60Hz flicker is far less annoying.
Ideally 240Hz would be great for a software BFI implementation, but HDMI can't handle that... :( Even 120Hz is not supported by most almost all TVs out there. Some computer monitors can do 120Hz, but who wants to watch movies on a computer monitor? Because of these limitations I'm not really eager to look into BFI at the moment. I don't want to spend time on something which only works on computer monitors...

Ideally, hardware-based BFI at 120Hz or 240Hz is even better. That eliminates the flicker annoyance for most. Some 120Hz monitors (http://www.blurbusters.com/faq/120hz-monitors) with LightBoost actually already do this, by turning off the backlight while waiting for LCD pixels to finish refreshing (high speed video (http://www.youtube.com/watch?v=hD5gjAs1A2s)), and strobing between the refreshes. They flicker like a 120Hz CRT, but that allows the CRT motion clarity effect during 120fps@120Hz videogames, as well as zero-motion-blur playback of 120fps videos (e.g. pre-interpolated first with SmoothVideo, etc).
Turning the backlight off while waiting for LCD pixels to finish refreshing is something you can't really do with a software based solution. So it's a good thing having that in hardware. However, I think probably the hardware BFI implementation in many displays is far from optimal and a good software implementation could potentially improve on that. If only TVs would support high refresh rates via HDMI. <sigh>

No, software. Stepping through frames shows IVTC to 24p is working properly (no repeated frames or combing) but smooth motion doesn't turn on unless I force it in the settings.
Have you added a bug entry to the bug tracker about this? If so, there's no need to discuss it here in the forum any further. No need to post a link to your tracker issue, either. I'll get to that when I find the time...

[Soukyuu]

The way the new subtitle interface works, madVR needs the GPU to perform subtitle alpha blending - and in 16bit. Unfortunately some older GPUs only support alpha blending in 8bit. For such GPUs you're likely to see an opaque black box instead of smoothly blended subtitles. I'm sorry, but at this point I don't see how I could fix it. It's a hardware limitation of your GPU and the only solution will probably be to upgrade your GPU to a newer model. I believe all newer GPUs can do alpha blending in 16bit without any problems. At least my HD4000 can.

Maybe I'll find a workaround in the long run. But in the short run I don't think I can solve this for old GPUs like yours... :(Aww man. Is my 260GTX really that old? I'm still mostly satisfied with its performance, but I guess it was just a matter of time hardware compatibility would become a problem =/

[madshi]

To be honest, I've no clue about which generation is 16bit alpha blend capable and which is not. That's not a spec that's advertized anywhere. I don't think there's even a "CAP" bit for that in Direct3D, so I can't even ask Direct3D or the GPU itself if it can do it. We'll have to collect information from users to see which GPU generations exactly support XySubFilter+madVR and which don't...

[DragonQ]

Have you added a bug entry to the bug tracker about this? If so, there's no need to discuss it here in the forum any further. No need to post a link to your tracker issue, either. I'll get to that when I find the time...
Will do, just wanted to see if it was already known.

[madshi]

Aww man. Is my 260GTX really that old? I'm still mostly satisfied with its performance, but I guess it was just a matter of time hardware compatibility would become a problem =/
Just tried with my old NVidia 9400 mainboard. I can reproduce the problem there. *However*, by enabling the option "use 10bit image buffer instead of 16bit" the problem goes away. So that should be a good workaround for older GPUs.

[Mark Rejhon]

I'm the operator of the Blur Busters Blog, and creator of TestUFO motion tests, so I've become an expert in motion blur...
First: I agree, low priority issue for now, but this is an early canary warning that this is going to be slowly a more important feature in the future.
However, I would like to inform you of a few things:

Ideally 240Hz would be great for a software BFI implementation, but HDMI can't handle that... :( Even 120Hz is not supported by most almost all TVs out there.Fortunately, that is gradually becoming easier for 2013-model HDTV's. See HDTV Refresh Rate Overclocking HOWTO: True 120Hz from PC to TV (http://www.blurbusters.com/faq/120hz-pc-to-tv). Several Panasonic, Sony, Vizio models worked. Some TV's such as the new SEIKI 4K HDTV's (4K@30Hz) actually natively support 120Hz@1080p so on those models, it's no longer overclocking. Did you hear about the new $699 SEIKI 39" 4K HDTV? The manufacturer actually officially supports 120Hz from a computer on this.

Some computer monitors can do 120Hz, but who wants to watch movies on a computer monitor? Because of these limitations I'm not really eager to look into BFI at the moment. I don't want to spend time on something which only works on computer monitors...Agreed, but see above.

Turning the backlight off while waiting for LCD pixels to finish refreshing is something you can't really do with a software based solution. So it's a good thing having that in hardware.You can combine hardware BFI (black frame equivalent by backlight) and software BFI (black frame as actual inserted blank frame). That's how a custom version of MAME (http://www.blurbusters.com/mame) does it. This is because LightBoost only strobes at 100-120Hz, it cannot strobe at 60Hz, so software-BFI is needed in addition.

Software based BFI (60fps on 120Hz) would do this on normal 120Hz LCD monitors:
--> 8.3ms visible:8.3ms dark == 50% reduction of motion blur

However for doing 60fps @ 120Hz software BFI on a LightBoost display, something rather interesting happens: less motion blur on LCD than plasma (LCD colors are still terrible, though)
LightBoost flashes the backlight for as short as 1.4ms, so over two frames (one 8.3ms visible, one 8.3ms black, total 16.7ms cycle) you are now are getting:
--> 1.4ms visible + 6.9ms backlight off + 1.4ms black frame + 6.9ms backlight off

So effectively, you've combined hardware-based BFI (simulted via backlight) and software-based BFI (actual black frame), you are now getting:
--> 1.4ms visible + 15.3ms black frame (combined hardware+software)

This is a 92% reduction in motion blur over a 60Hz LCD (1.4ms of sample-and-hold blur, instead of 16.7ms of sample-and-hold blur) (web animation of sample-and-hold blur (http://www.testufo.com/#test=eyetracking)). Which produces CRT-quality motion of 60Hz sources on LightBoost LCD displays; that's less motion blur than plasma displays. This is since plasma displays have about 5ms of motion blur, due to red/green phosphor decay, and plasma motion blur becomes noticeable during very fast pans (e.g. when viewing TestUFO: Moving Photo test (http://www.testufo.com/#test=photo&pps=1440) at 1440pixels/sec) while this Moving Photo test remains crystal sharp on a LightBoost display, especially at LightBoost=10%. I should point out that at 1440 pixels/sec, 1ms of sample-and-hold motion blur equals 1.4 pixels of motion blur. So 5ms does add up to quite a lot of motion blur during fast high-def pans where there is no source-based motion blur (e.g. soft focus, long exposure, overcompression). For 1080p 60fps high-bitrate video (especially 4K60fps downconverted to 1080p@60fps, to create ultra-sharp high-contrast edges), it is possible to begin to notice 1ms differences in motion blur during fast screen-width-per-second panning (1920 pixels/sec), because that's currently adding 1/1000th of 1920 = 2 pixels of motion blur. Very subtle. But human eyes can see 2 pixel differences at 1080p, if you're not too far from the display. So, it's in favour to expand the ratio of blackness to visible frame, shave off the milliseconds of visible frame, and that's where combining hardware BFI with software BFI can come in, to bypass a hardware BFI limitation (e.g. LightBoost not functioning at 60Hz).

The 92% reduction of blur is also confirmed via this graph too, during motion measurement tests (see graph (http://www.blurbusters.com/faq/60vs120vslb/#graph), as well as various testimonials (http://www.blurbusters.com/zero-motion-blur/testimonials/)). However, I agree, most people won't be watching movies on a computer monitor, but it bears worth pointing out that LightBoost computer monitors have recently leapfrogged over televisions in terms of achievable motion resolution, even though they have terrible color quality compared to televisions.

However, I think probably the hardware BFI implementation in many displays is far from optimal and a good software implementation could potentially improve on that.That CAN be true, but properly done hardware BFI (especially backlight assisted, not just via panel) can generally be superior to software BFI:

(1) You don't need to rapidly switch the LCD pixel color value from on/off/on/off. This can reduce contrast severely. Hardware BFI assisted via backlight method (rather than LCD panel method) doesn't necessarily need to turn on/off LCD pixels or even completely; it just turns on/off the backlight at precisely synchronized intervals (or segments of backlight at a time, as in scanning backlights) Although terminology should not be used interchangeably, BFI/scanning backlights/strobe backlights all essentially, at the human eye level, roughly equivalent -- these approaches simply adds a dark period between refreshes -- to reduce the sample and hold effect (animation demo of sample and hold effect (http://www.testufo.com/#test=eyetracking) -- this animation looks very different on CRT than LCD).
(2) BFI at the LCD panel level (the only method available to software) also potentially causes some nasty inversion artifacts on some panels, due to the interference with the positive/negative voltages (lagom pixel-walk (http://www.lagom.nl/lcd-test/inversion.php), techmind inversion explanation (http://www.techmind.org/lcd/index.html#inversion)). So fine-single-pixel checkerboard texture patterns can appear during BFI.
(3) Some LED backlights are designed to pulse brighter when on, so you can compensate for BFI dimness by using brighter backlight flashes to compensate.
Doing hardware-based BFI can avoid these problem. The hardware-based BFI avoids that by using the backlight instead, preventing the need for LCD pixels to go back-and-fourth.

If only TVs would support high refresh rates via HDMI. <sigh>Good news. This is changing already. Some of them now do (http://www.blurbusters.com/faq/120hz-pc-to-tv), and sometimes even semi-officially (e.g. SEIKI 4K).

I agree, low priority issue for now, but this message is an early canary warning that this is going to be slowly a more important feature in the future.

[Soukyuu]

Just tried with my old NVidia 9400 mainboard. I can reproduce the problem there. *However*, by enabling the option "use 10bit image buffer instead of 16bit" the problem goes away. So that should be a good workaround for older GPUs.
Yes, that fixes it for me as well. A friend with a 560 doesn't have this issues, so I guess all nVidia GPUs 400 series and up are not affected.

I couldn't see any difference in the image between having 10 and 16Bit buffer, what does it really affect?

[madshi]

Fortunately, that is gradually becoming easier for 2013-model HDTV's. See HDTV Refresh Rate Overclocking HOWTO: True 120Hz from PC to TV (http://www.blurbusters.com/faq/120hz-pc-to-tv). Several Panasonic, Sony, Vizio models worked. Some TV's such as the new SEIKI 4K HDTV's (4K@30Hz) actually natively support 120Hz@1080p so on those models, it's no longer overclocking. Did you hear about the new $699 SEIKI 39" 4K HDTV? The manufacturer actually officially supports 120Hz from a computer on this.
I'm not a friend of overclocking. It means components are probably running at a higher temperature. Which can be just fine in the short run, but could result in hardware failure in the long run. Anyway, it's good to hear that things are improving and that some TVs now officially support 1080p120. Didn't hear about that SEIKI yet, but I'm in Europe, and that SEIKI is probably not available here, I would guess...

hardware BFI can generally be superior to software BFI because:
(1) You don't need to rapidly switch the LCD pixel color value from on/off/on/off. This can reduce contrast severely. Hardware BFI via backlight method (rather than LCD panel method) doesn't need to turn on/off LCD pixels; it just turns on/off the backlight (or segments of backlight at a time, as in scanning backlights) Although terminology should not be used interchangeably, BFI/scanning backlights/strobe backlights all essentially, at the human eye level, roughly equivalent -- these approaches simply adds a dark period between refreshes -- to reduce the sample and hold effect (animation demo of sample and hold effect (http://www.testufo.com/#test=eyetracking) -- this looks very different on CRT than LCD).
(2) BFI at the LCD panel level (the only method available to software) also causes some nasty inversion artifacts, due to the interference with the positive/negative frames (lagom pixel-walk (http://www.lagom.nl/lcd-test/inversion.php), techmind inversion explanation (http://www.techmind.org/lcd/index.html#inversion)). So fine-single-pixel checkerboard texture patterns can appear during BFI.
(3) Some LED backlights are designed to pulse brighter when on, so you can compensate for BFI dimness by using brighter backlight flashes to compensate.
Doing hardware-based BFI can avoid these problem. The hardware-based BFI avoids that by using the backlight instead, preventing the need for LCD pixels to go back-and-fourth (LCD is slow, backlight is faster).
Hopefully OLED takes care of all these problems. Hopefully they'll get lifetime and image retention issues under control. Then BFI for OLED could be a nice solution.

I agree, low priority issue for now, but this message is an early canary warning that this is going to be slowly a more important feature in the future.
I've had BFI on my "long term" list since I started developing madVR. However, I'd need a 120Hz capable display to actually develop something, and I think I'm going to wait for OLED.

I couldn't see any difference in the image between having 10 and 16Bit buffer, what does it really affect?
Well, it's quite a bit lower accuracy. Might eventually result in slightly less smooth gradiants...

-------

So here's the new official xy-VSFilter / XySubFilter thread:

http://forum.doom9.org/showthread.php?t=168282

Please post subtitle specific questions there, I mean things like incorrectly formatted subtitles etcs. If you have subtitle questions that are more madVR related than XySubFilter, you can post them here instead, e.g. if you run into GPU performance problems with XySubFilter or things like that...

[leeperry]

I'm the operator of the Blur Busters Blog
Hi Mark, good to see you around :cool:

Apparently Sammy TV's don't do full frame BFI, just like a CRT, more like half/quarter vertical height or so....anyway yes, by the time mVR reaches 1.0 we will see whether there are more HDMI2/DP 120Hz+ displays on the market.

[cyberbeing]

I've made an official thread for the xy-VSFilter Project, which I'd encourage everybody to use for any XySubFilter discussions not directly related to madVR:

xy-VSFilter Project (High Performance VSFilter Compatible Subtitle Filters) (http://forum.doom9.org/showthread.php?t=168282)

[cyberbeing]

Just tried with my old NVidia 9400 mainboard. I can reproduce the problem there. *However*, by enabling the option "use 10bit image buffer instead of 16bit" the problem goes away. So that should be a good workaround for older GPUs.

Can you detect such GPUs and do this automatically?

[xabregas]

Some people are reporting ghosting. I don't know why, I don't see it myself. Either I'm just not as perceptive to this kind of artifacts, or maybe your display uses a different refresh rate than 60Hz. E.g. if your display internally uses 70Hz, then it will probably repeat some frames to get from 60Hz to 70Hz. If those repeated frames happen to be madVR blended frames, you will see obvious ghosting. But I don't really know if that's what's happening on your PC. Can your display handle 72Hz in native display resolution? If so, try that.

Reclock is not needed for smooth motion to work correctly.


If your PC is fast enough, uncheck all those "trade quality for performance" options.

You may want to change to XySubFilter, not to xy-vsfilter. XySubFilter allows you to render subtitles in the same quality as MPC's internal subtitle filter - but with less CPU consumption, with correct colors and with correctly scaled 3D effects. The MPC internal subtitle filter is slower, shows incorrect colors and has problems scaling certain ASS effects correctly.

Thanks for the answer madshi, i was testing it without catalyst crap installed and i had several ghosting, but i installed CCC again :sly: and disabled all the crap settings except deinterlancing method and pulldown detection. Now i have almost zero ghosting with smooth motion on.

As for the refresh rate, not sure whats the correct refresh rate of my screen but CTRL+ J gives me 60.0014HZ :eek: on my sony TV and 59,9475HZ on my DELL desktop monitor :angry:

Not sure what this means, but i have no ghosting whatsoever on my desktop monitor and on my sony tV i have :confused:

i`ve just installed xysub but i dont know where is the option to put subtitles in the botton black bar on 2.35 movies :(

TIA

[Soukyuu]

Currently, madvr sets both the decoding and subtitle queue via cpu queue parameter. Could we have those separated in the future (if it makes sense internally)?

edit: About the only reason I'd want to have it is for more responsive jumping, since the decoding queue never falls below 63/64 while subtitle queue does drop to 11/64 on some heavily styled subs (and that's on a 4GHz X4 970BE cpu...)

[dansrfe]

What exactly is the goal in setting queues? Should we bring it as low as possible or as high as possible? Does increasing the queue size add lag to initial render steps and/or skipping around the video? How do we know what the optimal queue sizes are for the machine we use?

[sneaker_ger]

There will be an official thread for it here on doom9 soon. FWIW, the new subtitle interface is not "mine", although I played a role defining it and the header is hosted on my server. Still, it's a BSD licensed open specification and can be used by anyone who's interested. An MPC-HC dev already hinted that maybe they could use it for the MPC-HC internal renderers, too, but that's far from a "done deal". In any case, it's not "madVR's new subtitle interface", but just the "new subtitle interface" and madVR is one of the first who supports it. Just to clarify the situation...

Yeah, sorry, my wording was indeed misleading.
I didn't want to open a new thread myself because I was hoping for an official one. It's better if a team member does it, so he can always update the start post with fresh info straight from the source.

[Soukyuu]

What exactly is the goal in setting queues? Should we bring it as low as possible or as high as possible? Does increasing the queue size add lag to initial render steps and/or skipping around the video? How do we know what the optimal queue sizes are for the machine we use?
The lower the queue, the faster it can be filled (=less delay on seeking, especially with "delay playback until queue full" option), but if your PC lacks processing power, you will get dropped frames once it hits a demanding place.

So to find out if your queue needs adjusting, play a video of your choice and look at queue numbers on the OSD. If you get dropped frames and any queue shows as empty (or dangerously low), try increasing it by a few steps. I imagine the default values are sufficient in the majority of scenarios.

So far I only saw ONE video with .ass subtitles that dropped frames on default settings (the same I had to increase my cpu queue to 64 for). Even those subs that used to be regarded as "very heavy" with VSFilter (not Xy-VSFilter) decode without problems on my machine on default settings.

[dukey]

For dvd if the render the menus over the video, and you have large render queues, the latency on the buttons is hilariously bad

[DragonQ]

madshi, I have a log where I closed MPC-HC 1-2 seconds after I saw a black frame using smooth motion. How much of the 3.25 GiB log do you need?

[cyberbeing]

i`ve just installed xysub but i dont know where is the option to put subtitles in the botton black bar on 2.35 movies :(

To-do list. The subtitle interface was explicitly designed to support this, but madVR doesn't yet expose this functionality to users.

[huhn]

will it only move subtitle without positioning information.

like top/bot subs but not typesets?

[xabregas]

To-do list. The subtitle interface was explicitly designed to support this, but madVR doesn't yet expose this functionality to users.

so for me MPC internal subtitle renderer is still needed :sly:

[cyberbeing]

will it only move subtitle without positioning information.

like top/bot subs but not typesets?

That's the plan. This is the next feature we plan to implement in XySubFilter.

[shimaflarex]

It seems the alpha blending bug only happens for me if I start playback with "optimize subtitle quality for performance instead of quality". Is that supposed to happen?
Changing it during playback has no effect.

As madshi said, using 10bit image buffer also fixes it.

[bugmen0t]

madshi, I have a log where I closed MPC-HC 1-2 seconds after I saw a black frame using smooth motion. How much of the 3.25 GiB log do you need?
Hallelujah! I tried several weeks without success. :eek:
Madshi only needs a short part around the black frame. So maybe the last 10 or 30 seconds will do. Note that the log can be heavily compressed by 7zip or WinRAR with text compression enabled.

[DragonQ]

Yeah but the timestamping isn't clear, I have no idea how much of it is the last 30 seconds. The final timestamp is labelled "00977632" and the penultimate line is "00977608". :/

[Soukyuu]

It seems the alpha blending bug only happens for me if I start playback with "optimize subtitle quality for performance instead of quality". Is that supposed to happen?
Changing it during playback has no effect.

As madshi said, using 10bit image buffer also fixes it.I noticed the same and made a new issue summarizing everything on madshi's bugtracker (http://bugs.madshi.net/view.php?id=115)

[madshi]

Can you detect such GPUs and do this automatically?
Yes.

As for the refresh rate, not sure whats the correct refresh rate of my screen but CTRL+ J gives me 60.0014HZ :eek: on my sony TV and 59,9475HZ on my DELL desktop monitor
That's the refresh rate your GPU is outputting. It's still possible that your display runs at a different internal refresh rate. If so, there's no way to find out, except by either trying to get the information from some tech forum, or by trying to output various refresh rates via your GPU and checking which one results in the ghosting going away...

i`ve just installed xysub but i dont know where is the option to put subtitles in the botton black bar on 2.35 movies :(
Not sure I understand that. What kind of subtitles are we talking about? ASS? SRT? VOB? PGS? Is that a Blu-Ray movie? Or a custom encode where the black bars have been removed and are not encoded in the movie?

Currently, madvr sets both the decoding and subtitle queue via cpu queue parameter. Could we have those separated in the future (if it makes sense internally)?

edit: About the only reason I'd want to have it is for more responsive jumping, since the decoding queue never falls below 63/64 while subtitle queue does drop to 11/64 on some heavily styled subs (and that's on a 4GHz X4 970BE cpu...)
So you want a larger subtitle queue than decoder queue? Not sure if I want to offer that functionality. I guess it depends on how many people really need it. It might be a better idea to get the XySubFilter devs to implement proper multi-threading.

madshi, I have a log where I closed MPC-HC 1-2 seconds after I saw a black frame using smooth motion. How much of the 3.25 GiB log do you need?
Great!! The last maybe 1-2 minutes of playback would be great.

Yeah but the timestamping isn't clear, I have no idea how much of it is the last 30 seconds. The final timestamp is labelled "00977632" and the penultimate line is "00977608". :/
Those are milliseconds. So if 00977632 is the last line then substract 2 * 60 * 1000 to get the last 2 minutes of playback. So that would be everything higher than 857632.

It seems the alpha blending bug only happens for me if I start playback with "optimize subtitle quality for performance instead of quality". Is that supposed to happen?
Changing it during playback has no effect.

As madshi said, using 10bit image buffer also fixes it.
This will be fixed in the next build.

[DragonQ]

Those are milliseconds. So if 00977632 is the last line then substract 2 * 60 * 1000 to get the last 2 minutes of playback. So that would be everything higher than 857632.

I see. Log is here. (http://www.aotplaza.com/Files/HTPC/madVR%20-%20log%20(2%20mins%20before%20black%20frame).7z)

[xabregas]

That's the refresh rate your GPU is outputting. It's still possible that your display runs at a different internal refresh rate. If so, there's no way to find out, except by either trying to get the information from some tech forum, or by trying to output various refresh rates via your GPU and checking which one results in the ghosting going away...


Not sure I understand that. What kind of subtitles are we talking about? ASS? SRT? VOB? PGS? Is that a Blu-Ray movie? Or a custom encode where the black bars have been removed and are not encoded in the movie?




The ghosting is like the 100HZ option that some LCDs come with, but much better, sometimes i cant even notice, only in very slow scenes like when people are talking, for instance, 007 skyfall when 007 is talking with Q in the Museum i notest some ghosting, but i didnt notice any ghosting in Avengers movie for example.

As for the subtitles its the last thing u said, like itunes 2.35 movies :sly:

MPC Int subtitle renderer has that option "Position subtitles relative to the video frame", with 3 options, uncheck, default and check, where default is a square inside the checking slot that makes srt subtitles go to the botton of the screen in 2.35 movies. Xy-Vsfilter only do that if i choose "Extend to 16:9" option in the menu, similar to what ffdshow subtitle filter does.

TIA

[madshi]

I see. Log is here. (http://www.aotplaza.com/Files/HTPC/madVR%20-%20log%20(2%20mins%20before%20black%20frame).7z)
Great - thanks! On a quick check the log seems to contain helpful information!

As for the subtitles its the last thing u said :sly:

MPC Int subtitle renderer has that option "Position subtitles relative to the video frame", with 3 options, uncheck, default and check, where default is a square inside the checking slot that makes srt subtitles go to the botton of the screen in 2.35 movies. Xy-Vsfilter only do that if i choose "Extend to 16:9" option in the menu, similar to what ffdshow subtitle filter does.
Currently madVR is limited to only render subtitles onto the active video rect. So if the black bars in a 2.35 movie are stripped from the encoding, madVR can't render the subtitles into them. This will be fixed in some future version. For now I don't have a quick fix available, unfortunately...