[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k] [s4s] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Home]
Board
Settings Home
/a/


File: vlcsnap-14430688.png (246 KB, 704x396)
246 KB
246 KB PNG
How do i get VLC to stop overlapping subtitles? I can't read subs because everything goes on top of each other
>>
MPC + CCCP
>>
You don't, that's why VLC is shit.

Switch to mplayer or mpc.
>>
I heard that they're going to fix this problem in VLC v0.9
>>
VLC is shit.
>>
Holy shit Red Garden is so awesome
>>
>>10181453
It's called MPC + CCCP.

Do it.
>>
Easy. Use something that doesn't have shitty subtitle support. AKA anything but VLC.
>>
>>10181453
>>10181523
>>10182608
samefaggot bumping his own thread
>>
MPC

Though I prefer GOM. It doesnt seem to work with .mkv though = /
>>
>>10181453
>VLC
That's your problem.
>>
>>10181460
QFT. I denied it, I tried other players, but in th e end, nothing beats MPC.
>>
People still use VLC? I thought everyone already knew by now that it was shit.
>>
>>10182699
Then you've never used mplayer.
>>
open source media player classic, cccp and coreavc.
>>
Install CCCP, comes with Media Player Classic and Zoomplayer.

Also, download CoreAVC. Disable FFDShow for AvC/H.264 files, use CoreAVC for them.

If you want another alternative, download MPlayer (full package)
>>
http://tirnanog.fate.jp/mirror/mplayer/mplayer2008.02.06.7z

Problem solved.
>>
>>10182689
Not OP, but I still do since CCCP doesn't work with my ancient OS.

In before "Download XP NAO!". Did that, quickly uninstalled. Latest crack doesn't work, and I'm not begging Microsoft for a fucking code to get XP to work, so screw it.
>>
>>10182811
That's why you use mplayer. Which is also based off libavcodec but doesn't suck dick.
>>
>>10182811
Winders 2000 huh? Look into
http://en.wikipedia.org/wiki/The_KMPlayer
>>
>>10182755

How does CoreAVC work better with H264 files than FFDShow?
>>
I'm using Windows 3.1
I can't use MPC
>>
>http://tirnanog.fate.jp/mirror/mplayer/mplayer2008.02.06.7z
>tirnanog
>Knights of Tirnanog

FUCK YES YOU ARE THE COOLEST POWER RANGERS SPINOFF EVER
>>
>>10182844
AND WHOSE FAULT IS THAT?
>>
>>10182829
ME, but same difference.

Giving it a shot. Thanks.
>>
>>10182842

It's MUCH faster. Like 50% or so. Less lag, ability to play higher res files without AV-desynch...

It also has fantastic SMP support, so decoding speed increases linearly with the number of CPU cores.

And, despite some clashes with x264-encoded files, its support for the H.264 standard is actually much better than ffmpeg's. Not that you are going to care about that, unless you use it to watch HD-DVDs, Blu-Rays or digital HDTV.
>>
>>10182842
Well if you have a slow computer it's still better when it comes to h.264 encoded without slices since the libavcodec parallelization is currently only slice based. But if you have anything close to a modern CPU you should be fine with just ffdshow or anything based off a recent version of libavcodec.
>>
>>10182890
not sure it'll do anything for you on ME. Windows 2000 doesn't require any product activation though... just saying.
>>
>>10182891
All of what you say ceased being true a few months back. CoreAVC is only more than marginally better at this point only if you have an ancient CPU.
>>
>>10182901

There's a patch floating around that separates entropy&MB decoding from transform&deblocking. It gives about a 30% speedup with 2 cores, and doesn't scale beyond that.

It is incorporated in the latest ffdshow-tryouts and CCCP.

I haven't tried it, I don't know if you have to do something to activate it in ffdshow, like you have for the MPEG1/2 decoders.

Still a far cry from CoreAVC's "perfect" frame-parallel decoding, which gives pretty much a 95% speedup.
>>
>>10182891
>And, despite some clashes with x264-encoded files, its support for the H.264 standard is actually much better than ffmpeg's.

Not really. The issue with x264 encodes is that CoreAVC just refuses to play anything with a huge MV range which is actually not something mandated by the spec. In any other case they are both compliant to the spec when it comes to decoding streams.
>>
>>10182946
>
There's a patch floating around that separates entropy&MB decoding from transform&deblocking. It gives about a 30% speedup with 2 cores, and doesn't scale beyond that.

Funny since I've been using the latest libavcodec on a quad core and it scales quite well beyond my old dual core.

>Still a far cry from CoreAVC's "perfect" frame-parallel decoding, which gives pretty much a 95% speedup.

You clearly have no clue what you're talking about with such nonsensical statements. You don't get a 95% speedup with CoreAVC. And if you think it does please provide your evidence.
>>
>>10182946
Combined with the slice based decoding and I can play back any h.264 encoded Blu-Ray/HD DVD much more efficiently than CoreAVC.
>>
>>10181453
Is Gurren Lagann your favourite anime by any chance?
>>
>It also has fantastic SMP support, so decoding speed increases linearly with the number of CPU cores.

>Still a far cry from CoreAVC's "perfect" frame-parallel decoding, which gives pretty much a 95% speedup.

Someone please tell me this guy is joking...
>>
>>10182916

Just benchmarked. TimeCodec, latest ffdshow-tryouts vs CoreAVC 1.5. 640x480, x264 encoded. Core 2 Duo, 2 GHz, 2 MB cache.

ffdshow: 264 fps
coreavc: 588 fps
coreavc limited to 1 CPU: 335 fps

As expected, CoreAVC still smokes ffdshow.
>>
Yea, h264 in an mkv container, 720p, blu-ray encoded, 80% faster frame-rate, benched at 120. These amateurs using xvid with vlc. they should die, dont got nothing on me with my zoomplayer.
>>
>How do i get VLC to stop overlapping subtitles?

>VLC

There! I found your problem.

Now get rid of it.
>>
>>10183081
You must be using an old version of ffdshow or your ffdshow isn't using the latest version of libavcodec. Any way I've benchmarked ffh264 through ffmpeg and my fps is around ~450 for the h.264 encoded Final Fantasy Blu-ray.
>>
/a/ - Technology
>>
>>10183081
Which isn't 50% or even the outrageous 95% speedup you claimed. Way to fail.
>>
>>10183090
ITT: we type random phrases because no-one really knows what anyone else is saying anyway.
>>
>>10183110
Oh and that was with other stuff also running in the background such as WoW.
>>
>>10183090
You can play h.264 in mkv in VLC. What exactly is your point?
>>
>>10183081
That's pretty crap for SD. I get around 300fps benchmarking with 1080p with libavcodec through mplayer.
>>
CCCP + ZOOM PLAYER.

DO it.
>>
>>10183081

Also, on the same machine. 1920x1080, interlaced content, very high birtate from digital satellite broadcast.

ffdshow: 18.7 fps
coreavc 1 core: 26.5 fps
coreavc 2 cores: 47.7 fps

CoreAVC is the only one that can play it back at real time. To ffdshow's credit, this sample didn't show any visible artifacts.
>>
You know that CoreAVC actually costs money?
>>
>>10183198
That's some pretty fail benchmarks. You must have a pretty slow cpu.
>>
>>10183198
Based on the ffdshow speeds you are clearly using a build without parallelization or don't have it enabled.
>>
>>10183261

CPU was stated above, Core 2 Duo T7xxx at 2GHz.

Keep in mind that these files are real content, not anime. Anime decodes much faster.

For example, a 720p MKV gives me 181 fps. If you extrapolate that to 1080p, that'd be 80fps.

Anime is a joke load compared to movies, and has much more variability (this is the reason people complain when the intro in some show desynchs, or when a rain scene kills them - Sudo's DTB ep6 IIRC comes to mind - you just got to a part where most of the picture actually changes every frame, and the bitrate went from 2Mbps to 6 or so; my sample was high in the 20s).

I just tested with mplayer-svn on the same machine, in benchmark mode. It got an impressive 22 fps on the second sample, bit it still doesn't match single-core CoreAVC, much less dual-core.

A quad core machine would only exacerbate the difference.
>>
>>10183462
>I just tested with mplayer-svn on the same machine, in benchmark mode. It got an impressive 22 fps on the second sample, bit it still doesn't match single-core CoreAVC, much less dual-core.

You obviously didn't enable the parallelization then. I can play 1080p60 real time with the latest version of mplayer.

>A quad core machine would only exacerbate the difference.

Only because you aren't benchmarking libavcodec correctly.
>>
>>10183462
>Keep in mind that these files are real content, not anime. Anime decodes much faster.

Showing that again you have no clue what you're talking about. The complexity of decoding has to do with the sizes of the decoded frames not the type of the content.
>>
>>10183462
>Keep in mind that these files are real content, not anime. Anime decodes much faster.

Showing that again you have no clue what you're talking about. The complexity of decoding has to do with the size of the frames that are being decoded not the type of the content.
>>
>>10183479

What kind of parallelization are you talking about?

If you are talking about slice-style, this sample has only one slice per frame and therefore it won't work.

Also, most releases use recent versions of x264, which always uses a single slice too, so it's pretty much useless for most folks here.
>>
>>10182811
You need to do whatever you can to gain some intelligence stats if you can't even pirate Windows XP.
>>
>>10183538
>What kind of parallelization are you talking about?

The slice-based decoding together with the entropy decoding split.

>If you are talking about slice-style, this sample has only one slice per frame and therefore it won't work.

Even without slices I can play back any 1080p24 content real time on my dual core.

>Also, most releases use recent versions of x264, which always uses a single slice too, so it's pretty much useless for most folks here.

Doesn't matter if it's x264 or not. You're still not benchmarking right.
>>
>>10183542
They sell xp?
>>
>>10183525

Haha. I'm not even going to try to argue with you.

Just, please, test this: encode a 1920x1080 unchanging black frame. Then some "live" content (not anime), upscaled if you wish, encoded at 20Mbps.

Test decoding performance.
>>
any of you CCCP fags know how to make it play without letterboxing. I want to watch my shit in actual full screen and VLC (while shitty) is the only one that does this.
>>
>>10183587
>Haha. I'm not even going to try to argue with you.

Good, because you'll fail pretty badly.

>Just, please, test this: encode a 1920x1080 unchanging black frame. Then some "live" content (not anime), upscaled if you wish, encoded at 20Mbps.

>Test decoding performance.

If the two things aren't encoded using the exact same settings at the exact same bitrate your comparison is completely worthless.

Of course the all black frames content will obviously play better because the frames will be compressed far more and hence it will have less decoding complexity. This silly little comparison you want doesn't do anything to help your case.
>>
bitchfags are bitching
>>
>>10183619
So you like your stuff looking like shit? Sorry, not going to help you. Stretch is crap.
>>
>>10183619
Please tell me this is a troll question... Otherwise you're one of the biggest idiots on earth...
>>
File: 1205307829640.png (514 KB, 801x599)
514 KB
514 KB PNG
>>55 posts omitted. Click Reply to view.
>>
>>10183643

Ok. Deep inside I hope you are a troll, but that won't do.

First you said "The complexity of decoding has to do with the size of the frames that are being decoded not the type of the content."; then you said "Of course the all black frames content will obviously play better because the frames will be compressed far more and hence it will have less decoding complexity.". That, good sir, is contradictory and doesn't mak any sense at all.

Bitrate has a STRONG correlation with decoding time, on all codecs, and more predominantly on H.264 using CABAC.

The evidence is so overwhelming that I really don't know what to tell. Just open your favorite episode and keep an eye on CPU usage during heavy scenes (opening perhaps) and then during mostly static scenes (people talking without camera pan). Good luck finding that straight line you're looking for.
>>
>>10183619

Using Media Player Classic, press the '6' key in your numpad, with NumLock activated. Repeat as necessary. Try the other numbers too, and also try holding CTRL while you do so.
>>
CCCP + MPC works great for me, just give it a try.



Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.