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. |
18th April 2017, 21:38 | #81 | Link |
Professional Code Monkey
Join Date: Jun 2003
Location: Kinnarps Chair
Posts: 2,555
|
Yes, delete all old ones. I forgot to bump the index version and using an old index=>old issues.
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet |
18th April 2017, 22:48 | #82 | Link |
Registered User
Join Date: Oct 2014
Posts: 268
|
The weird things I had at the start of my MTS files seem fixed now. Stepping through the files (forward) reveal nothing out of the ordinary an all good.
Stepping backwards still starts acting weird, but you might just call that 'by design and do not seek backwards' I don't know. If I start stepping backwards (after having stepped forward through the whole video) the first few frames seek very quick (cached I guess) then the frames seek back slower (to be expected, seeking backwards in a file that is not All-I). But after around 80 frames the picture 'freezes'. I can step back a few frames but they all look the same (they didn't while seeking forward through the file). This lasts for around 10 frames (GOP length?). Then it goes to slowly seeking backwards frame by frame again. Now, if I start stepping forward at this moment, those 10 'freeze frames' are still there. I have to step back a whole lot more and then step forward ago and now the exact same frames step forward OK. So is there actually some seeking gone bad here, or is the Vapoursynth cache being naughty? Or both? What appears to be happening is that after seeking +/- 80 frames backwards ffms2 produces 10 frames of 'freeze frame', and Vapoursynth caches those 10 frames, so stepping them forward again will not fix it while the bad frames are still in the cache. @Myrsloik: You still have my MTS example file right? Create a simple script that opens it with ffms2 with nothing more, open it in Vapoursynth Editor. Go to the last frame, then start holding down the arrow-down key to step backward. Around frame 192 the 'freeze frame' occurs. Also happens in VDFilterMod (to rule out a bug in Vapoursynth Editor or something). If I open the file and don't seek to the end but instead seek to frame +/- 220 and start holding down the arrow key, the glitch once again occurs at frame +/- 192 (same spot) so it does seem related to that position in the file, not the amount of frames I stepped backwards or anything. PS (And yes, I deleted .ffindex file before I started) PS2 (And I'm talking about test5, just to be clear) |
18th April 2017, 23:40 | #83 | Link | |
Registered User
Join Date: Jun 2002
Location: On thin ice
Posts: 6,837
|
Quote:
I want to encode this sample: http://www.mediafire.com/file/t51xbk...owing_Pains.ts Some script code and command lines to get me started would be helpful.
__________________
https://github.com/stax76/software-list https://www.youtube.com/@stax76/playlists |
|
19th April 2017, 10:32 | #84 | Link | |
Professional Code Monkey
Join Date: Jun 2003
Location: Kinnarps Chair
Posts: 2,555
|
Quote:
I wish I'd kept a bigger collection of samples.
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet |
|
27th April 2017, 19:43 | #85 | Link |
RipBot264 author
Join Date: May 2006
Location: Poland
Posts: 7,815
|
Is this normal that in latest ffms2000 test5 all frames in VC-1 are detected as keyframes?
Example Code:
# keyframe format v1 fps 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Code:
# keyframe format v1 fps 0 0 1 25 49 53 77 101 106 130 154 178 202 226 250 274 298 322 346 370 394 418 442 466 490
__________________
Windows 7 Image Updater - SkyLake\KabyLake\CoffeLake\Ryzen Threadripper |
27th April 2017, 19:56 | #86 | Link |
Professional Code Monkey
Join Date: Jun 2003
Location: Kinnarps Chair
Posts: 2,555
|
No, can I have a small sample?
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet |
27th April 2017, 20:01 | #87 | Link |
RipBot264 author
Join Date: May 2006
Location: Poland
Posts: 7,815
|
No problem give me 15 minutes...
__________________
Windows 7 Image Updater - SkyLake\KabyLake\CoffeLake\Ryzen Threadripper |
27th April 2017, 20:16 | #88 | Link |
RipBot264 author
Join Date: May 2006
Location: Poland
Posts: 7,815
|
__________________
Windows 7 Image Updater - SkyLake\KabyLake\CoffeLake\Ryzen Threadripper |
27th April 2017, 20:51 | #89 | Link | |
Professional Code Monkey
Join Date: Jun 2003
Location: Kinnarps Chair
Posts: 2,555
|
Quote:
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet |
|
13th May 2017, 08:55 | #91 | Link |
Professional Code Monkey
Join Date: Jun 2003
Location: Kinnarps Chair
Posts: 2,555
|
Audio handling wasn't changed at all. Oddly enough.
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet |
13th May 2017, 09:05 | #92 | Link |
Broadcast Encoder
Join Date: Nov 2013
Location: Royal Borough of Kensington & Chelsea, UK
Posts: 2,904
|
Uhm... Well, with the stable FFMpegSource2 I was able to open a file in avisynth, and pipe the audio using BePipe to NeroAACEnc, but I can't with ffms2000.
I think you can reproduce my error pretty easily with any source. Sample: FFMpegSource2("Ep1.avi", fpsnum=24000, fpsden=1001, atrack=-1) BePipe.exe --script "Import(^AVS Script.avs^)" | neroAacEnc.exe -lc -br 320000 -if - -of "audio.m4a" |
22nd May 2017, 02:31 | #93 | Link |
Broadcast Encoder
Join Date: Nov 2013
Location: Royal Borough of Kensington & Chelsea, UK
Posts: 2,904
|
Hi,
I have an .mka audio track which contains two tracks: the first one is a FLAC lossless audio track with 6 channels, and the second one is an AC3 stereo audio track. When I use FFAudioSource() it takes the second track (AC3) instead of the very first one. I think it should use the first available audio track, so FLAC. Is this a bug? Do you want me to upload the .mka file? General Unique ID : 174555563715229058114849233714037036678 (0x83522DC00F66393A809D00CAF83E9686) Complete name : O:\episodi\[VCB-Studio] Shingeki no Bahamut Genesis [Ma10p_1080p]\bahamutep2audiodolby.mka Format : Matroska Format version : Version 4 / Version 2 File size : 664 MiB Duration : 23 min 52 s Overall bit rate mode : Variable Overall bit rate : 3 890 kb/s Encoded date : UTC 2016-11-12 14:21:14 Writing application : mkvmerge v9.5.0 ('Quiet Fire') 64bit Writing library : libebml v1.3.4 + libmatroska v1.4.5 Audio #1 ID : 1 Format : FLAC Format/Info : Free Lossless Audio Codec Codec ID : A_FLAC Duration : 23 min 52 s Bit rate mode : Variable Bit rate : 3 249 kb/s Channel(s) : 6 channels Channel positions : Front: L C R, Side: L R, LFE Sampling rate : 48.0 kHz Frame rate : 11.719 FPS (4096 spf) Bit depth : 24 bits Stream size : 555 MiB (84%) Writing library : libFLAC 1.2.1 (UTC 2007-09-17) Language : Japanese Default : Yes Forced : No Audio #2 ID : 2 Format : AC-3 Format/Info : Audio Coding 3 Format settings, Endianness : Big Codec ID : A_AC3 Duration : 23 min 52 s Bit rate mode : Constant Bit rate : 640 kb/s Channel(s) : 2 channels Channel positions : Front: L R Sampling rate : 48.0 kHz Frame rate : 31.250 FPS (1536 spf) Bit depth : 16 bits Compression mode : Lossy Stream size : 109 MiB (16%) Language : Japanese Service kind : Complete Main Default : No Forced : No |
22nd May 2017, 09:24 | #94 | Link |
Professional Code Monkey
Join Date: Jun 2003
Location: Kinnarps Chair
Posts: 2,555
|
Are both tracks manually selectable?
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet |
23rd May 2017, 20:54 | #96 | Link |
Professional Code Monkey
Join Date: Jun 2003
Location: Kinnarps Chair
Posts: 2,555
|
Released test6. Download link in first post.
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet |
16th July 2017, 10:34 | #98 | Link |
Registered User
Join Date: Jun 2002
Location: On thin ice
Posts: 6,837
|
@Myrsloik
I believe there is a memory leak with the x64 build, I've not examined it in detail, in staxrip for every job memory grew 50-100 MB, I always thought it's staxrip and I've been too careless with resources, forgetting to detach event handlers and things, I changed my design and had major problems with it, by accident I noticed that it don't happen with l-smash, I've tried with VirtualDub, after every opening and closing circle memory grows 10 MB, for GUIs this is very problematic because they don't keep one script open but rather work with multiple scripts opening and closing them on demand. I've currently an issue on the tracker by somebody that is trying to batch process 3200 files in one go. :-)
__________________
https://github.com/stax76/software-list https://www.youtube.com/@stax76/playlists |
16th July 2017, 12:57 | #100 | Link |
Registered User
Join Date: Jun 2002
Location: On thin ice
Posts: 6,837
|
I knew it was coming.
__________________
https://github.com/stax76/software-list https://www.youtube.com/@stax76/playlists |
|
|