Home > encode, x264, x264-bin > x264 rev2208+677 tMod

x264 rev2208+677 tMod

21st July 2012, Saturday Leave a comment Go to comments

Sources on GitHub: https://github.com/astrataro/x264_tMod.

Use win32thread and fprofiled:
tMod/tMod-10bit/tMod+MixAQ/tMod+OreAQ:
Download:
x264_rev2208+677_tMod-Libav.7z : NMMGitHubMediaFire
x264_rev2208+677_tMod-FFmpeg.7z : NMMGitHubMediaFire
x264_rev2208+677_tMod-Lite.7z : NMMGitHubMediaFire

By default uses old style progress indicator, and won’t break GUI’s progress parser. If you like r2204 style progress indicator, try --stylish, but keep careful if you are using GUIs. The difference between r2204 and this --stylish is that console title always uses old style indicator, so you won’t get meaningless raw numbers without labels on console title.

Patches:
–—––—– Download: patches-rev2208.7z: NMMMediaFire
00: L-Smash(including audio encoding)
01: Add an option to use r2204 style progress indicator (New!!)
02: More detailed version head
03: Add back “touhou” tune
04: Film Grain Optimization
05: Fade Compensation
06: Remove stats before renaming
07: Add a parameter to set level of writing options in User Data Unregistered SEI
08: Utilize internal threading in lavf/ffms
09: Auto VBV settings
10: AVI output
11: filters hqdn3d pad vflip yadif patch
12: Log file
13: Encoding time
14: Level force
15: Profile force
16: Cosmetic
17: New experimental AQ mode (modification of Auto-variance AQ)
18: Print video info with lavf/ffms demuxer
19: lto & Ofast
20: AviSynth 16bit hack
21: Skip bit depth filter when possible
22: Use f3kdb for bit depth conversion when needed
23: Video filter: subtitles
24: Detect color matrix with lavf/ffms demuxer
25: Fix AviSynth color space converting matrix
26: Fix swscale color space converting matrix
27: Support “dgi” file extension
28: Variable progress precision, print encoded file size & estimated total size (Updated!!)
29: More detailed “zones” help
30: Unofficial --device
31: Correct framerate of AviSynth to proper NTSC fraction if applicable (New!!)
32: Add “libaacplus”
33-1: MixAQ-core(only in tMod+MixAQ version, remove AQDebug)
33-2: OreAQ-core(only in tMod+OreAQ version, remove AQDebug)

—————————————————-
My other tools list ( including some of the x264 builds ) : MediaFire

Categories: encode, x264, x264-bin
  1. esquall
    22nd July 2012, Sunday at 06,19am UTC

    thanks again😀

  2. easyfab
    22nd July 2012, Sunday at 08,02am UTC

    Hi,
    Thanks for this build.
    I have a question, with a .avs input I got this message scrolling endless when using the new libfdk_aac :

    [libfdk_aac @ 07107A40] extended_data is not set.

    What must I do ? what extended data must I set ( in avs or x264 ) ?

    • 22nd July 2012, Sunday at 12,21pm UTC

      It’s a warning in lavf as the decoder doesn’t get and send extended_data properly. I’ll try to solve it in the next stable release (possibly with next x264 version update). Yet it is not really harmful, as extended_data will be set automatically inside libavcodec, so for now just ignore it.

  3. Kouta84
    9th August 2012, Thursday at 04,55am UTC

    Hi taro,
    I have from some time ago, an endless prompt, can this be fixed too?

    Explaining image: http://www.imagebam.com/image/e886f1205052969

    • 9th August 2012, Thursday at 10,17pm UTC

      Right click on console title, press “Defaults”, and set width of screen buffer size to a larger number in “Layer” tab:

      Press OK to save, and you won’t meet endless prompt any more since the next time you launch the command line.

  4. 9th August 2012, Thursday at 10,27am UTC

    Thank you very much :3

  5. Kouta84
    10th August 2012, Friday at 09,38pm UTC

    Thank you very much me too!!

  6. easyfab
    12th August 2012, Sunday at 08,53am UTC

    Opus audio seems to become very popular and looks promising at low bitrate . As it’s now decoded with lav filters, is it possible to add libopus in your build ?

    • 12th August 2012, Sunday at 03,18pm UTC

      I plan to add opus decoding in the next build, but encoding is another thing, as spec of opus in matroska is just released recently, and even the official opus encoder is still changing its behaviour….

  7. Kouta84
    16th August 2012, Thursday at 12,34am UTC

    another quick question.
    Seems that, at least, x64 builds set framerate as vfr even when is crf, but I’m using ffvhuff as lossless codec, maybe coud this be the problem…

    • Kouta84
      16th August 2012, Thursday at 12,35am UTC

      *cfr

    • 17th August 2012, Friday at 08,33am UTC

      With demuxer=lavf x264 always treats input as vfr for many formats such as mkv, mp4, etc, because lavf doesn’t parse the whole input file before encoding, and thus won’t be able to know if the input file is real vfr content or not. While ffms seems to be not fully reliable for timestamp parsing in some cases. Many people have found timestamp issues with ffms, not only in tMod but also in vanilla builds, with which the encoded file has weird timestamps. For example, one reported such a timecodes of encoded file with ffms, while the input is pure 24/1.001fps with jitter(caused by mkvmerge’s default setting of --timecode-scale 1000000):
      # timecode format v1
      Assume 393.394609
      0,1,384.024946
      2,2,393.388206
      3,5,384.024946
      6,6,393.388206
      7,8,384.024946
      9,9,393.394609

      And with demuxer=lavf the result is correct. In most cases, this only occurs when the input file is broken and ffms fails to parse it correctly, but I’m not quite sure….

      Personally if encoding without AviSynth I prefer to use lavf demuxer with a timecodes file using --tcfile-in. It gives most reliable result, and won’t suffer jitter issue of mkv. Use fixtc in the previous post to correct timecodes extracted from mkv files. I’m writing a new timecodes tool with more features based on it, but fixtc is enough at least to fix such timecodes. The latest version of the new tool is here, still in development and some parts haven’t been finished, e.g., it accepts multiple files for appending function, but till now appending has not been implemented and it simply outputs all converted results, so don’t pass multiple files to it now….

      However, --tcfile-in has some problem working with subtitles vf. This problem comes from original direct264, and I’m not sure which part of codes go wrong so probably won’t fix it in the near future. So if subtitles vf is needed, use --demuxer=lavf but without --tcfile-in, and then manually remux the output file with timecodes. The jitter problem mainly occurs when encoding mkv to mp4 (as the input and output timestamp precisions are same when encoding mkv to mkv or mp4 to mp4), and it does not affect subtitles rendering or x264 encoding progress, so one only need to use timelineeditor/tc2mp4 to fix the timecodes of output file. Actually this jitter issue is only annoying for some ristricted playbacks like PSP. On PC players it is not likely to be noticeable or harmful.

  8. 2nd September 2012, Sunday at 04,11am UTC

    Can you build MixAQ and OreAQ with 10-bit output too?

    • 11th September 2012, Tuesday at 09,38pm UTC

      Yes, but with current codes, 10-bit version of MixAQ/OreAQ works quite different from 8-bit version, and are somehow weird. For example, OreAQ with default settings should allocate more bits to Middle areas, and reduce quality in Bright/Dark areas, but 10-bit OreAQ does not always follow this, so it may output unwanted results. Anyway, variance AQ already does pretty good job with high bit depth encoding. There is little sense to use MixAQ/OreAQ in 10-bit, so I did not look into those patches to fix high bit depth behaviour.

  9. Kouta84
    29th October 2012, Monday at 06,57pm UTC

    It treats input as vfr even with ffms input, are their mp4, avi or mkv, all pure cfr, I tryed some files.
    http://www.imagebam.com/image/af45fe217594811
    now I’m using –force-crf

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: