Skip to main content

Topic: .Ogg Vorbis aotuv (Read 328004 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • sauvage78
  • [*][*][*][*][*]
.Ogg Vorbis aotuv
Reply #450
Just tried to encode a 5.1 file (6 channels in F2K) [Dances With Wolves Speech Sample (DTS Master) 48Khz], -q4 I get 331Kbps (128x5/2=320Kbps expected) & quality seems ok (Didn't ABX)
So personnaly I cannot reproduce the bug (Doesn't mean there isn't any).

Edit:
I used john33's oggenc2 like capma, not venc like alter4.
  • Last Edit: 25 February, 2011, 06:46:40 AM by sauvage78
Rip & Check: EAC Secure [Low/C2]+CUETools [AR Confidence 2+]
NAS (Backup): Flac -4 (for Speed) | CDImage+CUE with F2K
DAP (Playback): Opus 128Kbps | Tracks with F2KM on Android (LG G5)
Video: VP10 (2160p30@24Mbps/2160p60@48Mbps) Asap !!!

  • OggY68
  • [*]
.Ogg Vorbis aotuv
Reply #451
I don't know if the BS (Block-Switching) issue has been addressed in this latest Beta v601 release, but I can still easily ABX the Autechre killer sample.

  • sauvage78
  • [*][*][*][*][*]
.Ogg Vorbis aotuv
Reply #452
The fact that such a bad sample for vorbis is still easyly ABXable is not really a surprise IMHO (unless you expect miracles like aotuv b2 at each aotuv release), the real question is more: is there any improvement even if still ABXable.
  • Last Edit: 25 February, 2011, 07:06:53 AM by sauvage78
Rip & Check: EAC Secure [Low/C2]+CUETools [AR Confidence 2+]
NAS (Backup): Flac -4 (for Speed) | CDImage+CUE with F2K
DAP (Playback): Opus 128Kbps | Tracks with F2KM on Android (LG G5)
Video: VP10 (2160p30@24Mbps/2160p60@48Mbps) Asap !!!

  • Nowings69
  • [*][*]
.Ogg Vorbis aotuv
Reply #453
@john33
Could you change your oggenc2's writing library in OGG COMMENT?

oggenc2(yours)
AO; aoTuV [20110223] (based on Xiph.Org's libVorbis)

venc
AO; aoTuV [20110223] (based on Xiph.Org's libVorbis)

I wanna diplay your oggenc2 for aotuv
like this

Are you saying that you want the 'AO: ' removed from the front of the Vendor String? If so, I'm not sure that I should considering Aoyumi created the modified libraries and that is his chosen Vendor String. To change it might suggest that I am using a different version of the libraries.


I wish to read from OGG COMMENT as TAG in writing library by Encoder at (or)Foobar
Im not talking bout BSD or ect
Well.i gave up it
Thanks jonn33

  • OggY68
  • [*]
.Ogg Vorbis aotuv
Reply #454
@sauvage78

Yes you're right this new release is definately better.  I ABXed the following samples and could only ABX Autechre:

01- DCT Killer Samples (Lossless)\01- Artefact+Context\01- Abfahrt Hinwil (Artefact+Context) Lossless.flac
01- DCT Killer Samples (Lossless)\01- Artefact+Context\02- Autechre (Artefact+Context) Lossless.flac
01- DCT Killer Samples (Lossless)\01- Artefact+Context\03- Castanets (Artefact+Context) Lossless.flac
01- DCT Killer Samples (Lossless)\01- Artefact+Context\04- Harlem (Artefact+Context) Lossless.flac
01- DCT Killer Samples (Lossless)\01- Artefact+Context\05- Kraftwerk (Artefact+Context) Lossless.flac
01- DCT Killer Samples (Lossless)\01- Artefact+Context\06- Rush (Artefact+Context) Lossless.flac
01___Pet_Shop_Boys___In_The_Night___cutted.flac
bibilolo.flac
emese.flac
Linchpin__Edit_.flac
Replica._Lossy.flac

So thanks Aoyumi for the splendid work.

  • dutch109
  • [*][*][*]
.Ogg Vorbis aotuv
Reply #455
@OggY68

Did the bitrate increase a lot for these samples (compared to previous aoTuV) ?
Vorbis -q2/5 (Android/PC) & WavPack -hhx6m
https://playnoise.com/

  • OggY68
  • [*]
.Ogg Vorbis aotuv
Reply #456
@OggY68

Did the bitrate increase a lot for these samples (compared to previous aoTuV) ?

I only tested them at default -q3 quallity setting.

  • lvqcl
  • [*][*][*][*][*]
  • Developer
.Ogg Vorbis aotuv
Reply #457
Other killer samples for vorbis are badvilbel and the_product.

  • alter4
  • [*][*][*]
.Ogg Vorbis aotuv
Reply #458
Hi ALL.

I am Back with 6ch encoding problems. The problem is sample(or maybe format) specific.
There is a one of problematic 6ch file.
Download sample and try to encode using aotuv beta6.01 to catch the bug.

Thx and hope somebody helps to fix.

P.S. HA! looks I find the problem! aotuv can't properly encode  44100 rate 6ch files, but can 48000 rate! Please confirm!
  • Last Edit: 25 February, 2011, 07:10:04 PM by alter4

  • capma
  • [*]
.Ogg Vorbis aotuv
Reply #459
Confirmed. Resampled file (to 48000) encodes correctly.
--------------------

  • alter4
  • [*][*][*]
.Ogg Vorbis aotuv
Reply #460
It is clear now. The question is closed, new bug report is open.

P.S. But the good news are 32000hz 6ch files convert properly as well, so I think it something like "slip of the pen" in aotuv code, and Aoyumi (if hi has free time and good opportunity  ) can easily fix it.
  • Last Edit: 26 February, 2011, 07:21:02 AM by alter4

  • alter4
  • [*][*][*]
.Ogg Vorbis aotuv
Reply #461
Quote
aoTuV  Beta6.02 [beta6.01  >> beta6.02] (2011/02/27)
# In the specific pattern (different from beta6.01), the bug that caused overflow was fixed.

New aotuv version is out, but the bug with 6ch files is still there.


Quote
C:\unknown>venc.exe -q4 6ch.wav
aoTuV Beta 6.02 - OggVorbis Encoder © 2003-2011 Aoyumi

Output file : "6ch.ogg"
Stream info.: 6ch/44.1kHz/16bit(int) -> 6ch/44.1kHz/32bit(float)

Quality 4 (nominal bitrate : 324kbps)
Processing...Done.

[Report of the encoded file]
  play time    : 17.842 sec.
  encode time  : 1.968 sec. (x9.066)
stream bitrate : 94.08 kbps
bit reduction  : 1 / 45


Thx for your work Aoyumi.
  • Last Edit: 27 February, 2011, 09:03:29 AM by alter4

  • Aoyumi
  • [*][*][*]
.Ogg Vorbis aotuv
Reply #462
About 5.1ch
I recognize the problem.
There are some causes, and time is necessary in investigation and a revision.
Sorry.

Aoyumi

  • john33
  • [*][*][*][*][*]
  • Developer
.Ogg Vorbis aotuv
Reply #463
aoTuVb6.02 compiles are now at Rarewares.
John
----------------------------------------------------------------
My compiles and utilities are at http://www.rarewares.org/

  • alter4
  • [*][*][*]
.Ogg Vorbis aotuv
Reply #464
About 5.1ch
I recognize the problem.
There are some causes, and time is necessary in investigation and a revision.
Sorry.

Aoyumi


OK. I think this bug is not critical, not all guys rip 6ch music from movies to ogg:)
But just for sure Aoyumi could I ask you? Will you profile fix ...as ready... or postpone the fix for a long term?

Sorry again if you feel me tactless.
  • Last Edit: 28 February, 2011, 12:43:51 PM by alter4

  • Aoyumi
  • [*][*][*]
.Ogg Vorbis aotuv
Reply #465
About 5.1ch
I recognize the problem.
There are some causes, and time is necessary in investigation and a revision.
Sorry.

Aoyumi


OK. I think this bug is not critical, not all guys rip 6ch music from movies to ogg:)
But just for sure Aoyumi could I ask you? Will you profile fix ...as ready... or postpone the fix for a long term?

Sorry again if you feel me tactless.

I want to release fixed version in the time early as possible.
But I cannot yet declare it about the time...

  • B7k
  • [*]
.Ogg Vorbis aotuv
Reply #466
I dont rip in 6ch but have been a big fan of youre work Aoyumi cant wait to try the new version out.



  • urak
  • [*]
.Ogg Vorbis aotuv
Reply #467
Thank you very much Aoyumi! 

  • lvqcl
  • [*][*][*][*][*]
  • Developer
.Ogg Vorbis aotuv
Reply #468
lead_voice sample:

1 - original
2 - encoded with aotuv b6.02 -q4
3 - resampled to 88kHz and encoded
4 - resampled to 96kHz and encoded


  • akapuma
  • [*]
.Ogg Vorbis aotuv
Reply #469
1 - original
2 - encoded with aotuv b6.02 -q4
3 - resampled to 88kHz and encoded
4 - resampled to 96kHz and encoded
Hello lvqcl,

thank you for your tests.

I think, number 2 is OK, and number 3 is not OK.

What is with number 4? Is it OK? There is some schlieren between the lines.

I think, the original is 44,1kHz. Have you tested 48kHz, too? This is the common sample rate for digital TV (DVB).

Best regards

akapuma

  • lvqcl
  • [*][*][*][*][*]
  • Developer
.Ogg Vorbis aotuv
Reply #470
I cannot see or hear problems with this sample resampled to 48kHz.

Quote
What is with number 4? Is it OK?

Numbers 3 and 4 aren't not OK IMHO (ABXed 8/8).

  • Aoyumi
  • [*][*][*]
.Ogg Vorbis aotuv
Reply #471
lead_voice sample:

1 - original
2 - encoded with aotuv b6.02 -q4
3 - resampled to 88kHz and encoded
4 - resampled to 96kHz and encoded

Thank you for a report.

If the problem that you recognize produces even "libvorbis 1.3.2", I understand the problem.
I intend to try the improvement of the problem in the future.

Aoyumi

  • lvqcl
  • [*][*][*][*][*]
  • Developer
.Ogg Vorbis aotuv
Reply #472
If the problem that you recognize produces even "libvorbis 1.3.2", I understand the problem.
I intend to try the improvement of the problem in the future.


Oops. Sorry that I didn't this before. Yes, libvorbis 1.3.2 also shows similar problems with 88 and 96kHz, but in other places of the track.
  • Last Edit: 07 March, 2011, 10:06:15 AM by lvqcl

  • akapuma
  • [*]
.Ogg Vorbis aotuv
Reply #473
Hello Aoyumi,

is everything OK with you? In your user profile is written, that you are living in Kanto. And I saw, that Kanto is near Fukushima.

My best wishes to you, to your family and the whole Japanese nation.

Best regards

akapuma

  • greynol
  • [*][*][*][*][*]
  • Global Moderator
.Ogg Vorbis aotuv
Reply #474
These types of messages should be done via PM, please read TOS #5 if you have any questions.

That said, my heart goes out to all those affected and to be affected by the catastrophe in Japan.
  • Last Edit: 17 March, 2011, 02:03:38 PM by greynol
13 February 2016: The world was blessed with the passing of a truly vile and wretched person.

Your eyes cannot hear.