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.

 

Go Back   Doom9's Forum > Programming and Hacking > Development

Reply
 
Thread Tools Search this Thread Display Modes
Old 2nd November 2011, 23:09   #481  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
AFAIK, VBR mode woks just fine without -b and -B. In that case LAME will simply use its default minimum/maximum bitrates for your 'VBR' quality value.

So IMO you shouldn't overwrite these, unless you have a very good reason to do so. The documentation even states that "the use of -B is NOT RECOMMENDED".

Also, as mentioned before, the lowest bitrate allowed in MP3, according to MPEG-1, is 32 kbps.

Even lower bitrates are only supported with MPEG-2 extensions. And even then only for 16, 22.05 and 24 kHz, but not for the more common sample rates, like 44.1 or 48 kHz.

Last but not least and most important:

I encoded a 44.1 kHz file with LAME and "-V9" (no other options set!). LAME encoded at 22.05 kHz. The output MP3 file did contain 8, 16 and 24 kbps frames!

Hence LAME will automatically downsample the input for such ultra-low quality settings. And, if it does so, it will make use of the lowest bitrates - even without '-b' option

(BTW: You may want to check out MPEG Audio Info, which gives detailed per-frame info, to analyze the output MP3 files.)
__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.



Last edited by LoRd_MuldeR; 2nd November 2011 at 23:45.
LoRd_MuldeR is offline   Reply With Quote
Old 2nd November 2011, 23:59   #482  |  Link
Taurus
Registered User
 
Taurus's Avatar
 
Join Date: Mar 2002
Location: Krautland
Posts: 824
Out of curiosity I gave the FHG aac encoder a shot.
Maybe this has been adressed before:
Why is the slider in the compression settings tab only working at insane intervalls?
From level 0.50 to 0.9 there is almost no increase in bitrate
and even 1.0 is giving me a file undersized compared to the source.
The FHG files are from the newest 5.622 Winamp.
Back to nero for now.
Taurus is offline   Reply With Quote
Old 3rd November 2011, 00:28   #483  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
Quote:
Originally Posted by Taurus View Post
Out of curiosity I gave the FHG aac encoder a shot.
Maybe this has been adressed before:
Why is the slider in the compression settings tab only working at insane intervalls?
From level 0.50 to 0.9 there is almost no increase in bitrate
and even 1.0 is giving me a file undersized compared to the source.
The FHG files are from the newest 5.622 Winamp.
Back to nero for now.
As the FHG encoder is ClosedSource, only one of the Winamp/FHG engineers could answer this question

All that LameXP can do is passing options to the API provided by the encoder. And the VBR quality selector of the FHG encoder is relatively coarse-grained:

The quality value can be selected between 1 and 5, integer (whole numbers) only. The quality selector of the Nero encoder is much more fine-grained, as you know...

(Note that the quality slider in LameXP was designed for Nero AAC. For FHG AAC the values in range 0.0 to 1.0 are mapped to 1...5 in a linear fashion)
__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.



Last edited by LoRd_MuldeR; 3rd November 2011 at 00:40.
LoRd_MuldeR is offline   Reply With Quote
Old 3rd November 2011, 01:39   #484  |  Link
Othniel Graichen
Linux Software Engineer
 
Join Date: Oct 2011
Location: San Antonio, TX
Posts: 4
Quote:
Originally Posted by LoRd_MuldeR View Post
AFAIK, VBR mode works just fine without -b and -B. In that case LAME will simply use its default minimum/maximum bitrates for your 'VBR' quality value.

Also, as mentioned before, the lowest bitrate allowed in MP3, according to MPEG-1, is 32 kbps.

Even lower bitrates are only supported with MPEG-2 extensions.
I stand corrected. Thank you. I like to learn.

However, I am aware that I am utilizing MPEG-2.5 extensions. Maybe LAMExp should have another checkbox under the Additional Options tab in the channel mode/sampling rate box called "Enable MPEG-2.5".

When such a box is checked, 8, 11.025 and 12 kHz should be available in the sample rate pull-down box. I suggest also that the minimum bitrate field be set to 8 when MPEG-2.5 is enabled.

The application is not theoretical but that of mp3 audiobooks and
voice recordings.
Othniel Graichen is offline   Reply With Quote
Old 3rd November 2011, 08:22   #485  |  Link
Taurus
Registered User
 
Taurus's Avatar
 
Join Date: Mar 2002
Location: Krautland
Posts: 824
Thank you
Taurus is offline   Reply With Quote
Old 4th November 2011, 11:20   #486  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
Quote:
Originally Posted by Othniel Graichen View Post
I stand corrected. Thank you. I like to learn.

However, I am aware that I am utilizing MPEG-2.5 extensions. Maybe LAMExp should have another checkbox under the Additional Options tab in the channel mode/sampling rate box called "Enable MPEG-2.5".

When such a box is checked, 8, 11.025 and 12 kHz should be available in the sample rate pull-down box. I suggest also that the minimum bitrate field be set to 8 when MPEG-2.5 is enabled.

The application is not theoretical but that of mp3 audiobooks and
voice recordings.
I don't think we need yet another option. LAME doesn't have a switch for that either. It's as simple as that: When encoding to a sample rate that is only supported in MPEG-1 (32, 44.1 or 48 kHz), LAME will create a MPEG-1 MP3 stream. When encoding to a sample rate that is only supported in MPEG-2 (16, 22.05 and 24 kHz), LAME will create a MPEG-2 MP3 stream. And when encoding to a sample rate that is only supported in MPEG-2.5 (8, 11.025 and 12 kHz), LAME will create a MPEG-2.5 MP3 stream. For 99.9% of the user this difference doesn't even matter (and thus we should confuse them with different MPEG standard versions). Moreover, in any case, LAME will make use of all supported bit rates (for the current sample rate), at least when encoding in VBR mode. There is no need to use -b or -B to "enable" additional bit rates. Instead these options can be used to further restrict the bit rates that VBR mode can choose from! And thus you should only be using them, if you have a very good reason to do so. Otherwise trust LAME's VBR algorithm...

__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.



Last edited by LoRd_MuldeR; 4th November 2011 at 11:53.
LoRd_MuldeR is offline   Reply With Quote
Old 6th November 2011, 10:17   #487  |  Link
MadRat
Registered User
 
MadRat's Avatar
 
Join Date: Jan 2008
Posts: 110
Quote:
Originally Posted by LoRd_MuldeR View Post
As always in this situation: Please re-scan the suspicious file at http://www.virustotal.com/ to be sure it isn't really infected.

Then report the FALSE POSITIVE to the developer of your AntiVirus software. And, if they don't fix the problem, switch to a better product.

(I can't do anything about the issue, because it is impossible to know what aspect of LAME triggers the failure in their software)
It seems to be a false positive.

Quote:
File already submitted: The file sent has already been analysed by VirusTotal in the past. This is same basic info regarding the sample itself and its last analysis:
MD5: 1aad0bc496cf8bd07b86ebc11bdb8a33
Date first seen: 2011-11-02 20:48:41 (UTC)
Date last seen: 2011-11-02 20:48:41 (UTC)
Detection ratio: 0/43
Thank you, for being patient.
MadRat is offline   Reply With Quote
Old 6th November 2011, 13:38   #488  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
Clearly a FALSE POSITIVE

Interestingly the "Symantec" scanner, which is the company behind "Norton", also confirms that the file is clean.

Maybe the issue has already been fixed ???
__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.



Last edited by LoRd_MuldeR; 6th November 2011 at 13:41.
LoRd_MuldeR is offline   Reply With Quote
Old 11th November 2011, 07:07   #489  |  Link
ikuban
Registered User
 
Join Date: Mar 2011
Posts: 6
Is the new RMS equalization mode similar to a dynamic range compression?

Last edited by ikuban; 11th November 2011 at 07:09.
ikuban is offline   Reply With Quote
Old 11th November 2011, 09:38   #490  |  Link
manolito
Registered User
 
manolito's Avatar
 
Join Date: Sep 2003
Posts: 2,321
Quote:
Originally Posted by ikuban View Post
Is the new RMS equalization mode similar to a dynamic range compression?
No, the dynamic range of the source is not touched by this option. What is getting changed is the balance between the different channels of a multichannel source. Here is a quote from the SoX manual regarding the gain -nb effect:
Quote:
Given the −e option, the levels of the audio channels of a multi-channel file are ‘equalised’, i.e.
gain is applied to all channels other than that with the highest peak level, such that all channels
attain the same peak level (but, without also giving −n, the audio is not ‘normalised’).

The −B (balance) option is similar to −e, but with −B, the RMS level is used instead of the peak
level. −B might be used to correct stereo imbalance caused by an imperfect record turntable car-
tridge. Note that unlike −e, −B might cause some clipping.

−b is similar to −B but has clipping protection, i.e. if necessary to prevent clipping whilst
balancing, attenuation is applied to all channels. Note, however, that in conjunction with −n, −B
and −b are synonymous.

Cheers
manolito
manolito is offline   Reply With Quote
Old 11th November 2011, 12:17   #491  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
Quote:
Originally Posted by ikuban View Post
Is the new RMS equalization mode similar to a dynamic range compression?
Well, kind of, I would say.

It does not compress the dynamic range "inside" on channel. That is: All samples in one channel will always be amplified by the same factor.

However one channel may be amplified stronger than the other ones, which possibly (not necessarily) changes the dynamic range "between" the channels - that may be desired or not.

(Be aware that, due to a bug(?) in SoX, the plain "-n" option files for some files, which was all the reason to use "-nb" by default)
__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.



Last edited by LoRd_MuldeR; 11th November 2011 at 13:48.
LoRd_MuldeR is offline   Reply With Quote
Old 11th November 2011, 14:27   #492  |  Link
manolito
Registered User
 
manolito's Avatar
 
Join Date: Sep 2003
Posts: 2,321
Quote:
(Be aware that, due to a bug(?) in SoX, the plain "-n" option files for some files, which was all the reason to use "-nb" by default)
Oh yes, this is a SoX bug which has been there for some time. Have a look here:
http://sox.sourceforge.net/Docs/Bugs

The recommended workaround is:
Quote:
Create a folder called `tmp' at the top level of the `C:' drive (and any other drive from which SoX may be invoked).
But this workaround does not work at all. And SoX has not been updated since February 2011, so there is not much hope that this bug will get fixed anytime soon...

Which makes me think if it would be a good idea to ditch SoX for normalizing altogether...


Cheers
manolito
manolito is offline   Reply With Quote
Old 11th November 2011, 15:12   #493  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
I think that's another issue.

LameXP already tells SoX to put temporary files into the LameXP Temp folder, which lies inside %TEMP% and therefore has no problems with file permissions.

The problem is more that normalization with just "-n" works most of the time, but sometimes creates an empty file without an error message...

(BTW: Whoever tries to create a sub-folder for temporary files in the Root folder of the system drive and then wonders about missing permissions doesn't have a clue. That wouldn't work on a Unix system either - at least not without using 'sudo'. It's certainly not a Vista/UAC-specific issue)
__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.



Last edited by LoRd_MuldeR; 11th November 2011 at 15:18.
LoRd_MuldeR is offline   Reply With Quote
Old 11th November 2011, 18:09   #494  |  Link
manolito
Registered User
 
manolito's Avatar
 
Join Date: Sep 2003
Posts: 2,321
Quote:
The problem is more that normalization with just "-n" works most of the time, but sometimes creates an empty file without an error message...
For me it does not look like sometimes. For a 6-ch wav file (no matter if the file is in the extensible format or not) SoX always creates an empty file without an error message, while for a 2-ch wav file SoX always works like it should.

I did many tests to determine this behavior, and these tests were done from the SoX command line (completely outside of LameXP). For 6-ch wav files the error was always something like "Could not create temporary device".


Cheers
manolito
manolito is offline   Reply With Quote
Old 11th November 2011, 18:29   #495  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
Well, I was able to reproduce the problem with some multi-channel (here I mean: more than two channels) files, but not with all. Also I never saw an error message, only an empty file.

If you see "Could not create temporary device", then please try with --temp . so that SoX will put the Temp file into the current directory...
__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.


LoRd_MuldeR is offline   Reply With Quote
Old 12th November 2011, 08:19   #496  |  Link
Motenai Yoda
Registered User
 
Motenai Yoda's Avatar
 
Join Date: Jan 2010
Posts: 688
I notice that sometimes normalizing by peak don't work well, analyzing the result file with audition it finds peaks over -0.5dB..

Edit: It's a lack of lossy compression
__________________
powered by Google Translator

Last edited by Motenai Yoda; 12th November 2011 at 09:02.
Motenai Yoda is offline   Reply With Quote
Old 12th November 2011, 10:04   #497  |  Link
Hobbe
Registered User
 
Join Date: Oct 2007
Posts: 42
Thanks for the new version... But I get this error when I tried to encode a mp3 with album art..
unsupported image: 'C:\Users\*****\AppData\Local\Temp\7f6f2bd55a3e4905a01d7d2723996b41\735d2eb984d746819396f5d3289e3cf8.jpg'.
Specify JPEG/PNG/GIF image

Exited with code: 0x0001

Does't lame support jpg at all? or is it just a problem with this file?
Hobbe is offline   Reply With Quote
Old 12th November 2011, 12:35   #498  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
Hello, Hobbe. As you didn't post the complete log, it's a bit hard to guess.

But it looks like LameXP (or MediaInfo to be more precise) found a JPEG image in the input file (or you manually added cover art in LameXP's MetaInfo editor), but then the encoder (LAME?) refused to embed that picture into the output file.

Unless you provide an example, I can only speculate that either the JPEG file is broken/incomplete somehow or what appears to be a JPEG file at first glance turns out to not be a JPEG file...

(You may want to open 'C:\Users\John Doe\AppData\Local\Temp\xxxx\yyyy.jpg' in IrfanView and check what happens)
__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.



Last edited by LoRd_MuldeR; 12th November 2011 at 12:38.
LoRd_MuldeR is offline   Reply With Quote
Old 12th November 2011, 13:16   #499  |  Link
Hobbe
Registered User
 
Join Date: Oct 2007
Posts: 42
Thanks... I did use mp3tag to add album art. This time I scaled down the picture to a smaller size and readded the art and now it works.. strange..

Can I add album art directly in LameXP? if so... how? =)

Any plans to make it possible to use Quicktime encoder?
Hobbe is offline   Reply With Quote
Old 12th November 2011, 13:30   #500  |  Link
LoRd_MuldeR
Software Developer
 
LoRd_MuldeR's Avatar
 
Join Date: Jun 2005
Location: Last House on Slunk Street
Posts: 12,933
Quote:
Originally Posted by Hobbe View Post
Thanks... I did use mp3tag to add album art. This time I scaled down the picture to a smaller size and readded the art and now it works.. strange..

Can I add album art directly in LameXP? if so... how? =)
Meta Data editor. Add the file, then click "View Details" and "Artwork". In the Artwork view, you can click "Edit" to load/change the album art.

Quote:
Originally Posted by Hobbe View Post
Any plans to make it possible to use Quicktime encoder?
You probably mean the QuickTime/iTunes AAC encoder?

Well, I certainly can't include the QuickTime AAC encoder with LameXP due to licensing issues. If there is some OpenSource CLI front-end available for QuickTime AAC, I could add support for that in a similar way to FHG AAC encoder. Which means that people would still have to download and install QuickTime from the official web-site. Then copy the encoder DLL's to the correct place manually.

But due to a deep-rooted antipathy against QuickTime products, that probably won't happen soon
__________________
There was of course no way of knowing whether you were being watched at any given moment.
How often, or on what system, the Thought Police plugged in on any individual wire was guesswork.



Last edited by LoRd_MuldeR; 12th November 2011 at 14:52.
LoRd_MuldeR is offline   Reply With Quote
Reply

Tags
aac, aotuv, flac, lame, lamexp, mp3, mp4, ogg, oggenc, opus, vorbis

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 21:29.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2018, vBulletin Solutions Inc.