WINANP Plugin for MP3 Surround playback.
Download it from here:(zip-file, 1 MB)
ftp://ftp1.fraunhofer.de/institute/iis/am...namp_Plugin.exe
Well done, everybody's going to vote "NO" 
QUOTE(Bodhi @ Jun 27 2007, 14:36) 
Well done, everybody's going to vote "NO"
I'd imagine that even if there was a "Yes" vote possible the results would change little, if at all....
QUOTE(zipr @ Jun 27 2007, 21:22)
QUOTE(Bodhi @ Jun 27 2007, 14:36)
Well done, everybody's going to vote "NO"
I'd imagine that even if there was a "Yes" vote possible the results would change little, if at all....
Agreed, the poll is comically accurate.
Why the hell should I have "surround" MP3 files? I want my MP3s to sound like original as much as possible.
QUOTE(PHOYO @ Jun 28 2007, 00:00)
Why the hell should I have "surround" MP3 files? I want my MP3s to sound like original as much as possible.
@Jimpin: Ask a moderator to fix the poll if you have troubles at setting it up properly.
Yes, i will, its the forum problem.
But guys, the quality of the Fraunhofer MP3 decoder is very high.
QUOTE(PHOYO @ Jun 28 2007, 00:00)
Why the hell should I have "surround" MP3 files? I want my MP3s to sound like original as much as possible.
Err... what about multichannel sources, e.g. Dolby Surround or 5.1 DVD-Audio? Or do you confuse the term "surround" with any pseudo-3D-effects added by some DSPs to simulate an areal sound?
QUOTE(Jimpin)
But guys, the quality of the Fraunhofer MP3 decoder is very high.
Personally I wouldn't go for this surround codec if I wanted to encode multichannel contents. The main reason for most people still using the MP3 format is its decent compatibility, not its quality, which is inferior to more modern formats at low bitrates. But even the compatibility argument isn't valid in this encoder's case, due to it being decoded in common stereo by most devices. AC3 is the most compatible multichannel solution, AAC and WMA 10 Professional provide supreme quality while still enjoying better 5.1 decoding support compared to MP3Surround.
QUOTE(Jimpin @ Jun 28 2007, 03:33)
Yes, i will, its the forum problem.
No, it's the user problem because obviously, you entered "yes" as question instead of choice.
Oh, yes, you right Sebastian Mares.
QUOTE(Junon @ Jun 28 2007, 12:49)
QUOTE(PHOYO @ Jun 28 2007, 00:00)
Why the hell should I have "surround" MP3 files? I want my MP3s to sound like original as much as possible.
Err... what about multichannel sources, e.g. Dolby Surround or 5.1 DVD-Audio? Or do you confuse the term "surround" with any pseudo-3D-effects added by some DSPs to simulate an areal sound?
Maybe I didn't completely understand your answer, but MP3 format doesn't support multichannel. For multichannel sources, I use different codec.
QUOTE(PHOYO @ Jun 28 2007, 19:28)
Maybe I didn't completely understand your answer, but MP3 format doesn't support multichannel.
From http://en.wikipedia.org/wiki/Mp3_surround
QUOTE
The encoder most likely creates MPEG-2 Layer 3 files in order to support multichannel sources.
QUOTE(Junon @ Jun 28 2007, 12:49)
Personally I wouldn't go for this surround codec if I wanted to encode multichannel contents. The main reason for most people still using the MP3 format is its decent compatibility, not its quality, which is inferior to more modern formats at low bitrates. But even the compatibility argument isn't valid in this encoder's case, due to it being decoded in common stereo by most devices. AC3 is the most compatible multichannel solution, AAC and WMA 10 Professional provide supreme quality while still enjoying better 5.1 decoding support compared to MP3Surround.
The situation could be different if FhG offered a DirectShow filter for MP3 surround decoding. Then the format could be used for creating compatible video soundtracks. The soundtrack would play on all standard systems as regular 2-channel MP3 and on a Windows PC it would be possible to play the multi-channel content with all DS players if the DS filter is installed. AFAIK, the only currently available video + MP3 surround player is DivX Player (www.divx.com).
Most likely codecs like AAC HE/LC, OGG Vorbis and WMA Pro provide better multi-channel audio quality at a similar bitrate (or at least better channel separation). However this has not been tested anywhere and no one has published personal test reports yet, so perhaps it would be better to not say anything about audio quality without providing proof. I could try some ABX testing and upload the decoded 5.1 wave test samples and post my test configuration. Unfortunately ABC/HR-Java does not work with multi-channel files so possibly the only usable test tool is foobar's ABX module.
QUOTE(Junon @ Jun 28 2007, 20:42)
QUOTE(PHOYO @ Jun 28 2007, 19:28)
Maybe I didn't completely understand your answer, but MP3 format doesn't support multichannel.
From http://en.wikipedia.org/wiki/Mp3_surround
QUOTE
The encoder most likely creates MPEG-2 Layer 3 files in order to support multichannel sources.
Standard decoders see the MP3S files as normal MPEG1 Layer 3 files. Encspot reports that the files use Joint Stereo mode. Only Fraunhofer's own decoder can extract the multi-channel content.
It has been said that the overhead is 16 kbps, so in theory a 192 kbps 5.1 channel MP3S file should provide better 2-channel stereo audio quality than a 160 kbps file that is encoded with a standard FhG encoder after downmixing the channels.
I knew it Alex B:
http://forums.winamp.com/showthread.php?threadid=273054
You must install this plugin for MP3 Surround files & if you want 2 use WINAMP
MP3 plugin, select the deactivate plugin in Fraunhofer IIS MP3 Surround Plugin option
And i converted stereo WAV & MP3 files 2 MP3 Surround & the quality is just stereo.
You must directly convert AC-3/DTS 5.1 files from DVDs 2 MP3 Surround to hear the real multi channel sound.
http://forums.winamp.com/showthread.php?threadid=273054
You must install this plugin for MP3 Surround files & if you want 2 use WINAMP
MP3 plugin, select the deactivate plugin in Fraunhofer IIS MP3 Surround Plugin option
And i converted stereo WAV & MP3 files 2 MP3 Surround & the quality is just stereo.
You must directly convert AC-3/DTS 5.1 files from DVDs 2 MP3 Surround to hear the real multi channel sound.
QUOTE(Jimpin @ Jun 28 2007, 21:30)
... You must directly convert AC-3/DTS files from DVDs 2 MP3 Surround to hear the real multi channel sound.
I too knew that. For my ABX test I would create some high quality 5.1 wave reference samples. Perhaps I can do that sometime during this summer.
BTW, I answered no to the poll question. However, the sitution has changed since then. I have now 54 tracks in MP3S format. I downloaded all available MP3S sample tracks from these addresses:
http://www.all4mp3.com/music/
ftp://ftp1.fraunhofer.de/institute/iis/am..._Demo_Songs.zip
Yes,you right 2, but MP3 Surround is useless for me.
Because the CD tracks are stereo.
Because the CD tracks are stereo.
QUOTE(Junon @ Jun 28 2007, 20:42)
QUOTE(PHOYO @ Jun 28 2007, 19:28)
Maybe I didn't completely understand your answer, but MP3 format doesn't support multichannel.
From http://en.wikipedia.org/wiki/Mp3_surround
QUOTE
The encoder most likely creates MPEG-2 Layer 3 files in order to support multichannel sources.
OK, I stand corrected. Just ignore my first post in this topic. I really thought it was DSP plugin or something.
QUOTE(Alex B @ Jun 28 2007, 20:10)
QUOTE(Junon @ Jun 28 2007, 20:42)
The encoder most likely creates MPEG-2 Layer 3 files in order to support multichannel sources.
Standard decoders see the MP3S files as normal MPEG1 Layer 3 files. Encspot reports that the files use Joint Stereo mode. Only Fraunhofer's own decoder can extract the multi-channel content.
It has been said that the overhead is 16 kbps, so in theory a 192 kbps 5.1 channel MP3S file should provide better 2-channel stereo audio quality than a 160 kbps file that is encoded with a standard FhG encoder after downmixing the channels.
The 5.1 code is saved outside the pre-existing frame data as "padding". This makes any mp3Surround file a standard MPEG-1 Layer-III file. The overhead is indeed 16kbps but in the case of CBR files the amount of padding increases to 32kbps (necessary if it wants to stay an MPEG-1 layer III file as only bitrates in 32kbps steps are supported by the specs). the 32kbps are in part 5.1 data and in part just filling bits.
A 5.1 channel MP3 will not provide better 2-channel stereo audio quality because it is actually a normal stereo MP3 with a third channel (hidden in the padding bits) for 5.1. That is why ANY Mp3-player can play these files without problems!
Hhhmm, interesting.
QUOTE(Ojay @ Jun 28 2007, 22:42)
The 5.1 code is saved outside the pre-existing frame data as "padding". This makes any mp3Surround file a standard MPEG-1 Layer-III file. The overhead is indeed 16kbps but in the case of CBR files the amount of padding increases to 32kbps (necessary if it wants to stay an MPEG-1 layer III file as only bitrates in 32kbps steps are supported by the specs). the 32kbps are in part 5.1 data and in part just filling bits.
A 5.1 channel MP3 will not provide better 2-channel stereo audio quality because it is actually a normal stereo MP3 with a third channel (hidden in the padding bits) for 5.1.
A 5.1 channel MP3 will not provide better 2-channel stereo audio quality because it is actually a normal stereo MP3 with a third channel (hidden in the padding bits) for 5.1.
At lower bitrates a 16 kbps step is possible, e.g. from 112 kbps to 128 kbps. However, my example was 192 kbps, so you are correct. If the overhead and padding consume 32 kbps then 192 kbps CBR MP3S should provide as good stereo quality as a 160 kbps standard stereo CBR file, not better quality.
However, I don't know if Fraunhofer provides any details the about channel downmix system that is used for the stereo part of the file. That info would be needed for creating a comparable 160 kbps stereo file that could be used in an ABX test.
I also wonder how much channel separation is still available after 6 discreet channels are encoded to MP3S and decoded back to 6 separate channels.
The new commandline MP3S encoder and decoder versions offer some interesting features:
http://www.all4mp3.com/tools/sw_fhg_cl.html#50
QUOTE
Whats New in Version 1.3:
New features - Encoder
OFL feature:
OFL (Original File Length) is a tool that puts information about the orignal file's length in the bitstream. A decoder that is aware of such OFL information can analyze it and then decode the bitstream having the same starting point and length as the original file (any codec delay or flushing at the end will be removed).
"-ofl" is the relevant switch.
VBR encoding:
the encoder can now produce bitstreams encoded with variable bitrate. Therefore five modes are provided that can be invoked with the "-m"-switch.
"-m 1..5" (where "1" represents highest quality and "5" lowest average bitrate).
Note: When encoding in VBR mode, "-br 0" must be used. CBR encoding is applied by using "-m 0" and the desired bitrate, e.g. "-br 128000" for 128kbit/s
VBRI header:
the encoder can prepend information about a VBR encoded bitstream by producing a so-called VBRI header (VBRI = VBR Index table).
"-vbri" is the relevant switch.
New features - Decoder
piping:
the decoder can read data from stdin and also write to stdout.
to read from stdin, "-if -" has to be used as input parameter,
to write to stdout, "-of -" is used accordingly.
decoding mp3sx bitstream:
the actual free command line decoder is now capable to decode
bitstreams that were produced with the mp3 sx converter tool.
If I understand this correctly the OFL feature provides info for gapless decoding.
Is MP3 Surround quality better than AAC, Ogg Vorbis, WMA multi channel?
QUOTE(Jimpin @ Jun 28 2007, 23:47)
Is MP3 Surround quality better than AAC, Ogg Vorbis, WMA multi channel?
We are not allowed to claim anything without providing objektive support.
A quote from HA's Terms Of Service:
QUOTE
8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support.
In any case, testing multi-channel files would be quite a bit more complicated than testing standard stereo files. I may try a personal test, but the usefulness of the test results would be quite limited anyway.
Some examples of the limiting factors:
My personal preferences, the used speaker system & room acoustics combination, the too small amount of different test samples and the selected encoding options (there is very little old knowledge available and I can't test all setting combinations and bitrates or quality levels).
Ok, right...
I found this interesting paper from fraunhofer's site:
An Introduction To MP3 Surround (pdf)
It contains a listening test report too.
An Introduction To MP3 Surround (pdf)
It contains a listening test report too.
Oh, the quality of Fraunhofer IIS MP3 Surround Plugin is very better than Nullsoft plugin.
goddammit...
QUOTE(Jimpin @ Jun 29 2007, 11:12)
Oh, the quality of Fraunhofer IIS MP3 Surround Plugin is very better than Nullsoft plugin.
That's funny, because the underlying C/C++ code for the 'normal' (non-surround) MP3 decoder for these two plugins are line-for-line identical.
QUOTE(Jimpin @ Jun 28 2007, 10:58)
Yes,you right 2, but MP3 Surround is useless for me.
Because the CD tracks are stereo.
Thank You...Duh!
The guy has been banned, people...
Let's forget this thread.
Let's forget this thread.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.