If you compiled the Vorbis source yourself, you may have run into a bug *with old versions of GCC* (up to EGCS 1.1.2/GCC 2.91.*) which leads to Vorbis compiling incorrectly, and producing files that sound like shit.
Check the Vorbis site for more information... the only correction is to upgrade to a more recent version of GCC, or (if you are using RedHat) to download the fixed up RPMs they have created.
FLAC is still working on the same track. I made it as aggressive as it would go, since a one-time CPU cost is worth a long-term storage savings
While laudable, this isn't a good idea with FLAC. The 'compress me harder option' -9 is about 10 times slower than the next one down, -8, and will normally save me about an extra 0.1%. Generally, you will get very little extra compression over -6 or so (which I what I use) -- so please don't get disheartened, just use a more realistic compression setting.
Like it or not, 128 kb/s is the rate that 90% of the people out there use to make music... And it's just as relevant to ask for the best 128k encoder as it is to ask for the best 64k encoder or 256k encoder.
There are other tests concentrating on the high end (for example R3Mix recently did a blind test comparing 8 >=192k LAME settings... one of the main results being that 192k is *not* transparent, and that VBR is better than CBR at the same bitrate).
The 'not perfect' was due to the sound systems & testing environments of the testers not being controlled -- but there is nothing you can do about that unless paying thousands for a rigorous trial at a proper lab.
Vorbis as a whole isn't at version 1, but both the file specification and the decoder *are*. This means that you can make a hardware decoder now, and it won't need to be updated until version 2 (like MP3 and MP3Pro).
Re-encoding is a good idea. *Transcoding* is a bad idea.
but I encode all my albums into Mp3s using VBR at "100%" quality
This doesn't make any sense -- "100% quality"? What encoder are you using?
does Ogg have variable bit rate capability?
Yes in a big way -- Ogg doesn't really have *constant* bit rate capability:) (although it will be able to be *more* constant than currently in RC3).
I think we've decided that "size doesn't matter".
Size still does matter. With FLAC, I'll get a lossless file about 60% of the original. With Ogg, I can get a file that sounds almost perfect about 15% of the size of the original. This means I can fit 4 more albums into the same size -- useful for storing all 400 albums on the hard disc so I can random play them.
Here is the link to the 'Official' Vorbis plugin for WinAmp. Despite the other message, it's still not included by default in a WinAmp install, but hopefully should be in the next version.
Instead of encoding MP3s at CBR 256, you should consider using one of the LAME MP3 presets designed for people like you.
The latest alpha versions of LAME have three presets from Dibrom: --dm-preset standard, extreme, and insane. These tune the parameters LAME uses for audiophines -- extreme will probably give you ~250kb/s, and tests indicate that they can actually sound better than CBR 256.
You get almost the same quality at half the size with WMA
Sadly, you can't. Listening tests have shown that WMA 8 has sacrified sound quality at medium/high bitrates over WMA 7 to improve quality at the low end. So it's great for music over a modem, but at 96k and above it is no better than Ogg Vorbis.
The tests that you link to were done incredibly badly, and should just be ignored.
Here is a test that, although not perfect, was at least semi-blind. The conclusions: at that bitrate, MPC ('MPegPlus' not 'MP3Pro') and AAC were the best, followed by LAME MP3, OGG & WMA8 all together, and finally the very worst was XING encoded MP3.
This test was run with Vorbis RC2. RC3 will be out in a week, with much improved noise masking. For a taste of RC3, you can check out the Vorbis CVS, which includes most of the RC3 fixes but encodes at a fixed rate of 128 kp/ps. This raises the low pass, improves the noise masking, and the stereo channel coupling code.
You must be using Windows.
Monkey's Audio is a Windows only format... and you should not trust your data to a single-OS file format (yes, I would say the same for Linux-only file formats). Use FLAC instead.
MPC has better licensing than Monkey's Audio: the *decoder* is open source (GPL even), so you will always be able to decode your music. *encoding* is only possible on Windows however (although there is an older binary version available for i386 Linux systems), and the encoder will be made shareware in the near future. This is a real pity, because tests have shown that even at 128kp/s MPC is up there with AAC (MPEG-4 audio).
There is a very good comparison of lossless audio encoders
here.
The 'shorten' format has some problems, such as seeking when playing. I'd advise you to use the FLAC format. It has all the plugins that Shorten has, but has better compression, and inbuilt support for seeking and streaming.
If you don't have a spare friend, you can use the ABX testing method to see if you can distinguish between two files. Take a file, compress it, save to a WAV, and then give the files to the ABX program, which acts as your spare friend:)
If you're running Windows, you can get ABX from
http://www.pcabx.com/. On UNIX systems, the LAME source code comes with an ABX program (in the misc/ directory, I think).
Here is an example of a test that took place using a slightly different testing methodology, more akin to MUSHRA (which is used to evaluate lots of encoders at the same time):
http://www.ff123.net/128tests.html.
I have a decent OS -- distantly based on a Mandrake 7 CD I installed a while back. That's what I do all my work on.
Windows is for games & my partner's work (she 'has' to use Office and write (shudder) Visual Basic programs).
And quite frankly, I'm not interesting in the roots of the problems my Windows installation has. Someone asked for an example of a IE bug, and I gave one.
Whenever I'm using Windows (my 'wife' has to for compatibility with her school) and I'm moderating on Slashdot, IE gets very confused about the placement of all the boxes, and completely fucks up the rendering... not to mention it eats all available memory up and won't let me open another browser window until I've closed the Slashdot window. This happens with IE 5.5 & 6 on Windows 98 SE.
Re:Its not faster than MOziilla, its not more stab
on
Mozilla 0.9.5
·
· Score: 1
Anyone know how Linus' book is selling? According to
http://www.amazon.co.uk/exec/obidos/ASIN/1587990 80 6/o/qid=1002929163/sr=2-2/ref=sr_bt_2/202-9281352- 4689406
'Just For Fun: The Story of an Accidental Revolutionary' by Linus Torvalds, David Diamond, is the 3769th best selling book in Amazon UK.
So a real mass market success...
Unless you're coming to talk about and demonstrate an illegal (in the US) thing you did or sell tons of pirated copies of Windoze and Office, you don't have anything to fear. Yet, at least.
From the Adobe copy protection debacle, it looks like you can be arrested in the US for doing something that's legal where you live, but illegal in the US... this is an interesting precedent.
I'm looking forward to the UK government arresting any US tourists that carry concealed weapons when they are at home in the US.
Looks like everything is still better value in the Americas...
Over here in the UK it's going to cost me £40 (that's about $55) per month to get NTL Cable, with 512k down & 128k up. That's the only broadband option -- ADSL is being completely mismanaged by BT.
But then again, from the sounds of it, the people over here could actually have the right idea about pricing -- at least they're not all going out of business.
Wow... are you lot still talking about 'invisible pink unicorns'? That was the 'god' substitute of choice on alt.atheism back in 1994 when I was a regular...
Most importantly, layout is all important. HTML was designed with the purpose of leaving layout of the document at the ultimate discretion of the browser. That's bad, especially if you have to save your mathematical formulae as PNGs in order to get them to display properly.
No, layout isn't import -- content is important. You have just argued very well for making papers available in TeX format... because then we can easily take chunks of information and use them easily.
Slashdot Mirror
If you compiled the Vorbis source yourself, you may have run into a bug *with old versions of GCC* (up to EGCS 1.1.2/GCC 2.91.*) which leads to Vorbis compiling incorrectly, and producing files that sound like shit.
Check the Vorbis site for more information... the only correction is to upgrade to a more recent version of GCC, or (if you are using RedHat) to download the fixed up RPMs they have created.
On Windows use either CDex or EAC. On UNIX systems use CDParanoia. I used to provide links, but this time I'll let you use a search engine.
FLAC is still working on the same track. I made it as aggressive as it would go, since a one-time CPU cost is worth a long-term storage savings
While laudable, this isn't a good idea with FLAC. The 'compress me harder option' -9 is about 10 times slower than the next one down, -8, and will normally save me about an extra 0.1%. Generally, you will get very little extra compression over -6 or so (which I what I use) -- so please don't get disheartened, just use a more realistic compression setting.
Like it or not, 128 kb/s is the rate that 90% of the people out there use to make music... And it's just as relevant to ask for the best 128k encoder as it is to ask for the best 64k encoder or 256k encoder.
There are other tests concentrating on the high end (for example R3Mix recently did a blind test comparing 8 >=192k LAME settings... one of the main results being that 192k is *not* transparent, and that VBR is better than CBR at the same bitrate).
The 'not perfect' was due to the sound systems & testing environments of the testers not being controlled -- but there is nothing you can do about that unless paying thousands for a rigorous trial at a proper lab.
Vorbis as a whole isn't at version 1, but both the file specification and the decoder *are*. This means that you can make a hardware decoder now, and it won't need to be updated until version 2 (like MP3 and MP3Pro).
Re-encoding is a good idea. *Transcoding* is a bad idea.
No, go with FLAC (or at least 'Shorten').
Monkey's Audio is Windows only, and it is a very stupid idea to archive data into a Windows-only format.
but I encode all my albums into Mp3s using VBR at "100%" quality
:) (although it will be able to be *more* constant than currently in RC3).
This doesn't make any sense -- "100% quality"? What encoder are you using?
does Ogg have variable bit rate capability?
Yes in a big way -- Ogg doesn't really have *constant* bit rate capability
I think we've decided that "size doesn't matter".
Size still does matter. With FLAC, I'll get a lossless file about 60% of the original. With Ogg, I can get a file that sounds almost perfect about 15% of the size of the original. This means I can fit 4 more albums into the same size -- useful for storing all 400 albums on the hard disc so I can random play them.
Here is the link to the 'Official' Vorbis plugin for WinAmp. Despite the other message, it's still not included by default in a WinAmp install, but hopefully should be in the next version.
Instead of encoding MP3s at CBR 256, you should consider using one of the LAME MP3 presets designed for people like you.
The latest alpha versions of LAME have three presets from Dibrom: --dm-preset standard, extreme, and insane. These tune the parameters LAME uses for audiophines -- extreme will probably give you ~250kb/s, and tests indicate that they can actually sound better than CBR 256.
You get almost the same quality at half the size with WMA
Sadly, you can't. Listening tests have shown that WMA 8 has sacrified sound quality at medium/high bitrates over WMA 7 to improve quality at the low end. So it's great for music over a modem, but at 96k and above it is no better than Ogg Vorbis.
The tests that you link to were done incredibly badly, and should just be ignored.
Here is a test that, although not perfect, was at least semi-blind. The conclusions: at that bitrate, MPC ('MPegPlus' not 'MP3Pro') and AAC were the best, followed by LAME MP3, OGG & WMA8 all together, and finally the very worst was XING encoded MP3.
This test was run with Vorbis RC2. RC3 will be out in a week, with much improved noise masking. For a taste of RC3, you can check out the Vorbis CVS, which includes most of the RC3 fixes but encodes at a fixed rate of 128 kp/ps. This raises the low pass, improves the noise masking, and the stereo channel coupling code.
MPC has better licensing than Monkey's Audio: the *decoder* is open source (GPL even), so you will always be able to decode your music. *encoding* is only possible on Windows however (although there is an older binary version available for i386 Linux systems), and the encoder will be made shareware in the near future. This is a real pity, because tests have shown that even at 128kp/s MPC is up there with AAC (MPEG-4 audio).
There is a very good comparison of lossless audio encoders here. The 'shorten' format has some problems, such as seeking when playing. I'd advise you to use the FLAC format. It has all the plugins that Shorten has, but has better compression, and inbuilt support for seeking and streaming.
If you're running Windows, you can get ABX from http://www.pcabx.com/. On UNIX systems, the LAME source code comes with an ABX program (in the misc/ directory, I think).
Here is an example of a test that took place using a slightly different testing methodology, more akin to MUSHRA (which is used to evaluate lots of encoders at the same time): http://www.ff123.net/128tests.html.
I have a decent OS -- distantly based on a Mandrake 7 CD I installed a while back. That's what I do all my work on.
Windows is for games & my partner's work (she 'has' to use Office and write (shudder) Visual Basic programs).
And quite frankly, I'm not interesting in the roots of the problems my Windows installation has. Someone asked for an example of a IE bug, and I gave one.
256 meg not enough memory to use a web browser these days?
And when I need to upgrade, it certainly won't be to NT.
Whenever I'm using Windows (my 'wife' has to for compatibility with her school) and I'm moderating on Slashdot, IE gets very confused about the placement of all the boxes, and completely fucks up the rendering... not to mention it eats all available memory up and won't let me open another browser window until I've closed the Slashdot window. This happens with IE 5.5 & 6 on Windows 98 SE.
but MOzilla is faster,
wrong
more powerful
wrong
and more stable
wrong again.
Plus Konq is just or linux
wrong
and KDE at that
at last he almost gets one right.
For example, take a look at APLus, a GPLed language created at Morgan Stanley and derived from APL.
For more information about J (another offshoot of ALP) and APL, take a look at J\APL - the journal of APL.
Anyone know how Linus' book is selling?0 80 6/o/qid=1002929163/sr=2-2/ref=sr_bt_2/202-9281352- 4689406
According to
http://www.amazon.co.uk/exec/obidos/ASIN/158799
'Just For Fun: The Story of an Accidental Revolutionary' by Linus Torvalds, David Diamond, is the 3769th best selling book in Amazon UK.
So a real mass market success...
Unless you're coming to talk about and demonstrate an illegal (in the US) thing you did or sell tons of pirated copies of Windoze and Office, you don't have anything to fear. Yet, at least.
From the Adobe copy protection debacle, it looks like you can be arrested in the US for doing something that's legal where you live, but illegal in the US... this is an interesting precedent.
I'm looking forward to the UK government arresting any US tourists that carry concealed weapons when they are at home in the US.
Looks like everything is still better value in the Americas...
Over here in the UK it's going to cost me £40 (that's about $55) per month to get NTL Cable, with 512k down & 128k up. That's the only broadband option -- ADSL is being completely mismanaged by BT.
But then again, from the sounds of it, the people over here could actually have the right idea about pricing -- at least they're not all going out of business.
Wow... are you lot still talking about 'invisible pink unicorns'? That was the 'god' substitute of choice on alt.atheism back in 1994 when I was a regular...
Some things just keep on rolling...
or forking out for a individual copy from the journal [£5 a shot in the UK]
Only £5? Loads of the journals I read have a $35 article copy/download fee these days... (the fees have skyrocketed in the last couple of years)
Most importantly, layout is all important. HTML was designed with the purpose of leaving layout of the document at the ultimate discretion of the browser. That's bad, especially if you have to save your mathematical formulae as PNGs in order to get them to display properly.
No, layout isn't import -- content is important. You have just argued very well for making papers available in TeX format... because then we can easily take chunks of information and use them easily.