Perhaps some American could explain to me why so many of you have such a problem with left of centre politics. There are many countries in the world where 'communist' isn't an insult -- why is it one in America?
Ask the people in Sweden for example, how they balance their very high taxes against their incredible high quality of life. I'd prefer to be taxed and happy than untaxed and disease ridden. I live in the UK, and as far as I'm concerned it doesn't tax enough -- and what it does tax is in the wrong areas (VAT -- sales tax -- is, relatively, a much bigger tax on the poor than on the rich).
The problem with Postscript is also that it is turing complete -- leading to the possibility of malicious code.
I'm in favour of domain specific languages being non-Turing complete -- it makes them simple to analyse, and much easier to optimise (the classic example being SQL).
If it wasn't based around the bloated monstrosity that is XML, I'd suggest SVG as an interesting alternative to DPS as a good way for a graphic layer to describe its objects.
My partner is a secondary school teacher here in the UK, and the only sites which kids can access are those which have been pre-approved. I find it incredible that any school would do otherwise.
Mozilla has much better compatibility than Konqueror, but the KDE team have about 1% of the resources of Netscape, and with that 1% have managed to produce a browser with 95% of the features of Mozilla.
They are slowly working toward implementing all the CSS2 features -- but there are only 24 hours in the day, and they are much more interested in implementing the features which people actually use.
If you believe that the three features you mention are 'fundamental', then submit bug reports to the KDE site (bugs.kde.org), together with examples of sites which Konqueror fails to display because of a lack of support for these features. A similar burst of bug reports happened after KDE 2.1, which showed the team which areas to work on next.
Additionally, GCC 3.1 will not be binary compatible with GCC 3.0 (for C++ anyhow), so it's not a bad idea to wait. Sadly, I can see Red Hat releasing a new major release (8? it's up to 8 already?) with GCC 3.0 as the compiler, a couple of months before GCC 3.1 is released, purely to keep their continual upgrade cycle going. We'd then have to wait for ANOTHER release cycle for a compiler with (please please please god) precompiled headers and a stable ABI.
It doesn't work with Netscape 4, but if you're interested in seeing a chess playing program in 5k of ECMAscript, take a look at
this entry to this years 5K website competition (there are some other stunningly creative sites on there, as well).
Please tell me you meant "...store the audio on disk in WAV format..." and not "...store the audio on disk in WMA format..."!
If you want the best audio quality, then forget about lossy compression, and get one of the lossless compressors (FLAC for example) which can compress WAVs into about 50% of normal size -- about equivalent to a 700 kpbs file.
Yes, disc space is much cheaper now: cheap enough that I can affort to set aside about 15 Gig for my music collection. If I stored my CDs completely uncompressed, I would be able to fit about 25 CDs in that space, 30 if I losslessly compress them. Using Vorbis at a bitrate which isn't transparent, but good enough that my ears don't hurt when I listen to the music (anything from 96 to 160 depending on the type of music, although I've not found anything I can distingush 128 and 160 between on my speakers yet), I can fit 300 CDs.
You sound like you have very discerning hearing (unlike me -- too much standing near the front at discos and concerts:-). Why don't you find a section of the music which Vorbis encodes badly, and tell the developers about it? This isn't the final release, just a tuning release, and they're always pathetically grateful to find people who can find cases where the encoder fails *and explain what sounds wrong*.
There are quite a few cropping up in the #vorbis IRC channel as we speak.
Yes, sorry about that - I should have mentioned that there is still some tuning still to be done. That said, RC2 is better at encoding everything I've tried at every bitrate I've tried than beta4. Given that a lot of people have been discussing Ogg Vorbis based in very old versions of the encoder (beta2 even), I was just concerned with getting people to try the newer version.
And hey, mass market testing is what the encoder needs...:)
I'm amazed you can still hear anything after sitting close to the brass section:). We just play as loud as we can to make up for the boredom of counting yet another 250 bars rest...
If we want a decent language for the new millennium, we should go for
APL. Fast, concise, and an ideal teaching language (note that the link is to APlus, a blend of APL and C which is apparently used by the Morgan Stanley investment bank). True, modern computers don't come with keyboards with enough keys to program it, but we can overcome that with overlays.
Alternatively, I'd love to see someone try web scripting in Brainfuck (advantages: simple, clear, easy to write an interpreter), or
Befunge (break out of the depressing paradigms of modern languages! Truely adds a new dimension to program code).
The payload is a random file from their computer, with the virus tacked onto the front. Remove the first however many (about 128K) bytes, and you get a peek into the world of an idiot that clicks on everything they are sent via email.
Sadly, nothing I've been sent by SirCam has been interesting.
if they are songs he downloaded, what choice does he have?
The best choice he has is to leave the MP3s exactly as they are. The last thing I want is for people to go around converting their MP3s to Vorbis. The two formats can co-exist perfectly. i know I have both.OGG and.MP3 files kicking around on my computer.
Do you have a short segment in WAV format that you claim is much better encoded to MP3 than OGG? If so, I'm sure the Vorbis developers would be eager to hear from you, and given that the 1.0 encoder isn't out for a couple of weeks, you've got a chance to improve the quality of Vorbis for everyone.
The release candidate for version 1 of Vorbis came out a while back, and should play back any Ogg files you find out there, even those encoded with beta releases (the release candidate of the encoder is still not finished, but hopefully will be released in the next couple of weeks).
So no, you won't HAVE to rerip all your files.
On the other hand, you might WANT to rerip them: the beta 4 encoder has no channel coupling, whereas the 1.0 encoder will. What does this mean? At least 10% smaller stereo files (or better sound quality for the same file size). There are also some sound quality issues with beta 4 which have apparently been fixed in the unreleased encoder.
Beta 4 is already on a par with MP3. The final encoder will tip the balance.
I also would change my entire collection to.ogg if it was supported in the same players/encoders as MP3 is.
I don't know what players and encoders you are using, but I can tell you that the ones I use in both Windows and Linux support Vorbis, and have done for quite a while.
Yes, Ogg is the name of the general stream specification (just like QuickTime or AVI). Vorbis is just the first codec developed which interfaces with Ogg: others under consideration/development include Squish (a lossless audio compressor) and Tarkin (patent free video codec).
In theory you can wrap any file format you like inside an Ogg stream, and conversely you could have a Vorbis file which isn't in an Ogg stream (although this would probably confuse every player out there:).
LAME is the best MP3 encoder available. It's free, open source, and multiplatform.
On Windows, the best ripper/encoder is
CDex.
On Linux, look for packages for Lame, CDParanoia and Grip for your distribution.
For discussion of the best MP3 encoders, take at look at
R3Mix (and in
particular at the forums).
Re:KDE Rocks!
on
KDE 2.2 Tagged
·
· Score: 3, Informative
KDE 3.0 will basically be what you would imagine KDE 2.3 would be (i.e. no world shaking new changes), but ported to QT 3. This will enable much better handling of 'foreign' languages, and a rewritten styling/themeing engine, plus other extras (data aware widgets for example).
Re:I'm a disappointed GNOME user...
on
KDE 2.2 Tagged
·
· Score: 1
Just a little note to say that Sawfish and KDE work together quite well:) All KDE wants is a windowmanager that supports the new window manager specification developed by both KDE, GNOME, and other window manager developers.
MP3Pro claims to be transparent at 64 kb/s, but let's keep in mind what they're doing - the encoder first does a lowpass at 10khz, and the decoder guesses to try to create those frequencies again
My god - *10* KHz? I thought it was 15... that explains why MP3Pro has sounded so god-awful when I've tried it.
Not only that, but with encoder 1.0, I should be able to bring that bitrate down to 96, and for certain music, possibly less if we're lucky. Think about that, 96 kb/s transparent stereo audio. Absolutely amazing.
It won't be quite as amazing as you think -- the current plans are for the default behaviour at 96kbps to include a 16 kbps lowpass filter. Mind you, this won't be noticable on standard pop music, or by people with bog standard ears:), but you'll need to use 112 at least if that matters to you. I'd dispute your claims about 'transparency' as well... but it's certainly good enough for listening on my speakers:)
Why aren't you just using CDex as a 1 step process to go from CD to.OGG? It includes the Vorbis beta 4 encoder, and recent betas include CDParanoia for ripping, so you don't need to use Exact Audio Copy either.
You sound like the sort of person that would be more interested in LPAC or Monkeys Audio: lossless compression of WAVs (generally to about 1/2 the original file size). 2:1 compression and absolutely no loss of quality.
One of the projects for Xiph after they release version 1.0 of Vorbis is Ogg Squish, which will be a lossless audio compression format, wrapped in an.OGG stream just like Vorbis.
Did you check that these 'artifacts' weren't present in the.wav file you ripped from before encoding? I've yet to find a file which Vorbis adds any *significant* artifacts to (although there are some issues with pre-echo in beta 4).
To make sure that you've ripped your CDs correctly, use a decent ripper like CDParanoia.
Slashdot Mirror
Perhaps some American could explain to me why so many of you have such a problem with left of centre politics. There are many countries in the world where 'communist' isn't an insult -- why is it one in America?
Ask the people in Sweden for example, how they balance their very high taxes against their incredible high quality of life. I'd prefer to be taxed and happy than untaxed and disease ridden. I live in the UK, and as far as I'm concerned it doesn't tax enough -- and what it does tax is in the wrong areas (VAT -- sales tax -- is, relatively, a much bigger tax on the poor than on the rich).
No, sane politicians do exist. The oxymoron is 'honest politician'.
The problem with Postscript is also that it is turing complete -- leading to the possibility of malicious code.
I'm in favour of domain specific languages being non-Turing complete -- it makes them simple to analyse, and much easier to optimise (the classic example being SQL).
If it wasn't based around the bloated monstrosity that is XML, I'd suggest SVG as an interesting alternative to DPS as a good way for a graphic layer to describe its objects.
'law' no, 'policy' yes.
My partner is a secondary school teacher here in the UK, and the only sites which kids can access are those which have been pre-approved. I find it incredible that any school would do otherwise.
On what grounds was this comment modded up on?
Probably because it's correct. You might not like it, but school resources are there for learning, not for hacking or playing stupid flash games.
Schools should have a policy where the only sites students can access are those which have been positively vetted by a teacher.
Mozilla has much better compatibility than Konqueror, but the KDE team have about 1% of the resources of Netscape, and with that 1% have managed to produce a browser with 95% of the features of Mozilla.
They are slowly working toward implementing all the CSS2 features -- but there are only 24 hours in the day, and they are much more interested in implementing the features which people actually use.
If you believe that the three features you mention are 'fundamental', then submit bug reports to the KDE site (bugs.kde.org), together with examples of sites which Konqueror fails to display because of a lack of support for these features. A similar burst of bug reports happened after KDE 2.1, which showed the team which areas to work on next.
Additionally, GCC 3.1 will not be binary compatible with GCC 3.0 (for C++ anyhow), so it's not a bad idea to wait. Sadly, I can see Red Hat releasing a new major release (8? it's up to 8 already?) with GCC 3.0 as the compiler, a couple of months before GCC 3.1 is released, purely to keep their continual upgrade cycle going. We'd then have to wait for ANOTHER release cycle for a compiler with (please please please god) precompiled headers and a stable ABI.
It doesn't work with Netscape 4, but if you're interested in seeing a chess playing program in 5k of ECMAscript, take a look at this entry to this years 5K website competition (there are some other stunningly creative sites on there, as well).
Please tell me you meant "...store the audio on disk in WAV format..." and not "...store the audio on disk in WMA format..."!
If you want the best audio quality, then forget about lossy compression, and get one of the lossless compressors (FLAC for example) which can compress WAVs into about 50% of normal size -- about equivalent to a 700 kpbs file.
Yes, disc space is much cheaper now: cheap enough that I can affort to set aside about 15 Gig for my music collection. If I stored my CDs completely uncompressed, I would be able to fit about 25 CDs in that space, 30 if I losslessly compress them. Using Vorbis at a bitrate which isn't transparent, but good enough that my ears don't hurt when I listen to the music (anything from 96 to 160 depending on the type of music, although I've not found anything I can distingush 128 and 160 between on my speakers yet), I can fit 300 CDs.
You sound like you have very discerning hearing (unlike me -- too much standing near the front at discos and concerts :-). Why don't you find a section of the music which Vorbis encodes badly, and tell the developers about it? This isn't the final release, just a tuning release, and they're always pathetically grateful to find people who can find cases where the encoder fails *and explain what sounds wrong*.
There are quite a few cropping up in the #vorbis IRC channel as we speak.
Yes, sorry about that - I should have mentioned that there is still some tuning still to be done. That said, RC2 is better at encoding everything I've tried at every bitrate I've tried than beta4. Given that a lot of people have been discussing Ogg Vorbis based in very old versions of the encoder (beta2 even), I was just concerned with getting people to try the newer version.
:)
And hey, mass market testing is what the encoder needs...
I'm amazed you can still hear anything after sitting close to the brass section :). We just play as loud as we can to make up for the boredom of counting yet another 250 bars rest...
If we want a decent language for the new millennium, we should go for APL. Fast, concise, and an ideal teaching language (note that the link is to APlus, a blend of APL and C which is apparently used by the Morgan Stanley investment bank). True, modern computers don't come with keyboards with enough keys to program it, but we can overcome that with overlays.
Alternatively, I'd love to see someone try web scripting in Brainfuck (advantages: simple, clear, easy to write an interpreter), or Befunge (break out of the depressing paradigms of modern languages! Truely adds a new dimension to program code).
The payload is a random file from their computer, with the virus tacked onto the front. Remove the first however many (about 128K) bytes, and you get a peek into the world of an idiot that clicks on everything they are sent via email.
Sadly, nothing I've been sent by SirCam has been interesting.
if they are songs he downloaded, what choice does he have?
.OGG and .MP3 files kicking around on my computer.
The best choice he has is to leave the MP3s exactly as they are. The last thing I want is for people to go around converting their MP3s to Vorbis. The two formats can co-exist perfectly. i know I have both
Do you have a short segment in WAV format that you claim is much better encoded to MP3 than OGG? If so, I'm sure the Vorbis developers would be eager to hear from you, and given that the 1.0 encoder isn't out for a couple of weeks, you've got a chance to improve the quality of Vorbis for everyone.
Or are you just mindlessly trolling?
The release candidate for version 1 of Vorbis came out a while back, and should play back any Ogg files you find out there, even those encoded with beta releases (the release candidate of the encoder is still not finished, but hopefully will be released in the next couple of weeks).
.ogg if it was supported in the same players/encoders as MP3 is.
So no, you won't HAVE to rerip all your files.
On the other hand, you might WANT to rerip them: the beta 4 encoder has no channel coupling, whereas the 1.0 encoder will. What does this mean? At least 10% smaller stereo files (or better sound quality for the same file size). There are also some sound quality issues with beta 4 which have apparently been fixed in the unreleased encoder.
Beta 4 is already on a par with MP3. The final encoder will tip the balance.
I also would change my entire collection to
I don't know what players and encoders you are using, but I can tell you that the ones I use in both Windows and Linux support Vorbis, and have done for quite a while.
Yes, Ogg is the name of the general stream specification (just like QuickTime or AVI). Vorbis is just the first codec developed which interfaces with Ogg: others under consideration/development include Squish (a lossless audio compressor) and Tarkin (patent free video codec).
:).
In theory you can wrap any file format you like inside an Ogg stream, and conversely you could have a Vorbis file which isn't in an Ogg stream (although this would probably confuse every player out there
LAME is the best MP3 encoder available. It's free, open source, and multiplatform.
On Windows, the best ripper/encoder is CDex. On Linux, look for packages for Lame, CDParanoia and Grip for your distribution.
For discussion of the best MP3 encoders, take at look at R3Mix (and in particular at the forums).
KDE 3.0 will basically be what you would imagine KDE 2.3 would be (i.e. no world shaking new changes), but ported to QT 3. This will enable much better handling of 'foreign' languages, and a rewritten styling/themeing engine, plus other extras (data aware widgets for example).
Just a little note to say that Sawfish and KDE work together quite well :) All KDE wants is a windowmanager that supports the new window manager specification developed by both KDE, GNOME, and other window manager developers.
MP3Pro claims to be transparent at 64 kb/s, but let's keep in mind what they're doing - the encoder first does a lowpass at 10khz, and the decoder guesses to try to create those frequencies again
My god - *10* KHz? I thought it was 15... that explains why MP3Pro has sounded so god-awful when I've tried it.
Not only that, but with encoder 1.0, I should be able to bring that bitrate down to 96, and for certain music, possibly less if we're lucky. Think about that, 96 kb/s transparent stereo audio. Absolutely amazing.
It won't be quite as amazing as you think -- the current plans are for the default behaviour at 96kbps to include a 16 kbps lowpass filter. Mind you, this won't be noticable on standard pop music, or by people with bog standard ears
Why aren't you just using CDex as a 1 step process to go from CD to .OGG? It includes the Vorbis beta 4 encoder, and recent betas include CDParanoia for ripping, so you don't need to use Exact Audio Copy either.
You sound like the sort of person that would be more interested in LPAC or Monkeys Audio: lossless compression of WAVs (generally to about 1/2 the original file size). 2:1 compression and absolutely no loss of quality.
.OGG stream just like Vorbis.
One of the projects for Xiph after they release version 1.0 of Vorbis is Ogg Squish, which will be a lossless audio compression format, wrapped in an
Did you check that these 'artifacts' weren't present in the .wav file you ripped from before encoding? I've yet to find a file which Vorbis adds any *significant* artifacts to (although there are some issues with pre-echo in beta 4).
To make sure that you've ripped your CDs correctly, use a decent ripper like CDParanoia.