1) AC3 *is* lossy
2) 350kpbs *is* lossy from a CD input.
If you want lossless compression of audio, you'll generally be able to get down no further than about 60% of the original file size (approx 850kpbs for a standard stereo CD input). Anything less than that *has* to be lossy.
Lossy isn't a bad thing -- the whole point of audio compression is to remove the parts of the sound which the human ear can't hear. What is debatable is at what point this 'transparency' happens with different encoders (some would argue that no MP3 is ever transparent. These tend to be people that will only listen to vinyl via valve amps).
RC2 will go down to 64kpbs on stereo 44.1kHz input.
RC3 will go lower, although it's not tuned for less than 64kpbs (56kpbs sounds ok, 48kpbs ok on easy to encode music, less than that and you start getting random underwater bubble sounds).
RC3 will also go *higher* than RC2 can: its highest quality level is nominally 480 kpbs, and you can set a higher rate than that manually.
Vorbis can't really go below 64kpbs (although 56kpbs is sounding quite decent with RC3) and still sound decent *at 44.1kHz*. Resample the input to 22kHz, say, and you'll be able to get quite a bit lower.
Note that RC3 will be more of an experimental release than RC2, in that it will only have full support for 44.1kHz.
Here is KDE's version increase plan:
a.b.c
'a' gets incremented when it will be binary incompatible with the previous version.
'b' gets incremented when there are additional binary compatible features
'c' gets incremented when bugs are fixed (although some features slip in from time to time)
Now, I don't know if Gnome have broken binary compatibility 7 times, as you imply. If they have, that's not something to be proud of.
Are you sure you haven't cut and pasted this from a discussion before KDE 2 came out?
CORBA is a nice idea, but grossly over designed. DCOP + KParts work exceptionally well, as evidenced by the ease with which they have been extending and developing Konqueror this last year. The KDE team are open to using CORBA, but only once GNOME have shown that it is an actual viable solution.
Multi-language support is already good in KDE 2, and will be better in KDE 3.
GNOME may be the best desktop alternative in the long term, but it certainly isn't in the short term, and I have my doubts about the medium term as well.
Are you from some odd parallel universe where the tests I linked to were COMPLETELY DIFFERENT? There is no anti-Microsoft bias or pro-Ogg bias in those tests at all.
The tests were blind for the listeners (they knew which formats were being tested, but not which files corresponded to which format), unlike the Washington Post study, which told the listeners *before they listened* what the format was. FF123's test is also comparing quality at the same (as far as possible within the format) bitrate, unlike the Washington Post study, which used different bitrates for the different formats. He has also provided you with the raw data returned from the listeners, so you can double check that he has not influenced the results.
Also, the graphs were in the 'informal analysis' section: the formal analysis concluded that the codecs in the first test were divided into three groups (from best to worst):
1) AAC and MPC
2) LAME encoded MP3, Ogg Vorbis RC2, WMA8
3) Xing encoded MP3
and this was arrived at via a researched and applicable statistical test, including guarantees of statistical power, completely unlike the Washington Post study.
Just because something is in a newspaper doesn't make it true.
What the fuck? I previewed this and all the links were fine. In fact, I've just gone and checked the HTML in the browser cache and it's fine there as well...
Here's the text again without and links for Slashdot to screw up:
This test is actually being analysed properly, using the insights gained from his previous 128kbps test, whose results you can find at http://ff123.net/dogies/dogies_plots.html). You can see the preliminary results here: http://ff123.net/128test/interim.html. In particular, look at the results of the last of the three test files. With overall 95% confidence we can say that on this test clip:
mpc is better than xing
ogg is better than xing
lame is better than xing
aac is better than xing
mpc is better than wma8
ogg is better than wma8
lame is better than wma8
aac is better than wma8
Aside from this, listening tests have previously shown that WMA8 is better than WMA7 at the low end (sub 64kpbs), at the expense of being worse at the middle to high end (112kpbs upwards). For listening to music on anything that costs more than $10, you would do best to stay away from WMA (and, looking at the other results on the page I previously mentioned, from Xing encoded MP3s as well...).
There was one recent newspaper test, done very badly, comparing files encoded at different bitrates, non-blind, which said WMA was best. This was about as far from scientific as it is possible for an 'objective' test to be. Luckily, there is a properly organised test underway to compare WMA, MPC, AAC, MP3 and Ogg at 128kpbs (which is still a very popular bitrate). To participate, go to
http://ff123.net/128test/instruct.html
Aside from this, listening tests have previously shown that WMA8 is better than WMA7 at the low end (sub 64kpbs), at the expense of being worse at the middle to high end (112kpbs upwards). For listening to music on anything that costs more than $10, you would do best to stay away from WMA (and, looking at the other results on the page I previously mentioned, from Xing encoded MP3s as well...).
Yes. Looking at the programs I listed, I've just realised that I've only listed graphical word processors... how I could have forgotten Wordperfect (version 5.1, which I still have on floppies somewhere) I really don't know -- it was the WP we used at our school (on our enormously expensive IBM PS/2 Model 50s...)
Yes - it's amazing that people are still using the NSA altering the S-boxes as an example of them trying to *weaken* crypto.
The original paper by Biham and Shamir is available at CiteSeer:
Differential Cryptanalysis of DES-like Cryptosystems. It's a classic paper that everyone interested in the field should read. There are some wonderful analyses of systems which people had suggested as alternatives to DES. For example (from page 72):
This section describes how to cryptanalyse GDES [Generalised DES]. The special case... which is suggested [by Schaumuller-Bichl] as a faster and more secure alternative to DES is breakable with just six [!] ciphertexts in a fraction of a second on a personal computer.
I *have* used other word processors... from Impression on the school Acorn Archimedes (still one of the best word processors/DTP packages, and quite similar in use to KWord), to Pagemaker, to AmiPro, to StarOffice/OpenOffice, to Word.
... or am I not allowed to actually *like* something made my Microsoft?
(your dig at BT aside, although I like the sentiment)
No, I think it's because we were missed out on the internet hype boom, so our providers are giving us the service at a cost *they* can afford. It's not as good a deal for the customers in the short term, but at least the connection shouldn't be going away in the medium term because of the company going bankrupt.
Why all this bashing of Microsoft Word? When it came out, it was better than anything else out there -- small, relatively fast, and relatively featurefull (ignoring word count:). I still have copies of Word 2 and Word 6 on floppies somewhere gathering dust (Word 2 was my favorite version -- I used to have to write macros for it). Word 95 introduced red underlining to indicate spelling errors, which personally I think is a wonderful innovation.
True, since Office 95 they've gone the route of changing the UI every pointless revision to milk money out of the gullible, but that seems to be the fate of any large successful mature commercial software package.
People forget that MS products get there initially because they are attractive to people. Some of this is monopoly abuse, if you want to call it that, but some is also because *what they produce is attractive and useful to people*.
I very rarely see the music I like (Fridge, Lamb, Broadcast, to take 3 examples) on file sharing apps, and certainly not in my audio format of choice (Ogg Vorbis). So, I'm quite happy listening to the music I have. However, I have seen most of the music I like at one time or another on the alt.binaries.sounds.* newsgroups, which seem to have slipped under the radar of the corporate witchhunters.
So why the hell don't they document exactly what their protocol is? Always we're told that this or the other app is the next big file sharing thing, but it has either no Linux client, or a binary only functionally restricted download. One of the best things to happen with Gnutella was that AOL shut the development down very early, when the protocol was still very simple. This gives people the chance to actually develop an open and documented protocol, which is more than you can say for most of the others.
It's not too bad... Xolox was only a Gnutella client, after all. Gnucleus is a much better one (yes, it's for Windows), and it's GPL as well.
On a wider note, the history of P2P file sharing seems to indicate that centralized systems will always be faster than something like Gnutella, but conversely easier for the big people to shut down. Freenet and Mojonation were two great white hopes a couple of years ago, but they currently seem to be mired in development hell, without a significant new release for a very long time. The only system that's going on stronger than ever is Gnutella.
Slashdot Mirror
Xolox is no better than Gnucleus
http://www.gnucleus.net/
which is also a Gnutella client for Windows, and also happens to be open source.
1) AC3 *is* lossy
2) 350kpbs *is* lossy from a CD input.
If you want lossless compression of audio, you'll generally be able to get down no further than about 60% of the original file size (approx 850kpbs for a standard stereo CD input). Anything less than that *has* to be lossy.
Lossy isn't a bad thing -- the whole point of audio compression is to remove the parts of the sound which the human ear can't hear. What is debatable is at what point this 'transparency' happens with different encoders (some would argue that no MP3 is ever transparent. These tend to be people that will only listen to vinyl via valve amps).
I own all the music I have on my computer, thank you.
Ripping music does not equal stealing music.
RC2 will go down to 64kpbs on stereo 44.1kHz input.
RC3 will go lower, although it's not tuned for less than 64kpbs (56kpbs sounds ok, 48kpbs ok on easy to encode music, less than that and you start getting random underwater bubble sounds).
RC3 will also go *higher* than RC2 can: its highest quality level is nominally 480 kpbs, and you can set a higher rate than that manually.
Vorbis can't really go below 64kpbs (although 56kpbs is sounding quite decent with RC3) and still sound decent *at 44.1kHz*. Resample the input to 22kHz, say, and you'll be able to get quite a bit lower.
Note that RC3 will be more of an experimental release than RC2, in that it will only have full support for 44.1kHz.
... but please, feel free to mention it in the same breath as 'The Patriot'.
Here is KDE's version increase plan:
a.b.c
'a' gets incremented when it will be binary incompatible with the previous version.
'b' gets incremented when there are additional binary compatible features
'c' gets incremented when bugs are fixed (although some features slip in from time to time)
Now, I don't know if Gnome have broken binary compatibility 7 times, as you imply. If they have, that's not something to be proud of.
Are you sure you haven't cut and pasted this from a discussion before KDE 2 came out?
CORBA is a nice idea, but grossly over designed. DCOP + KParts work exceptionally well, as evidenced by the ease with which they have been extending and developing Konqueror this last year. The KDE team are open to using CORBA, but only once GNOME have shown that it is an actual viable solution.
Multi-language support is already good in KDE 2, and will be better in KDE 3.
GNOME may be the best desktop alternative in the long term, but it certainly isn't in the short term, and I have my doubts about the medium term as well.
Are you from some odd parallel universe where the tests I linked to were COMPLETELY DIFFERENT? There is no anti-Microsoft bias or pro-Ogg bias in those tests at all.
The tests were blind for the listeners (they knew which formats were being tested, but not which files corresponded to which format), unlike the Washington Post study, which told the listeners *before they listened* what the format was. FF123's test is also comparing quality at the same (as far as possible within the format) bitrate, unlike the Washington Post study, which used different bitrates for the different formats. He has also provided you with the raw data returned from the listeners, so you can double check that he has not influenced the results.
Also, the graphs were in the 'informal analysis' section: the formal analysis concluded that the codecs in the first test were divided into three groups (from best to worst):
1) AAC and MPC
2) LAME encoded MP3, Ogg Vorbis RC2, WMA8
3) Xing encoded MP3
and this was arrived at via a researched and applicable statistical test, including guarantees of statistical power, completely unlike the Washington Post study.
Just because something is in a newspaper doesn't make it true.
This test is actually being analysed properly, using the insights gained from his previous 128kbps test, whose results you can find at http://ff123.net/dogies/dogies_plots.html). You can see the preliminary results here: http://ff123.net/128test/interim.html. In particular, look at the results of the last of the three test files. With overall 95% confidence we can say that on this test clip:
mpc is better than xing
ogg is better than xing
lame is better than xing
aac is better than xing
mpc is better than wma8
ogg is better than wma8
lame is better than wma8
aac is better than wma8
Aside from this, listening tests have previously shown that WMA8 is better than WMA7 at the low end (sub 64kpbs), at the expense of being worse at the middle to high end (112kpbs upwards). For listening to music on anything that costs more than $10, you would do best to stay away from WMA (and, looking at the other results on the page I previously mentioned, from Xing encoded MP3s as well...).
http://ff123.net/128test/instruct.html
This test is actually being analysed properly, using the insights gained from his previous 128kbps test, whose results you can find here). You can see the preliminary results here: http://ff123.net/128test/interim.html. In particular, look at the results of the last of the three test files. With overall 95% confidence we can say that on this test clip:
mpc is better than xing
ogg is better than xing
lame is better than xing
aac is better than xing
mpc is better than wma8
ogg is better than wma8
lame is better than wma8
aac is better than wma8
Aside from this, listening tests have previously shown that WMA8 is better than WMA7 at the low end (sub 64kpbs), at the expense of being worse at the middle to high end (112kpbs upwards). For listening to music on anything that costs more than $10, you would do best to stay away from WMA (and, looking at the other results on the page I previously mentioned, from Xing encoded MP3s as well...).
They are both equally valid variants. Look it up on dictionary.com: http://www.dictionary.com/cgi-bin/dict.pl?term=jud gement
Yes. Looking at the programs I listed, I've just realised that I've only listed graphical word processors... how I could have forgotten Wordperfect (version 5.1, which I still have on floppies somewhere) I really don't know -- it was the WP we used at our school (on our enormously expensive IBM PS/2 Model 50s...)
that's classified.
seriously.
The original paper by Biham and Shamir is available at CiteSeer:
Differential Cryptanalysis of DES-like Cryptosystems. It's a classic paper that everyone interested in the field should read. There are some wonderful analyses of systems which people had suggested as alternatives to DES. For example (from page 72):
I *have* used other word processors... from Impression on the school Acorn Archimedes (still one of the best word processors/DTP packages, and quite similar in use to KWord), to Pagemaker, to AmiPro, to StarOffice/OpenOffice, to Word.
... or am I not allowed to actually *like* something made my Microsoft?
(your dig at BT aside, although I like the sentiment)
No, I think it's because we were missed out on the internet hype boom, so our providers are giving us the service at a cost *they* can afford. It's not as good a deal for the customers in the short term, but at least the connection shouldn't be going away in the medium term because of the company going bankrupt.
Why all this bashing of Microsoft Word? When it came out, it was better than anything else out there -- small, relatively fast, and relatively featurefull (ignoring word count :). I still have copies of Word 2 and Word 6 on floppies somewhere gathering dust (Word 2 was my favorite version -- I used to have to write macros for it). Word 95 introduced red underlining to indicate spelling errors, which personally I think is a wonderful innovation.
True, since Office 95 they've gone the route of changing the UI every pointless revision to milk money out of the gullible, but that seems to be the fate of any large successful mature commercial software package.
People forget that MS products get there initially because they are attractive to people. Some of this is monopoly abuse, if you want to call it that, but some is also because *what they produce is attractive and useful to people*.
So what do you pay for the 1.5Mpbs service you're complaining so much about? Over here in the UK it's £40-50 ($60-70) a month for 512 down / 128 up.
In the UNIX security model, you have to explicitly set an execution bit on a file in order to run it.
So, if you download blah.py or blah.sh, by default it's just a text file.
... oh, and what is with this 'obviously it doesn't have attachments' comment?
I stand corrected.
No no: Java combines the speed of Visual Basic with the elegance and simplicity of C++.
I very rarely see the music I like (Fridge, Lamb, Broadcast, to take 3 examples) on file sharing apps, and certainly not in my audio format of choice (Ogg Vorbis). So, I'm quite happy listening to the music I have. However, I have seen most of the music I like at one time or another on the alt.binaries.sounds.* newsgroups, which seem to have slipped under the radar of the corporate witchhunters.
So why the hell don't they document exactly what their protocol is? Always we're told that this or the other app is the next big file sharing thing, but it has either no Linux client, or a binary only functionally restricted download. One of the best things to happen with Gnutella was that AOL shut the development down very early, when the protocol was still very simple. This gives people the chance to actually develop an open and documented protocol, which is more than you can say for most of the others.
It's not too bad... Xolox was only a Gnutella client, after all. Gnucleus is a much better one (yes, it's for Windows), and it's GPL as well.
On a wider note, the history of P2P file sharing seems to indicate that centralized systems will always be faster than something like Gnutella, but conversely easier for the big people to shut down. Freenet and Mojonation were two great white hopes a couple of years ago, but they currently seem to be mired in development hell, without a significant new release for a very long time. The only system that's going on stronger than ever is Gnutella.