aacPlus is just a marketing name for the HE-AAC standard.
There are GPL'ed implementations of HE-AAC decoders, for example at http://www.audiocoding.com, so these streams should be playable on open source systems, too.
Btw. Some of technical details in the article (notably about parametric stereo) are *complete bollocks*. What they describe is Mide-Side stereo.
Parametric stereo transmits only a mono channel plus a very small amount of sideband information that describes how to reconstruct the stereo image (via decorrelation and fading).
Hmm, I guess they don't count Jot as handwriting, which is at least on the Sony Ericssons for a while. It's close enough that it tends to "just work". And it's probably more reliable than handwriting recognition. I mean, even as a human I can't decode most peoples handwriting:)
That said, being a true IT person, I got a P910i with a keyboard, since I can't remember how handwriting was supposed to work, anyway...
>The Crafty license is not an Open Source or >Free Software license.
It's not because some zealots redefined what "open source" or "free software" are supposed to mean that the rest of the world has to follow their perverted definition.
The source is available for anyone. Hence open source.
Re:WPA was the source of my woes
on
Linux Unwired
·
· Score: 1
Please see the fix for this I posted somewhere else down here. And note that WEP can be trivially cracked with Airsnort.
Another Windows XP wireless bug
on
Linux Unwired
·
· Score: 1
There are, besides the SSID broadcasting issue, several other bugs in Windows XP wifi.
One of them happens with WPA (which I do hope you are using, given that WEP is so easily hacked), and causes you to disconnect after a few minutes. After reconnecting there are no further problems.
This appears to be a working fix, needs a registry edit:
>> This will solve your WiFi problem! although this >> should be considered a workaround, >> rather than a fix. With AuthMode set to 2 it means >> Machine authentication only. >> Whenever a user logs in, it has no effect on the >> connection. 802.1X authentication is >> performed using machine credentials only.
[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\EAPOL\Par am eters\General\Global] "AuthMode"=dword:00000002
Re:64-benchmarks wont be good
on
AMD64 Preview
·
· Score: 1
It's worse than that. On all operating systems save Windows Server 2003 and Linux 2.6.x, enabling hyperthreading on a dual will _lower_ performance in most cases, because the OS cannot differentiate real from hyperthreading CPUs.
And yes, this is despite Windows XP being marketed as 'HT optimized'.
>One thing I see, though, is that hyperthreading >would probably not do any good for such a game >b/c all of the integer ALUs on a processor would >be used by one thread, so there wouldn't be any >ALUs open for another thread. I think in this >sort of application of the Xenon, turning >hyperthreading off would help boost performance, >although I can't be 100% sure of it. Just a >thought.
Hyperthreading generally improves the performance of a chessprogram in terms of number of positions analyzed per second. But the parallel efficiency of most programs is not very high, so overall it ends up being about break-even.
Parallelizing chess is MUCH harder than most people here seem to realize.
Most of the really good Winamp coders left already. The guy that wrote almost all the core plugins, Peter Pawlowksi, quit because he didn't like Winamp 3 design and thought it was a dead end. Because AOL still owns his code, some of the plugins are 'dead' now, and the code can't be used any more. Bummer.
He wrote his own player instead, which is, eh, quite different from Winamp, Foobar2000.
Anyway, Frankel has little to complain about. Nullsoft was bought out for almost 86M$. For that much money, he'll never have to code, err, express himself ever again.
>I say the psychoacoustics principles are known >and take modelling as a signalling issue. Or >have the ogg guys measured the human ear >themselve?
Ha, this is a matter of definition. As I stated above, the basic principles are known, the finer work is not. This is still psychoacoustics in my book, but apparently it's not in yours. But I assume we actually agree about the state of things.
All of the lower level work and research on the psymodels was done by the Vorbis people themselves, which includes some refutations of scientifically published data.
>Perhaps you should read some scientific articles >instead of keeping your nose in C-code. (I don't >need to come up with arguments as you don't do >that either?)
Ok, I'm curious how you reliably (key word here) judge quality of a psychoacoustic audio codec with spectrographs. Pointers to articles or links are fine. Methods that 'somewhat' or 'usually' work are not! I know that none of the developers for Vorbis, AAC and LAME and MusePack use spectrographs for any serious work. And I'm sure they read a lot more scientific articles than me.
>This is partly correct, my nick was a bit of >troll four years ago (grew a bit tired of the >anti ms comments, which were 50% of the posts), >however not anymore, just kept it.
>note: MPC/MP+ ONLY does VBR, unlike MP3 where >VBR was added later and Vorbis where it is >optional).
Vorbis is natively VBR. You can do CBR if you really want to shoot yourself in the foot, but it also supports managed bitrate for sane people. MPC can ONLY do VBR.
How can this parent be +5 insightful? It is wrong and uninformative, and probably a troll anyway (see the nic and remember this is/.)
>I worked with MPEG4 (AAC) and OGG a lot (for my >phd. thesis) and spectral analysis IS very important.
It's interesting so many unsupported and blatantly false claims are made by people that claim to have PhD's.
>However, these hearing curves and integration times are already known
Really? Correctly modelling those is one of the greatest challenges in improving psychoacoustic encoders. We know how things work on a coarse level, but finemodelling needs a lot of work still.
>Most profit nowadays is in clever signal processing.
Not at all. Better modeling of human psychoacoustics, more efficient encoding (e.g. TNS in AAC, VQ codebooks in Vorbis), and generally 'tricks' like lossy stereo or perceptual noise substitution provide the largest gain. (It's arguable whether these are 'clever signal processing', but you were saying it was not due to psychoacoustics, which is just wrong)
>The spectrum of a decoded signal shows almost all artifacts very well and is therefore >something which helps a lot in showing artifacts in a coding scheme.
This is blatantly false and if you had ever seriously worked in this field you would know it.
>Theoratically AAC and OGG are rather similar, but AAC has a few nice extra's like the Temporal >Noise Shaper
And so does Vorbis have many things that AAC does not have, moreover, most of the similarity is only superficial.
>No.. no... the virus not only downloaded the >porn, but also used his credit card information >to sign up for the site, confirmed his >subscription via email,
If it got his credit card details, registering to a porn site would be no problem. You don't even need his email.
>Speaking of which - one of Ogg Vorbis' strongest >selling points is bitrate peeling - you >can "peel" a 192 kbps file to 128 kbps and the >resulting file will sound just as good as if it >were encoded directly off the original CD/wave >file.
Almost. Peeling will not give *exactly* the same quality, but much better than decode/reencode, and it will be faster too.
>But there is no tool yet. When can we expect to >see one?
There is a proof-of-concept tool available right now, but it does not get good quality yet. There hasn't been much demand for it yet so developers have focussed elsewhere - maybe with the portables out this will change.
>Uhm, if cam was NULL after the first check, >wouldn't cam->busy_lock cause a segfault
I don't feel bad about being flamed by people that missed my point, but I do feel bad about this. You are 100% right - the way I read the code can't possibly be correct.
>You're one of those guys who thinks that anyone >who doesn't grasp precisely all the different >technical fields you do, must be an fool.
Not quite. I did actually read the paper and even the authors resume before posting here, but I still did because the snippet and the comments surrounding it didn't make any sense. After seeing that the original code in 2.4.1 is different from the snippet in a critical way, I'm guessing they slightly rearranged it to fit their paper better. In that sense, my original annoyance towards their comments at the original author don't hold, since it's not what he wrote in the first place. (And that's was exactly the reason why I made the original post). But I think the end result is that the example is now particularly unfortunate.
There is some irony in that while I critiqued the statement that the original author was a dummy, I'm now being attacked because I supposedly think the papers author is a fool. Bzzz, eh, I didn't say anything.
I'm guessing being wrong and still moderated up to 5 doesn't help either.
Slashdot Mirror
aacPlus is just a marketing name for the HE-AAC standard.
There are GPL'ed implementations of HE-AAC decoders, for example at http://www.audiocoding.com, so these streams should be playable on open source systems, too.
Btw. Some of technical details in the article (notably about parametric stereo) are *complete bollocks*. What they describe is Mide-Side stereo.
Parametric stereo transmits only a mono channel plus a very small amount of sideband information that describes how to reconstruct the stereo image (via decorrelation and fading).
AAC is *much LESS* expensive than MP3. Just compare the licensing costs from Vialicensing (AAC) vs Thomson (MP3).
The parent is plain wrong. ("Don't believe all you read on the internet, kids")
Hmm, I guess they don't count Jot as handwriting, which is at least on the Sony Ericssons for a while. It's close enough that it tends to "just work". And it's probably more reliable than handwriting recognition. I mean, even as a human I can't decode most peoples handwriting :)
That said, being a true IT person, I got a P910i with a keyboard, since I can't remember how handwriting was supposed to work, anyway...
http://foobar2000.org
Simple. Works.
The author is the original author of the Winamp Vorbis support...
You should have gotten a genuine AMD64 instead of one of them crappy Intel clones ;-)
There was a good scientific test linked on slashdot a while ago, comparing spamfilters and including DSPAM and SpamAssassin.
Contrary to DSPAM author's claims, both it and and CRM-114 (another package which likes to self-hype) performed quite a bit worse than SpamAssassin.
Then again, I've heard people being happy with DSPAM that were not happy with SA.
Guess it depends on the mailfeed you get.
I've been waiting for 2 years for decent Windows drivers and my patience is ALSO up.
From my experience the only usable gratis 64 bit distribution is Gentoo, which works really well.
Debian 64 bit is plain unusable, and simply outpaced by Gentoo speedy tracking of new stuff.
>The Crafty license is not an Open Source or
>Free Software license.
It's not because some zealots redefined what
"open source" or "free software" are supposed
to mean that the rest of the world has to follow
their perverted definition.
The source is available for anyone. Hence open
source.
It's a quad OPTERON 850, not an 850Mhz machine.
The Opteron 850 runs at 2.4Ghz.
How exactly is that not open source?
Please see the fix for this I posted somewhere else down here. And note that WEP can be trivially cracked with Airsnort.
There are, besides the SSID broadcasting issue, several other bugs in Windows XP wifi.
r am eters\General\Global]
One of them happens with WPA (which I do hope you are using, given that WEP is so easily hacked), and causes you to disconnect after a few minutes. After reconnecting there are no further problems.
This appears to be a working fix, needs a registry edit:
>> This will solve your WiFi problem! although this
>> should be considered a workaround,
>> rather than a fix. With AuthMode set to 2 it means
>> Machine authentication only.
>> Whenever a user logs in, it has no effect on the
>> connection. 802.1X authentication is
>> performed using machine credentials only.
[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\EAPOL\Pa
"AuthMode"=dword:00000002
It's worse than that. On all operating systems save Windows Server 2003 and Linux 2.6.x, enabling hyperthreading on a dual will _lower_ performance in most cases, because the OS cannot differentiate real from hyperthreading CPUs.
And yes, this is despite Windows XP being marketed as 'HT optimized'.
>One thing I see, though, is that hyperthreading
>would probably not do any good for such a game
>b/c all of the integer ALUs on a processor would
>be used by one thread, so there wouldn't be any
>ALUs open for another thread. I think in this
>sort of application of the Xenon, turning
>hyperthreading off would help boost performance,
>although I can't be 100% sure of it. Just a
>thought.
Hyperthreading generally improves the performance of a chessprogram in terms of number of positions analyzed per second. But the parallel efficiency of most programs is not very high, so overall it ends up being about break-even.
Parallelizing chess is MUCH harder than most people here seem to realize.
Someone please mod the parent up to +5, Insightful. It's the only intelligent post to be made so far.
Most of the really good Winamp coders left already. The guy that wrote almost all the core plugins, Peter Pawlowksi, quit because he didn't like Winamp 3 design and thought it was a dead end. Because AOL still owns his code, some of the plugins are 'dead' now, and the code can't be used any more. Bummer.
He wrote his own player instead, which is, eh, quite different from Winamp, Foobar2000.
Anyway, Frankel has little to complain about. Nullsoft was bought out for almost 86M$. For that much money, he'll never have to code, err, express himself ever again.
>I say the psychoacoustics principles are known
>and take modelling as a signalling issue. Or
>have the ogg guys measured the human ear
>themselve?
Ha, this is a matter of definition. As I stated above, the basic principles are known, the finer work is not. This is still psychoacoustics in my book, but apparently it's not in yours. But I assume we actually agree about the state of things.
All of the lower level work and research on the psymodels was done by the Vorbis people themselves, which includes some refutations of scientifically published data.
>Perhaps you should read some scientific articles
>instead of keeping your nose in C-code. (I don't
>need to come up with arguments as you don't do
>that either?)
Ok, I'm curious how you reliably (key word here) judge quality of a psychoacoustic audio codec with spectrographs. Pointers to articles or links are fine. Methods that 'somewhat' or 'usually' work are not! I know that none of the developers for Vorbis, AAC and LAME and MusePack use spectrographs for any serious work. And I'm sure they read a lot more scientific articles than me.
>This is partly correct, my nick was a bit of
>troll four years ago (grew a bit tired of the
>anti ms comments, which were 50% of the posts),
>however not anymore, just kept it.
Ok, my bad.
--
GCP
>note: MPC/MP+ ONLY does VBR, unlike MP3 where
>VBR was added later and Vorbis where it is
>optional).
Vorbis is natively VBR. You can do CBR if you really want to shoot yourself in the foot, but it also supports managed bitrate for sane people. MPC can ONLY do VBR.
--
GCP
How can this parent be +5 insightful? It is wrong and uninformative, and probably a troll anyway (see the nic and remember this is /.)
>I worked with MPEG4 (AAC) and OGG a lot (for my
>phd. thesis) and spectral analysis IS very important.
It's interesting so many unsupported and blatantly false claims are made by people that claim to have PhD's.
>However, these hearing curves and integration times are already known
Really? Correctly modelling those is one of the greatest challenges in improving psychoacoustic encoders. We know how things work on a coarse level, but finemodelling needs a lot of work still.
>Most profit nowadays is in clever signal processing.
Not at all. Better modeling of human psychoacoustics, more efficient encoding (e.g. TNS in AAC, VQ codebooks in Vorbis), and generally 'tricks' like lossy stereo or perceptual noise substitution provide the largest gain. (It's arguable whether these are 'clever signal processing', but you were saying it was not due to psychoacoustics, which is just wrong)
>The spectrum of a decoded signal shows almost all artifacts very well and is therefore
>something which helps a lot in showing artifacts in a coding scheme.
This is blatantly false and if you had ever seriously worked in this field you would know it.
>Theoratically AAC and OGG are rather similar, but AAC has a few nice extra's like the Temporal
>Noise Shaper
And so does Vorbis have many things that AAC does not have, moreover, most of the similarity is only superficial.
--
GCP
>No.. no... the virus not only downloaded the
>porn, but also used his credit card information
>to sign up for the site, confirmed his
>subscription via email,
If it got his credit card details, registering to a porn site would be no problem. You don't even need his email.
This would be a viable defense IMHO.
--
GCP
>Assuming they use a 64bit key, so that Sun could
>legally export it from America
Why would they need to export encryption from the US? AES was invented in Belgium.
--
GCP
>Speaking of which - one of Ogg Vorbis' strongest
>selling points is bitrate peeling - you
>can "peel" a 192 kbps file to 128 kbps and the
>resulting file will sound just as good as if it
>were encoded directly off the original CD/wave
>file.
Almost. Peeling will not give *exactly* the same quality, but much better than decode/reencode, and it will be faster too.
>But there is no tool yet. When can we expect to
>see one?
There is a proof-of-concept tool available right now, but it does not get good quality yet. There hasn't been much demand for it yet so developers have focussed elsewhere - maybe with the portables out this will change.
--
GCP
>Uhm, if cam was NULL after the first check,
>wouldn't cam->busy_lock cause a segfault
I don't feel bad about being flamed by people that missed my point, but I do feel bad about this. You are 100% right - the way I read the code can't possibly be correct.
--
GCP
>You're one of those guys who thinks that anyone
>who doesn't grasp precisely all the different
>technical fields you do, must be an fool.
Not quite. I did actually read the paper and even the authors resume before posting here, but I still did because the snippet and the comments surrounding it didn't make any sense. After seeing that the original code in 2.4.1 is different from the snippet in a critical way, I'm guessing they slightly rearranged it to fit their paper better. In that sense, my original annoyance towards their comments at the original author don't hold, since it's not what he wrote in the first place. (And that's was exactly the reason why I made the original post). But I think the end result is that the example is now particularly unfortunate.
There is some irony in that while I critiqued the statement that the original author was a dummy, I'm now being attacked because I supposedly think the papers author is a fool. Bzzz, eh, I didn't say anything.
I'm guessing being wrong and still moderated up to 5 doesn't help either.
--
GCP