View Full Version : How about using AAC for Music Collection
Tuning
20th September 2003, 16:54
As the relatively new audio compression called AAC is here how about a poll about people interested in it.what lossy audio compression format do u like most.
Sirber
20th September 2003, 17:06
I voted MP3 but I mainly use mp3PRO, mainly mecause I have a RCA Lyra HDD 20GB that supports it, and also I hate to waste precious hard disk space :D
SeeMoreDigital
20th September 2003, 17:10
Almost all of my music collection is stored in the Mp3pro format @ 96kbps.
However, I voted for AAC because I do think that AAC @ 96kbps sounds better. And AAC+ sounds better still.
It will be great when applications begin to appear that will allow you to rip audio CD's to AAC/AAC+ files. In much the same way the audio CD to Mp3/Mp3pro applications do now!
Cheers
EDIT
I also like the fact that my Xcard can play relpay AAC files perfectly but not my Mp3pro files........ bummer!
ChristianHJW
20th September 2003, 17:11
For music collections, only lossless ( FLAC, Wavpack, etc. ) or MPC are sensible choices IMHO, both are missing here ...
Nibor
20th September 2003, 17:21
I don't have the best sound system, so Lame MP3 @ ABR 160 KBit/s is enough for me... I can't hear the difference to the original!
And even if I could, IMO it's not important that it sounds exactly like the source, it's enough when it sounds GOOD and I can't hear 'the compression' out of it!!
(Oh oh, some people from hydrogenaudio would strike me for these words ;) :D)
Sirber
20th September 2003, 17:27
@Nibor
We share the same opinion, that's why my music vary between 40kbps CBR(Vivaldi) and 144kbps VBR (Apocaliptica) in mp3PRO :D
@All
about WMA, it's supported by most hardware, but sound Q isn't there. Also, IMHO, I don't like using M$ products...
Animaniac
20th September 2003, 17:52
Nero AAC quality level "audiophile::highest" sounds transparent to me in most cases after some informal testing. That yeilds bitrates of approximately 256 kb/s. So, Musepack still is the most viable lossy format. I still use lossless (APE) for my music.
Sirber
20th September 2003, 17:58
APE is the only lossless codec I used. Quite efficient IMHO :)
Tuning
20th September 2003, 18:09
I was only concerning about lossy compression formats.So i have edited the question.Thanks.But i found mp3Pro was build to rip CDs @64kbps -> at this bitrate the audio is of poor quality.And u cannot decode to a good WAV file.
BoNz1
20th September 2003, 18:32
AAC is by far the most interesting for me. Good implementations can be transparent to most people at 128kbps. I usually encode with quicktime at 128kbps and I can never tell the difference between the original. Also, HE AAC is very nice for low bitrates. AAC also allows for multichannel of course the only other codec which can properly do multichannel in this poll is WMA. Although vorbis would be interesting if it could do it with proper channel coupling. But, IMO AAC is by far the most complete codec ATM.
Tuning
20th September 2003, 18:34
The dB poweramp uses mp4 plugin by which any one can convert existing mp3 collections to mp4.Though the transcoding process takes a lot.checkout www.dbpoweramp.com (http://www.dbpoweramp.com/codec-central.htm) :rolleyes:
ssjkakaroto
20th September 2003, 21:32
@Tuning: if you dont care about hardware support, you should really give MPC (Musepack) a try
Ramirez
20th September 2003, 23:46
Well, if I had to store my music collection in any of the loose compression format I'd definitely choose vorbis which is IMHO sounds better then AAC or MP3PRO at any bitrate.
2 sample song, you'll hear the difference unless you're suffering from serious hearing problems.:D
Vorbis 110KBPS VBR (http://storm.wronger.com/Vorbis 110KBPS VBR Matia Bazar Ti Sento.zip)
AAC 110KBPS VBR (http://storm.wronger.com/AAC 110KBPS VBR Matia Bazar Ti Sento.zip)
Gaia
21st September 2003, 00:51
Well, if I had to store my music collection in any of the loose compression format I'd definitely choose vorbis which is IMHO sounds better then AAC or MP3PRO at any bitrate.
Placebo effect...
Ramirez
21st September 2003, 01:12
So what are implying Mr. Gaia? Who gave me that placebo exactly?
For you information I'm testing thing very thoroughly and only then
I draw my opinion on them, plz don't patronize me and address me directly next time.
*sigh* :sly:
Sirber
21st September 2003, 03:09
Originally posted by BoNz1
AAC also allows for multichannel of course the only other codec which can properly do multichannel in this poll is WMA. Although vorbis would be interesting if it could do it with proper channel coupling.RealAudio can do 5.1 from 96kbps to 253kbps, with excellent quality. Vorbis is the best above 80kbps, 2 channels.
jcsston
21st September 2003, 03:49
I choose Vorbis because it's the only free audio compression format in the list. :D
Sirber
21st September 2003, 03:55
Free vs non-free, who cares?. Let's talk about quality :)
bond
21st September 2003, 05:24
you want answers which audio codec is best at, lets say, 128kbps?
checkout this (http://audio.ciara.us/test/128extension/results.html)!
the rest are personal preferences and placebo effects...
Tuning
21st September 2003, 05:43
you want answers which audio codec is best at, lets say, 128kbps?
checkout this!
the rest are personal preferences and placebo effects...
I visited the link provided and found my favourite to be better than others.Thanks for the comments.
@BonZi
AAC also allows for multichannel of course
But u do not have DS filter to play in mediaplayer.
@jcsston
As u r fond of opensource projects,may i inform u: there are many open source encoders available for AAC/mp4.
@gaia
U need to re-think which is best.I learned that AAC@96kbps == mp3@128 kbps.And AAC @128 kbps == AC3@192kbps(stereo).
Look from where i got it
www.telos-systems.com/?/techtalk/default.htm (http://www.telos-systems.com/?/techtalk/default.htm)
Neo Neko
21st September 2003, 06:17
Originally posted by Tuning
As u r fond of opensource projects,may i inform u: there are many open source encoders available for AAC/mp4.
What are their names and how does their quality varry? AFAIK there are very few Opensource ones. And of those ones only a select few are even still being developed. And of all of them quality can vary wildly.
Originally posted by Tuning
@gaia
U need to re-think which is best.I learned that AAC@96kbps == mp3@128 kbps.And AAC @128 kbps == AC3@192kbps(stereo).
Then quite frankly you need to re-think what the word best means and how one arrives at this assessment. When dealing with lossy perceptual audio coding there is no reliable factors that can be quantised to give perfect assessmnts of quality. Because as the name implies quality is directly related to a persons personal perception and not some provable undeniable measurable effect. Which means that everyone has their own oppinion on what is best and why it is the best. AAC@96kbps!=MP3@128kbps. Not for everyone at least. To some peoples perception perhaps. But then there are those with more and less sensitive hearing.
Tuning
21st September 2003, 06:29
I didn't informed u that the observations that i presented here was not mine it was the from the listening test conducted by BBC/NHK in 1999(i think).
Now the exact link is provided below and understand urselves.
http://www.telos-systems.com/techtalk/aacpaper_2/aacpaper_8.htm.Then make ur replies.
jcsston
21st September 2003, 08:32
Originally posted by Tuning
@jcsston
As u r fond of opensource projects,may i inform u: there are many open source encoders available for AAC/mp4. [/B]
Yes, but are they open-source and legal in the USA?
It may also be that VDubMOD, CDex, and Winamp support Vorbis very well that I prefer it over AAC. :)
Tuning
21st September 2003, 09:05
That i don't know.But i can say many audio ripping softwares that are legal has started providing support to AAC.For eg:dBpoweramp,nero wave editor,cooledit pro etc.
It may also be that VDubMOD, CDex, and Winamp support Vorbis very well that I prefer it over AAC.
That is true regarding VDM.And i don't know about CDex.But winamp 2.x and 3 is currently supporting AAC by the plugins offered at Rarewares (http://rarewares.hydrogenaudio.org/aac.html)
Gaia
21st September 2003, 10:29
Originally posted by Tuning
I didn't informed u that the observations that i presented here was not mine it was the from the listening test conducted by BBC/NHK in 1999(i think).
Now the exact link is provided below and understand urselves.
http://www.telos-systems.com/techtalk/aacpaper_2/aacpaper_8.htm.Then make ur replies.
Very old listening test!
Check out http://www.hydrogenaudio.org/index.php?&act=portal
More recent blind tests:
http://audio.ciara.us/test/index.html
Soulhunter
21st September 2003, 11:23
@SeeMoreDigital
It will be great when applications begin to appear that will allow you to rip audio CD's to AAC/AAC+ files. In much the same way the audio CD to Mp3/Mp3pro applications do now!
Maybe have a look here (http://www.poikosoft.com/) ! ;)
Bye
bond
21st September 2003, 12:59
Originally posted by Tuning
But u do not have DS filter to play in mediaplayer.there is a perfectly working ds filter: CoreAAC (http://coreaac.corecodec.org/) (compiles available at rarewares)
As u r fond of opensource projects,may i inform u: there are many open source encoders available for AAC/mp4.hm, many?
i am only aware of one: FAAC (http://sourceforge.net/projects/faac)
if you want a comparison of the quality of available AAC encoders look here (http://audio.ciara.us/test/aac128test/results.html)
it is of course also already possible to directly rip from cd to aac (for example with cdex, eac...)
most of them use the psytel or faac encoder (which are available for free)
although the best results will be achieved with quicktime (but the usability of qt is s*)
he-aac/aac+ is only supported in nero bruning rom by now and only for low bitrates (max: 70kbps or so)
Tuning
21st September 2003, 14:08
@bond
there is a perfectly working ds filter: CoreAAC
But it can't be used for AVI.:(.And there is no muxer to AVI.No VDubMod support.
bond
21st September 2003, 14:09
Originally posted by Tuning
But it can't be used for AVIaac cant be put into avi (till now) ;)
that has nothing to do with the ds-filter...
Tuning
21st September 2003, 14:13
Thanks bond for ur comments.Can u suggest which format(OGM/MKV) is best supported to AAC ?
bond
21st September 2003, 14:17
Originally posted by Tuning
Can u suggest which format(OGM/MKV) is best supported to AACit should work with both (you can of course also put video into mp4 container...)
Tuning
21st September 2003, 14:21
and softwares supporting it?
SeeMoreDigital
21st September 2003, 14:32
Hi Bond,
So am I still correct in assuming that there are still no 'easy to use' CD audio to AAC encoders that don't use either the psytel or faac codecs?
And that dbpowerAMP and CD-DA extractor still use these codecs?
I too prefer the QuickTime pro AAC encodes, but like you say, the useability is awful.
Cheers
bond
21st September 2003, 14:32
Originally posted by SeeMoreDigital
So am I still correct in assuming that there are still no 'easy to use' CD audio to AAC encoders that don't use either the psytel or faac codecs?there is a commandline tool available, which uses the nero aac dlls for encoding (of course you need nero installed), called Nencoder (get it at rarewares)...
you can use that in eac for example!
And that dbpowerAMP and CD-DA extractor still use these codecs?i am not sure if dbpoweramp already supports the nero commanline tool...
@tuning
hey, why not searching the forum a little bit ;)
ogm: ogg directshow muxer in graphedit (you need the aac to be in the mp4 container and 3ivx installed (because of the mp4 splitter))
matroska: mkvmerge
mp4: mp4ui
Sirber
21st September 2003, 14:48
It's a shame that we can't use mp3PRO in movies... :(
SeeMoreDigital
21st September 2003, 16:54
Originally posted by Sirber
It's a shame that we can't use mp3PRO in movies... :( Agreed! Surely some 'clever coder' could sort this out!
Cheers
Ramirez
21st September 2003, 17:13
Originally posted by bond
the rest are personal preferences and placebo effects...
yeah yeah, IMO you and your buddies should stop repeatedly say that nonsense, there is nothing psychological about the fact that Vorbis is better codec then AAC, and if you'd live your AAC obsession for a minute you'd clearly see that. (Probably not :P)
Learn to respect other people opinions!:devil:
All these "pseudo" scientific quality tests are meaningless to me; I'd rather use my own ears instead lol.
Sirber
21st September 2003, 17:28
@Ramirez
I agree with you. From my lastests tests, vorbis had wider stereo range and sounded better than mp3PRO and AAC (96kbps).
Tuning
21st September 2003, 17:46
Comments are creeping up and i made up my mind on conducting my own comparison test.Most members have voted for Ogg Vorbis and the second position is only AAC.The thread was started to find out who are interested in AAC.Finally i think i'm going to withdraw my interest in AAC and look forward to use Ogg vorbis.Let's see what is better !;)
Teegedeck
21st September 2003, 17:59
There's even more room for subjective errors (placebo effects) in comparing audio than in comparing video. Fortunately enough, there are also applications that make blind-tests for audio-files easy. So there's no excuse for just stating 'codec x is better than codec y' if you haven't done one, with abc-hr for example. Full stop.
The only statement that I would consider safe without testing is that MP3pro is worse that both Ogg and AAC. Simply because its design is older and much more restricted. But most of us probably can't tell the difference anyway for 99% of all music (partly due to our mediocre ears, partly due to our mediocre audio equipment). Wait for the results of rjamorim's public 64-kpbs-listening-test (due tomorrow) which I'm pretty confident will show how little difference to the original we really hear in a compressed file.
bond
21st September 2003, 18:59
Originally posted by Ramirez
yeah yeah, IMO you and your buddies should stop repeatedly say that nonsense, there is nothing psychological about the fact that Vorbis is better codec then AAC, and if you'd live your AAC obsession for a minute you'd clearly see that.hell, i didnt mean to offend you (so no need to offend me) and i have nothing to add to teegdeck's post about "placebo effects" cause he is definitely right...
and i didnt say that aac is better than vorbis, i just pointed to a listening test where aac and vorbis were equal qualitywise (and thats my opinion too)...
All these "pseudo" scientific quality tests are meaningless to me; I'd rather use my own ears insteadthese blind (!) listening tests are at least more scientific than your non-blind ones.
and everybody can participate these tests (at the moment there is one running for 64kbps)...
Learn to respect other people opinions!it would be better for you to think about the meaning of this sentence again, cause in contrary to you, i didnt claim to know what audio codec is best (and than call this personal prefernce a "fact")...
JohnV
21st September 2003, 19:14
Originally posted by Ramirez
Well, if I had to store my music collection in any of the loose compression format I'd definitely choose vorbis which is IMHO sounds better then AAC or MP3PRO at any bitrate.
2 sample song, you'll hear the difference unless you're suffering from serious hearing problems.:D
Vorbis 110KBPS VBR (http://storm.wronger.com/Vorbis 110KBPS VBR Matia Bazar Ti Sento.zip)
AAC 110KBPS VBR (http://storm.wronger.com/AAC 110KBPS VBR Matia Bazar Ti Sento.zip) Providing 1 sample track tells absolutely nothing. Even worse is that the AAC encoder or the settings/version aren't mentioned.
The least what should be done is to provide also the original sample.
rjamorim
21st September 2003, 19:38
Originally posted by Ramirez
All these "pseudo" scientific quality tests are meaningless to me; I'd rather use my own ears instead lol.
LOOOL!
People already called my tests everything (flow flawed to useless), but it's the first time people call it "pseudo scientific"
(I could rant for hours and hours about how scientific is a comparision of only one sample, without testing conditions or comparision against the reference, but I'd rather point you to this (http://www.itu.int/rec/recommendation.asp?type=folders&lang=e&parent=R-REC-BS.1116) document instead.
THAT is how scientific tests are conduced)
JohnV
21st September 2003, 19:46
Originally posted by Ramirez
yeah yeah, IMO you and your buddies should stop repeatedly say that nonsense, there is nothing psychological about the fact that Vorbis is better codec then AAC, and if you'd live your AAC obsession for a minute you'd clearly see that. (Probably not :P)
"You'd clearly see that"? Are you using freq response graphs or what? :p
Well, I don't share your opinion, but maybe you haven't tested the latest AAC encoders like the latest Nero AAC or QT 6.3.
Talking only about "AAC" is very unspecific, because there are dozens of different encoder implementation ranging from very bad to very good quality.
Gotta give Vorbis the credit that it is decent quality codec from low to higher bitrates, but it's not state of the art anymore. Even Monty admits that, although he's planning to strike back.. ;)
unmei
21st September 2003, 19:47
it seems to me people are a bit more religious about audio than about video issues :)
i couldnt decide on one, only exclude two (that's clear enough :). I think altho aac and vorbis belong to about the same generation of codecs, vorbis took off early and has a unchanged good quality for some time now but i havent seen updates (well there seem to be, but official windows binaries are still the same). AAC had a very slow start, there were binaries, but they provided not that great quality for a long time and it's only recently that people here seem to start using it (SBR and NeroDigital days) and now that the fuzz is on for AAC i think it might achieve some quality steps - i think it has great "theoretical" potential (even in non-SBR mode) and it might have a bright future as mpeg-4 (system) evolves further, it just looks like slow take-off is a trademark of MPEG children :)
SeeMoreDigital
21st September 2003, 20:13
By following SoulHunter's link I've just installed and encoded an entire audio CD to .Mp4 using the dbPowerAMP application.
I'm not sure about the manufacturer of the .Mp4 codec but I'm very pleased with the results all the same!
I generated a series of two encodes from the same CD source. One at 64kbps and the other at 96kbps. Ofcourse the 96kbps encodes were by far the better sounding.
Very nice. And at last I'm able to playback good sounding compressed audio files via my Xcard..... great!
Things can only get better!
Cheers
EDIT.... What no votes for WMA. Bill won't be very impressed!
Ramirez
22nd September 2003, 00:18
@Bond:
"Placebo Effect" as I understand it.
"You're only thinking that you know the truth while in fact you're completely wrong!" :D
This is why I've reacted that way; I had no idea about the true meaning of your comment, I apologize.
P.S
I still believe that vorbis is the best codec available today, and even 100 blind tests in a row won't convince me otherwise.:D
Tuesday
22nd September 2003, 00:28
I've been using mp3 encoded with various versions of LAME for years and its only been very rare that ive had any quality problems.
I used 128kbps CBR for years as it was largly claimed to be "CD Quality" when i first ventured into encoding.
After researching into various listeing tests i now use 192kbps ABR with LAME encoding with the .dll supplied with CDex for just plain music for my collection.
However, i have used OggVorbis audio at 0.1q (~80kbps) for a long time on my DVD rips as i find because the audio is mostly vocal in film i cant hear any distortion and the extra bitrate for the video is always good, as is the lower over head of .ogm compared to .avi
just my two-peneth.....
Ramirez
22nd September 2003, 00:30
Originally posted by JohnV
Providing 1 sample track tells absolutely nothing. Even worse is that the AAC encoder or the settings/version aren't mentioned.
The least what should be done is to provide also the original sample.
It's wasn't my intention at all to conduct some sort of all inclusive grandiose test,I just provided 2 sample files for better illustration of the diff between these two. A quick comparison if you will..
As for the settings/version, here we go.
AAC Sample
Length: 5:30
File size: 4,627,371
Format: MPEG-4 AAC
Samplerate: 48000
: 2Channels:
Preset: VBR/Stereo - Internet 90/100 Kb/s LC-AAC
BitsPerSample: 16
Bitrate: 108674
Made with Nero AAC Plugin /Nero 6.0.0.1.5
Vorbis Sample
Length: 5:30
File size: 4,285,628 bytes
Average bitrate: 103 kbps
Sampling rate: 48000 Hz
Channels: 2
Nominal bitrate: 100 kbps
Max bitrate: 110 kbps
Serial number: 27548
Xiph.Org libVorbis I 20020717
Made with BeSweet v1.5b19
For the obvious reasons I couldn't upload also the 70MB source file (no Webspace)
Originally posted by JohnV
"You'd clearly see that"? Are you using freq response graphs or what? :p
Just a wrong choice of word lol, you should place "Hear" instead of "see" there.
As you can clearly see, English isn't my native tongue ;)
Originally posted by JohnV
Well, I don't share your opinion, but maybe you haven't tested the latest AAC encoders like the latest Nero AAC or QT 6.3.
Talking only about "AAC" is very unspecific, because there are dozens of different encoder implementation ranging from very bad to very good quality.
So far I've tried AACmachine (psytel)/ Nero AAC plugin and QT6 AAC Export plugin; still in all cases Vorbis produced better sound Quality then corresponding AAC File, and believe me, it was lot more then just "one track"
If you're thinking that I should change something in my setup or use a diff encoder then plz advice, I'll give it a go.
rjamorim
22nd September 2003, 00:32
Originally posted by Ramirez
P.S
I still believe that vorbis is the best codec available today, and even 100 blind tests in a row won't convince me otherwise.:D
That is usually called "closed-mindedness"
Really, that statemend is like you are asking not to be taken seriously.
vBulletin® v3.8.11, Copyright ©2000-2026, vBulletin Solutions Inc.