what is the best bitrate and encoder to archive my cd
i figured out that quicktime has enormous quality. i did some 320, 256 and 192 aac encoding with quicktime 6.3 for windows. even orchestral tracks with wide dynamic ranges sound great at 192 at best encoder settings. no reduction in the stereo pan. is there any need for more than 192 ? this aac stuff is new to me. well i worked as an audio engineer in big studios so i just can say what i hear. what do you guys figured out ist the best quality encoder setting ? i dont think i have to go to 320 right ? do i have to consider something else before i sell all my original cds ?
Full Version: high end aac encoding
QUOTE(spirit303 @ Jun 30 2003 - 05:33 AM)
what is the best bitrate and encoder to archive my cd
If you want to archive your CD collection and have high quality you may want to consider MPC (MusePack) as an alternative option. Take a look at the MPC forum for more info.
Considering that the Quicktime (Dolby) Codec only supports CBR encoding at the moment, you might have to go a little higher than 192kbps to ensure transaparency in most cases. You probably should consider using Nero's latest AAC implementation (or its precedessor Psytel AACenc), which features VBR encoding for optimal quality.
dev0
dev0
QUOTE(dev0 @ Jun 30 2003 - 09:17 AM)
Considering that the Quicktime (Dolby) Codec only supports CBR encoding at the moment, you might have to go a little higher than 192kbps to ensure transaparency in most cases. You probably should consider using Nero's latest AAC implementation (or its precedessor Psytel AACenc), which features VBR encoding for optimal quality.
dev0
dev0
Based on my ABX testing, quicktime at 160kps is generally indistinguisable from the original. Have you done any blind testing that shows a different result?
QUOTE
Based on my ABX testing, quicktime at 160kps is generally indistinguisable from the original. Have you done any blind testing that shows a different result?
Fatboy clip is one example, AAC needs at least 190-200 kbps for transparency for that clip.
Well, fatboy is a well-known problem case, and people can't base their bitrate preferences on this kind of sample, or they will overkill on all other samples.
For my own listening tastes and gear, QuickTime at 160 is surely enough.
For my own listening tastes and gear, QuickTime at 160 is surely enough.
QUOTE(Ivan Dimkovic @ Jun 30 2003 - 03:22 PM)
Fatboy clip is one example, AAC needs at least 190-200 kbps for transparency for that clip.
Is LAME -aps transparent for that clip?
QUOTE(richard123 @ Jul 1 2003 - 12:53 AM)
QUOTE(Ivan Dimkovic @ Jun 30 2003 - 03:22 PM)
Fatboy clip is one example, AAC needs at least 190-200 kbps for transparency for that clip.
Is LAME -aps transparent for that clip?
No, it isn't. It is probably the easiest clip to ABX.
In that case, anyone disagree with: QT @ 160k is at least as good as LAME -aps?
QUOTE(richard123 @ Jun 30 2003 - 10:32 PM)
In that case, anyone disagree with: QT @ 160k is at least as good as LAME -aps?
It's difficult to compare these two, since one is ABR, and the other is VBR.
But yeah, I would say their quality is often on par.
QUOTE(rjamorim @ Jun 30 2003 - 08:45 PM)
It's difficult to compare these two, since one is ABR, and the other is VBR.
How does ABR or VBR affect your ability to compare two encodings? You can compare wav to ABR or ogg to lame or anything to anything.
Have you found a real world song for which LAME -aps sounds better than QT 160k?
QUOTE(richard123 @ Jun 30 2003 - 11:34 PM)
How does ABR or VBR affect your ability to compare two encodings? You can compare wav to ABR or ogg to lame or anything to anything.
VBR encodes paying attention to a quality setting. CBR/ABR encodes paying attention to a fixed bitrate setting. That's why it's difficult to compare these two.
QUOTE
Have you found a real world song for which LAME -aps sounds better than QT 160k?
I didn't bother searching. I don't use Lame.
And whats about OGG ? I thought OGG would be better then AAC ?
QUOTE(dominik @ Jun 30 2003 - 11:46 PM)
And whats about OGG ? I thought OGG would be better then AAC ?
There's no conclusive comparision between OGG and AAC at 128kbps so far. I think you'll have to wait for the test I plan to conduce by the end of July.
QUOTE(rjamorim @ Jun 30 2003 - 09:37 PM)
QUOTE(richard123 @ Jun 30 2003 - 11:34 PM)
How does ABR or VBR affect your ability to compare two encodings? You can compare wav to ABR or ogg to lame or anything to anything.
VBR encodes paying attention to a quality setting. CBR/ABR encodes paying attention to a fixed bitrate setting. That's why it's difficult to compare these two.
However two files are prepared, you can still hear both of them and decide which sounds better.
well guys
after all it seems that quicktime ist really the BEST aac encoder. i really don't hear a difference bewteen a 192 AAC file and the original wav over a studio monitoring system (and headphones) ! with nero aac i got very bad results. i think its time to come to a conclusion !
HIGHEST END AAC --> QUICKTIME 6.3, 192, 44.1, Encoding Quality: BEST)
and thats it !
after all it seems that quicktime ist really the BEST aac encoder. i really don't hear a difference bewteen a 192 AAC file and the original wav over a studio monitoring system (and headphones) ! with nero aac i got very bad results. i think its time to come to a conclusion !
HIGHEST END AAC --> QUICKTIME 6.3, 192, 44.1, Encoding Quality: BEST)
and thats it !
QUOTE
with nero aac i got very bad results
You got bad results at 192 kbps... can you provide us with the samples, version of the NERO, and ABX results - I would like to hear those files.
Me too
auldyin
auldyin
QUOTE(spirit303 @ Jul 1 2003 - 07:34 AM)
after all it seems that quicktime ist really the BEST aac encoder. i really don't hear a difference bewteen a 192 AAC file and the original wav over a studio monitoring system (and headphones) ! with nero aac i got very bad results. i think its time to come to a conclusion !
HIGHEST END AAC --> QUICKTIME 6.3, 192, 44.1, Encoding Quality: BEST)
HIGHEST END AAC --> QUICKTIME 6.3, 192, 44.1, Encoding Quality: BEST)
Have you tested at 160?
I've just performed a quick test on one, well-recorded harpsichord sample.
http://membres.lycos.fr/guruboolez/AUDIO/s...chpsichord.flac
Encoded at 160 and 192 kbps - BEST - with QuickTime 6.3 for Windows.
http://membres.lycos.fr/guruboolez/AUDIO/s...chpsichord.flac
Encoded at 160 and 192 kbps - BEST - with QuickTime 6.3 for Windows.
CODE
1R = F:\sample\Allemande\Bachpsichord_QT192.wav
2L = F:\sample\Allemande\Bachpsichord_QT160.wav
---------------------------------------
General Comments:
---------------------------------------
1R File: F:\sample\Allemande\Bachpsichord_QT192.wav
1R Rating: 4.0
1R Comment:
---------------------------------------
2L File: F:\sample\Allemande\Bachpsichord_QT160.wav
2L Rating: 2.7
2L Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\Bachpsichord_QT192.wav
15 out of 16, pval < 0.001
Original vs F:\sample\Allemande\Bachpsichord_QT160.wav
16 out of 16, pval < 0.001
2L = F:\sample\Allemande\Bachpsichord_QT160.wav
---------------------------------------
General Comments:
---------------------------------------
1R File: F:\sample\Allemande\Bachpsichord_QT192.wav
1R Rating: 4.0
1R Comment:
---------------------------------------
2L File: F:\sample\Allemande\Bachpsichord_QT160.wav
2L Rating: 2.7
2L Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\Bachpsichord_QT192.wav
15 out of 16, pval < 0.001
Original vs F:\sample\Allemande\Bachpsichord_QT160.wav
16 out of 16, pval < 0.001
- AAC QT 160 is audibly distorted, especially on the beginning (I focused my attention on the three first second). On a good headphone, this distortion annoyed me. ABX test was easy to perform without any mistake.
- AAC QT 192 is not totally correct. Distortion isn't really perceptible, but the instrument sound dull, and lifeless. Not annoying, except maybe on careful listening, or for people that really enjoy great recordings and this complex instrument. ABX test need more concentration (I had to listen each time to the original before decision).
For comparison, I launched another test based on the same sample, and opposed QT to Ahead Nero (which still has problems with harpsichord that QT doesn't have) and to mpc (1.14) standard. This last encoder is very generous in bitrate allocation (~250 kbps, i.e. +30% than usual).
CODE
1L = F:\sample\Allemande\Bachpsichord_QT192.wav
2R = F:\sample\Allemande\Bachpsichord - nero experimental wss_off extreme.wav
3L = F:\sample\Allemande\Bachpsichord - mpc standard.wav
---------------------------------------
General Comments:
---------------------------------------
1L File: F:\sample\Allemande\Bachpsichord_QT192.wav
1L Rating: 3.8
1L Comment:
---------------------------------------
2R File: F:\sample\Allemande\Bachpsichord - nero experimental wss_off extreme.wav
2R Rating: 2.0
2R Comment:
---------------------------------------
3L File: F:\sample\Allemande\Bachpsichord - mpc standard.wav
3L Rating: 4.6
3L Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\Bachpsichord_QT192.wav
12 out of 12, pval < 0.001
Original vs F:\sample\Allemande\Bachpsichord - nero experimental wss_off extreme.wav
12 out of 12, pval < 0.001
Original vs F:\sample\Allemande\Bachpsichord - mpc standard.wav
7 out of 12, pval = 0.387
2R = F:\sample\Allemande\Bachpsichord - nero experimental wss_off extreme.wav
3L = F:\sample\Allemande\Bachpsichord - mpc standard.wav
---------------------------------------
General Comments:
---------------------------------------
1L File: F:\sample\Allemande\Bachpsichord_QT192.wav
1L Rating: 3.8
1L Comment:
---------------------------------------
2R File: F:\sample\Allemande\Bachpsichord - nero experimental wss_off extreme.wav
2R Rating: 2.0
2R Comment:
---------------------------------------
3L File: F:\sample\Allemande\Bachpsichord - mpc standard.wav
3L Rating: 4.6
3L Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\Bachpsichord_QT192.wav
12 out of 12, pval < 0.001
Original vs F:\sample\Allemande\Bachpsichord - nero experimental wss_off extreme.wav
12 out of 12, pval < 0.001
Original vs F:\sample\Allemande\Bachpsichord - mpc standard.wav
7 out of 12, pval = 0.387
- Ahead Nero (experimental version) VBR -extreme : very easy to ABX. Instrument is distorded. Quick and successful ABX test.
- Quicktime CBR 192 : easiest now to ABX ; doesn't need anymore to listen to the original in order to be sure of my decision. Instrument is really lifeless, like synthetic.
- Musepack 1.14 --standard --xlevel : slight difference, difficult to ABX on quick condition. Very good reproduction.
wow that results surprise me !!
hmmm. did u use the 6.3 QT PRO ? Soo. when i got you right, QT 192 is better that 160. But finally what is THE PERFECT AAC RESULUTION ? I have a dream. all my favourite music portable with the new ipod 2003. so i have to use AAC. WAV ist to big to handle.
it would be nice if we all together find out the perfect way !
hmmm. did u use the 6.3 QT PRO ? Soo. when i got you right, QT 192 is better that 160. But finally what is THE PERFECT AAC RESULUTION ? I have a dream. all my favourite music portable with the new ipod 2003. so i have to use AAC. WAV ist to big to handle.
it would be nice if we all together find out the perfect way !
I suspected the lifeless feeling to be explained by a slight difference in volume. So, I decided to send all encoding to wavegain before ABXing QT 160 -> 320 and mpc again.
I'm surprised by the poor result of mpc here. I'm maybe more familiar with the bachpsichord sample now. but there's maybe another explanation : mpc is chocking when compared to 5 quicktime encoding. QT files are lifeless (even at 320 kbps), like blurred ; mpc is on the opposite a bit dirty, dusty.
I initially gave the highest note to the MPC encoding ; then I passed the ABX tests from sample 6 (easiest according to ABA tests) to sample 1. Therefore, I ABXed MPC after the 5 QT encoding ; surprisingly, I find this file to be really worse than what I heard immediatly, and lowered his notation from ~4.5 to 2.9.
note about The QT224 ABX bad result : thanks to my neighboor, and his stupid boomy bass system that irritated me a lot...
EDIT : Listening equipment :
· Terratec DMX6Fire
· Beyerdynamic DT-531
(· noisy CPU fan)
CODE
1L = F:\sample\Allemande\test wavegain\Bachpsichord - mpc standard.wav
2R = F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
3R = F:\sample\Allemande\test wavegain\Bachpsichord_QT320.wav
4R = F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
5R = F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
6R = F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
---------------------------------------
General Comments:
---------------------------------------
1L File: F:\sample\Allemande\test wavegain\Bachpsichord - mpc standard.wav
1L Rating: 2.9
1L Comment:
---------------------------------------
2R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
2R Rating: 3.4
2R Comment:
---------------------------------------
3R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT320.wav
3R Rating: 4.0
3R Comment:
---------------------------------------
4R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
4R Rating: 4.0
4R Comment:
---------------------------------------
5R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
5R Rating: 1.9
5R Comment:
---------------------------------------
6R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
6R Rating: 1.6
6R Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\test wavegain\Bachpsichord - mpc standard.wav
12 out of 12, pval < 0.001
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
6 out of 12, pval = 0.613
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT320.wav
10 out of 12, pval = 0.019
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
8 out of 12, pval = 0.194
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
12 out of 12, pval < 0.001
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
10 out of 12, pval = 0.019
2R = F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
3R = F:\sample\Allemande\test wavegain\Bachpsichord_QT320.wav
4R = F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
5R = F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
6R = F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
---------------------------------------
General Comments:
---------------------------------------
1L File: F:\sample\Allemande\test wavegain\Bachpsichord - mpc standard.wav
1L Rating: 2.9
1L Comment:
---------------------------------------
2R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
2R Rating: 3.4
2R Comment:
---------------------------------------
3R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT320.wav
3R Rating: 4.0
3R Comment:
---------------------------------------
4R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
4R Rating: 4.0
4R Comment:
---------------------------------------
5R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
5R Rating: 1.9
5R Comment:
---------------------------------------
6R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
6R Rating: 1.6
6R Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\test wavegain\Bachpsichord - mpc standard.wav
12 out of 12, pval < 0.001
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
6 out of 12, pval = 0.613
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT320.wav
10 out of 12, pval = 0.019
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
8 out of 12, pval = 0.194
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
12 out of 12, pval < 0.001
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
10 out of 12, pval = 0.019
I'm surprised by the poor result of mpc here. I'm maybe more familiar with the bachpsichord sample now. but there's maybe another explanation : mpc is chocking when compared to 5 quicktime encoding. QT files are lifeless (even at 320 kbps), like blurred ; mpc is on the opposite a bit dirty, dusty.
I initially gave the highest note to the MPC encoding ; then I passed the ABX tests from sample 6 (easiest according to ABA tests) to sample 1. Therefore, I ABXed MPC after the 5 QT encoding ; surprisingly, I find this file to be really worse than what I heard immediatly, and lowered his notation from ~4.5 to 2.9.
note about The QT224 ABX bad result : thanks to my neighboor, and his stupid boomy bass system that irritated me a lot...
EDIT : Listening equipment :
· Terratec DMX6Fire
· Beyerdynamic DT-531
(· noisy CPU fan)
guruboolez
I'd really appreciate it if you would compare LAME -aps to QT.
thanks
I'd really appreciate it if you would compare LAME -aps to QT.
thanks
QUOTE(richard123 @ Jul 1 2003 - 07:03 PM)
I'd really appreciate it if you would compare LAME -aps to QT.
CODE
1R = F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
2R = F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
3L = F:\sample\Allemande\test wavegain\Bachpsichord_LAME 3.93.3 APS.wav
4L = F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
---------------------------------------
General Comments:
---------------------------------------
1R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
1R Rating: 2.5
1R Comment:
---------------------------------------
2R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
2R Rating: 4.0
2R Comment:
---------------------------------------
3L File: F:\sample\Allemande\test wavegain\Bachpsichord_LAME 3.93.3 APS.wav
3L Rating: 1.7
3L Comment:
---------------------------------------
4L File: F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
4L Rating: 4.3
4L Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
8 out of 8, pval = 0.004
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
7 out of 8, pval = 0.035
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_LAME 3.93.3 APS.wav
8 out of 8, pval = 0.004
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
6 out of 8, pval = 0.145
2R = F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
3L = F:\sample\Allemande\test wavegain\Bachpsichord_LAME 3.93.3 APS.wav
4L = F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
---------------------------------------
General Comments:
---------------------------------------
1R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
1R Rating: 2.5
1R Comment:
---------------------------------------
2R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
2R Rating: 4.0
2R Comment:
---------------------------------------
3L File: F:\sample\Allemande\test wavegain\Bachpsichord_LAME 3.93.3 APS.wav
3L Rating: 1.7
3L Comment:
---------------------------------------
4L File: F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
4L Rating: 4.3
4L Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT160.wav
8 out of 8, pval = 0.004
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT192.wav
7 out of 8, pval = 0.035
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_LAME 3.93.3 APS.wav
8 out of 8, pval = 0.004
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT224.wav
6 out of 8, pval = 0.145
· Lame is worse than QT 160 on this sample : more distorted.
· But don't conclude anything on an harpsichord sample, which is very special (wma is a winner here)
· I noticed more than one year ago than mp3 (and especially lame) isn't a good challenger for this instrument
· I noticed too that QuickTIme is the best AAC encoder for this instrument.
Therefore, Nero MP4 encoder against Fraunhofer Fastenc VBR might be a more well-balanced fighting.
QUOTE(guruboolez @ Jul 1 2003 - 01:15 PM)
· Lame is worse than QT 160 on this sample : more distorted.
· But don't conclude anything on an harpsichord sample, which is very special (wma is a winner here)
· I noticed more than one year ago than mp3 (and especially lame) isn't a good challenger for this instrument
· I noticed too that QuickTIme is the best AAC encoder for this instrument.
· But don't conclude anything on an harpsichord sample, which is very special (wma is a winner here)
· I noticed more than one year ago than mp3 (and especially lame) isn't a good challenger for this instrument
· I noticed too that QuickTIme is the best AAC encoder for this instrument.
Thank you for the response.
Have you done more general testing on Lame compared to QT?
In my testing, QT 160 was at least as good as Lame -aps, but I'm always interested in what others hear, especially someone who appears to be very good at comparative listening. My testing has been on a variety of rock, jazz and classical.
QUOTE(richard123 @ Jul 1 2003 - 08:29 PM)
Have you done more general testing on Lame compared to QT ?
No, I definitively left mp3 for archiving purpose for 18 monthes. I'm using mpc, and I'm happy with it.
It's great to know these results, Guruboolez (as usual 
). Thanks a lot for them.
). Thanks a lot for them.
QUOTE(guruboolez @ Jul 1 2003 - 02:36 PM)
No, I definitively left mp3 for archiving purpose for 18 monthes. I'm using mpc, and I'm happy with it.
Alas, mpc does not work on an ipod. Ipod only supports mp3 and mp4 (and wav)
QUOTE(richard123 @ Jul 1 2003 - 09:06 PM)
Alas, mpc does not work on an ipod. Ipod only supports mp3 and mp4 (and wav)
Is the iPod your main purpose ? Then encode your music with MP4 at 128 or 160 kbps : most files will be transparent (i.e. no artifact, except maybe a small and discreet lowpass, difficult to hear on cheap headphone). If you are annoyed on some complex files, reencode them, and only them.
guruboolez - have you tested Psytel 2.15 in the -extreme or -archive modes? I have found it to be much better than QT at higher bitrates. I completely agree with you that QT sounds synthetic and flat at the higher bitrates.
PsyTEL 2.15 VBR -archive (265 kbps) vs QuickTime 6.3 CBR 256 kbps
· PsyTEL AAC ABXtest was very fast : less than 60 seconds for 16 trials. Beginning is suffering from obvious pre-echo and brilliance.
· QuickTime ABX test : total failure. No concentration at all : I tried to perform the same blitz-test, without success... Nevertheless, find the good one on ABA test.
EDIT : I tried again, with more serious :
Not easy to ABX QT at 256 kbps...
CODE
1R = F:\sample\Allemande\test wavegain\Bachpsichord - psytel ARCHIVE.wav
2R = F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
---------------------------------------
General Comments:
---------------------------------------
1R File: F:\sample\Allemande\test wavegain\Bachpsichord - psytel ARCHIVE.wav
1R Rating: 2.0
1R Comment:
---------------------------------------
2R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
2R Rating: 4.2
2R Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\test wavegain\Bachpsichord - psytel ARCHIVE.wav
16 out of 16, pval < 0.001
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
4 out of 16, pval = 0.989
2R = F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
---------------------------------------
General Comments:
---------------------------------------
1R File: F:\sample\Allemande\test wavegain\Bachpsichord - psytel ARCHIVE.wav
1R Rating: 2.0
1R Comment:
---------------------------------------
2R File: F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
2R Rating: 4.2
2R Comment:
---------------------------------------
ABX Results:
Original vs F:\sample\Allemande\test wavegain\Bachpsichord - psytel ARCHIVE.wav
16 out of 16, pval < 0.001
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
4 out of 16, pval = 0.989
· PsyTEL AAC ABXtest was very fast : less than 60 seconds for 16 trials. Beginning is suffering from obvious pre-echo and brilliance.
· QuickTime ABX test : total failure. No concentration at all : I tried to perform the same blitz-test, without success... Nevertheless, find the good one on ABA test.
EDIT : I tried again, with more serious :
CODE
ABX Results:
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
11 out of 16, pval = 0.105
Original vs F:\sample\Allemande\test wavegain\Bachpsichord - psytel ARCHIVE.wav
16 out of 16, pval < 0.001
Original vs F:\sample\Allemande\test wavegain\Bachpsichord_QT256.wav
11 out of 16, pval = 0.105
Original vs F:\sample\Allemande\test wavegain\Bachpsichord - psytel ARCHIVE.wav
16 out of 16, pval < 0.001
Not easy to ABX QT at 256 kbps...
QUOTE(guruboolez @ Jul 1 2003 - 03:37 PM)
Is the iPod your main purpose ? Then encode your music with MP4 at 128 or 160 kbps : most files will be transparent (i.e. no artifact, except maybe a small and discreet lowpass, difficult to hear on cheap headphone). If you are annoyed on some complex files, reencode them, and only them.
Ipod is a major part of the purpose. The problem is that I've spent too much time at www.head-fi.org and seem to be moving to better and better phones. QT @ 160 appears ok on both ipod and computer. I fear that someday I'll find myself with Etymotic 4's or something better and wish I'd encoded using a higher bit rate or a better encoder.
Most of the time that I'm annoyed with an encoding, I find the original CD was the cause of the annoyance, not the encoding.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.