Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Pre-Test thread (Read 55454 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Pre-Test thread

Reply #75
Quote
i would hate to see you choose ahead just because its the latest to throw its hat in the ring to the exclusion of qt -- which was i think a surprise in the last two tests, the first for winning outright and the second for placing 2d with cbr. on that basis i would assume it merits inclusion without comment.


Well, I would be interested in Ahead's AAC, BECAUSE it is new and BECAUSE QT was tested and found to be realy good. I would like to see how Ahead fares at 64kpps, so you could roughly compare the two (though this is not realy valid, given the bitrate difference).
I also think that there is a growing userbase for Aheads implementation of the codec, so let's give it a try.

Quote
however, if you personally abx qt vs. ahead and conclude that ahead shows more promise, i for one would not object to you kicking qt out. fwiw


I thought is wasn't about chosing the best codecs out of the bunch, but more to include some well known ones... or am I getting something wrong here?!   

edit: damn, Rjamorim has smuggeld a post in before mine... 
Nothing but a Heartache - Since I found my Baby ;)

Pre-Test thread

Reply #76
Quote
I thought is wasn't about chosing the best codecs out of the bunch, but more to include some well known ones... or am I getting something wrong here?!   

My tests were *always*about*quality*

Other features like price, support, license, popularity, etc.  come far behind.

Quote
edit: damn, Rjamorim has smuggeld a post in before mine... 


Well, nothing has been decided so far

Pre-Test thread

Reply #77
For mp3-64m I think that blade would be less interesting, as it does not reflect a real usage.
Lame or FhG would be more interesting.
Using FhG is an interesting suggestion if we consider that it is claimed to be the best mp3 codec in this bitrate area. (btw I am wondering how much better than Lame it really is)

Pre-Test thread

Reply #78
Quote
I don't take listening tests. But if someone wants to test Ahead 64 vs. QT 64, the results would be very welcome.

I'm interested. To keep it lean some questions to AAC experts here:

Quicktime 6.3:
- Is it a good idea to resample to 32kHz at 64kbps

Ahead AAC (Using NeroMix 1.4.0.4):
- Should PNS be enabled at 64kbps or not?
- Using VBR is recommendable - correct?
- Any settings/hidden preferences/etc. I could have missed?

Thanks in advance.

If anyone's interested I can upload the encoded mp4 samples.
__________________

What about lame --alt-preset 64 vs. fhg? I'm curious about this too. Is there a better fhg encoder for that bitrate than Cool Edit Pro 2.1's?
Let's suppose that rain washes out a picnic. Who is feeling negative? The rain? Or YOU? What's causing the negative feeling? The rain or your reaction? - Anthony De Mello

Pre-Test thread

Reply #79
Quote
Using FhG is an interesting suggestion if we consider that it is claimed to be the best mp3 codec in this bitrate area. (btw I am wondering how much better than Lame it really is)

Well, all I know are claims that FhG is better at 64 because it supports IS, and LAME doesn't.

Anybody caring to perform a quick test Lame vs. FhG?

Pre-Test thread

Reply #80
Quote
- Is it a good idea to resample to 32kHz at 64kbps

I don't think it would be. That's definitely preprocessing, and it's bias favourable to QuickTime.

If we resample for one of the codecs, we must resample for all of them.

Quote
- Should PNS be enabled at 64kbps or not?


No. The only reason LC is being included is because it's popular and has hardware support. If we enable PNS, both reasons become moot.

Quote
- Using VBR is recommendable - correct?


Yes, probably the Tape preset.

Quote
- Any settings/hidden preferences/etc. I could have missed?


Probably not.

Quote
What about lame --alt-preset 64 vs. fhg? I'm curious about this too. Is there a better fhg encoder for that bitrate than Cool Edit Pro 2.1's?


I think, traditionally, CoolEdit and Musicmatch always use the latest FhG versions.

So, both should be OK.

Now, it's worth wondering what codec will output the best quality
"Current - Best Quality" or "Legacy - High Quality (Slow)"

Regards;

Roberto.

Pre-Test thread

Reply #81
I'll do a test with ABC/HR on few of samples from this test.
(FhG HQ vs FhG Current vs Lame -ap 64)
ruxvilti'a

Pre-Test thread

Reply #82
Quote
I'll do a test with ABC/HR on few of samples from this test.
(FhG HQ vs FhG Current vs Lame -ap 64)

Thanks

I think you'll have to try Quality 1 with FhG in Audition. The lowest preset (10) is announced as 80-95kbps, which is too high for the test.

If even 1 is too high, I might have to go with CBR 64.



BTW, I would like to use that obscuring BAT file you made for the 128 test. Is it OK?

Sorry about not using it on the 128test, but I have a policy of not changing the test unless it's absolutely necessary. Else, some people might believe the results they already submitted were invalid, and that definitely wouldn't be the case.

Regards;

Roberto.

Pre-Test thread

Reply #83
Quote
Quote
- Is it a good idea to resample to 32kHz at 64kbps

I don't think it would be. That's definitely preprocessing, and it's bias favourable to QuickTime.

If we resample for one of the codecs, we must resample for all of them.

Que? I would assume that any good codec would resample internally when it is appropriate. You're not saying that you will force the codecs to use 44.1 kHz, are you? For one it wouldn't reflect real world usage to force codecs to use a specific sample rate. Normally you would leave that decision to the encoder...

Pre-Test thread

Reply #84
Quote
Que? I would assume that any good codec would resample internally when it is appropriate. You're not saying that you will force the codecs to use 44.1 kHz, are you? For one it wouldn't reflect real world usage to force codecs to use a specific sample rate. Normally you would leave that decision to the encoder...

Sure, sure. What I said:
"If we resample for one of the codecs..."

So, I won't resample samples before feeding them to their codecs. But if the codecs do that, it's their business.

Besides, I don't think that a codec resampling is preprocessing. It's just another step in the codec compression flow.

Pre-Test thread

Reply #85
Rjamorim,
you are hereby granted the rights to use my obscured-batch-file-test-method.
Consider it as BSDv2 licensed.

Quality 1 produces too large files, unfortunately.
I have prepared the files and will take the test right now.
ruxvilti'a

Pre-Test thread

Reply #86
ABC/HR 0.9b doesn't save the results when I try to setup a new test.
I get a blank results file with the title of the new test!
I had to do the test twice (except Waiting) because of the bug.

Does anybody have a working link to Friedman?

Results:
Code: [Select]
ABC/HR Version 0.9b, 30 August 2002
Testname: Waiting MP3@64k

1R = D:\test\L-Waiting.wav
2L = D:\test\FH-Waiting.wav
3L = D:\test\FC-Waiting.wav

---------------------------------------
General Comments:

---------------------------------------
1R File: D:\test\L-Waiting.wav
1R Rating: 1.5
1R Comment: Major ringing, slight dropouts
---------------------------------------
2L File: D:\test\FH-Waiting.wav
2L Rating: 2.0
2L Comment: Underwater
---------------------------------------
3L File: D:\test\FC-Waiting.wav
3L Rating: 1.7
3L Comment: Bad ringing
---------------------------------------
ABX Results:

ABC/HR Version 0.9b, 30 August 2002
Testname: Big Yellow Taxi MP3@64k

1L = D:\test\FH-bigye.wav
2L = D:\test\FC-bigye.wav
3R = D:\test\L-bigye.wav

---------------------------------------
General Comments:

---------------------------------------
1L File: D:\test\FH-bigye.wav
1L Rating: 3.0
1L Comment: Not bad, except slight underwatery quality and lowpass
---------------------------------------
2L File: D:\test\FC-bigye.wav
2L Rating: 2.5
2L Comment: Less lowpass than 1, but more artifacts
---------------------------------------
3R File: D:\test\L-bigye.wav
3R Rating: 1.5
3R Comment: Dropouts in guitar + tons of other
---------------------------------------
ABX Results:

ABC/HR Version 0.9b, 30 August 2002
Testname: Gone MP3@64k

1R = D:\test\L-gone.wav
2L = D:\test\FC-gone.wav
3L = D:\test\FH-gone.wav

---------------------------------------
General Comments:

---------------------------------------
1R File: D:\test\L-gone.wav
1R Rating: 1.0
1R Comment: Bad dropouts  and ringing
---------------------------------------
2L File: D:\test\FC-gone.wav
2L Rating: 1.4
2L Comment:
---------------------------------------
3L File: D:\test\FH-gone.wav
3L Rating: 1.3
3L Comment: Less ringing than 2, but too much lowpass
---------------------------------------
ABX Results:

ABC/HR Version 0.9b, 30 August 2002
Testname: Polonaise MP3@64k

1R = D:\test\FC-ChopinPolonaiseDMoll.wav
2L = D:\test\FH-ChopinPolonaiseDMoll.wav
3R = D:\test\L-ChopinPolonaiseDMoll.wav

---------------------------------------
General Comments:

---------------------------------------
1R File: D:\test\FC-ChopinPolonaiseDMoll.wav
1R Rating: 2.5
1R Comment: Slightly more ringing than 2
---------------------------------------
2L File: D:\test\FH-ChopinPolonaiseDMoll.wav
2L Rating: 3.0
2L Comment:
---------------------------------------
3R File: D:\test\L-ChopinPolonaiseDMoll.wav
3R Rating: 1.0
3R Comment: Totally destroyed - sounds flangy, tons of dropouts
---------------------------------------
ABX Results:

ABC/HR Version 0.9b, 30 August 2002
Testname: Experiencia MP3@64k

1L = D:\test\FC-experiencia.wav
2R = D:\test\FH-experiencia.wav
3L = D:\test\L-experiencia.wav

---------------------------------------
General Comments:

---------------------------------------
1L File: D:\test\FC-experiencia.wav
1L Rating: 2.5
1L Comment: Slightly less ringing than 3
---------------------------------------
2R File: D:\test\FH-experiencia.wav
2R Rating: 3.5
2R Comment: Sounds very good for 64k
---------------------------------------
3L File: D:\test\L-experiencia.wav
3L Rating: 2.3
3L Comment: Ringing
---------------------------------------
ABX Results:


/EDIT\ Cleanup and modifiaction - results same \EDIT/
ruxvilti'a


Pre-Test thread

Reply #88
Don't forget to enable all additional options (like narrowing of stereo image) and disable CRC generation.

/EDIT\
Additional information about my micro-test:
- all samples were decoded with Foobar2000 0.7RC7 without replaygain to 16bit and dithered
- encoded with LAME 3.90.3 --alt-preset 64 and Adobe Audition 1.0 without CRC and with all stereo options enabled.
\EDIT/
ruxvilti'a

Pre-Test thread

Reply #89
Quote
Quote
- Is it a good idea to resample to 32kHz at 64kbps

I don't think it would be. That's definitely preprocessing, and it's bias favourable to QuickTime.

If we resample for one of the codecs, we must resample for all of them.

Quicktime has a switch to choose sampling rate - as lame (22.05kHz sampling rate is integrated in --preset 64). So it'd be fair to choose the best sounding encoding setting.
Let's suppose that rain washes out a picnic. Who is feeling negative? The rain? Or YOU? What's causing the negative feeling? The rain or your reaction? - Anthony De Mello

Pre-Test thread

Reply #90
Results
Settings used:

<samplename>_CEP_dev.wav
CEP2.1 fhg mp3 encoder
64kbps CBR, HQ, 22050Hz Samplingrate, default lowpass (10349Hz), m/s, intensity stereo, narrowing enabled.

<samplename>_CEP_LP11.wav
CEP2.1 fhg mp3 encoder
64kbps CBR, HQ, 22050Hz Samplingrate, lowpass 11025 Hz, m/s, intensity stereo, narrowing enabled.

<samplename>.wav.wav
lame 3.90.3
--preset 64

<samplename>_qt_44.wav
Quicktime Pro 6.3. AAC
64kbps CBR, Audio Track: "Music"; Stereo; Best Quality; Sample rate 44100Hz

<samplename>_qt_32.wav
Quicktime Pro 6.3. AAC
64kbps CBR, Audio Track: "Music"; Stereo; Best Quality; Sample rate 32000Hz

<samplename>_Nero.wav
NeroMix 1.4.0.4: Ahead AAC
VBR Tape; HQ; LC


Results so far:
Code: [Select]
ABC/HR Version 0.9b, 30 August 2002
Testname: 001 riteofspring 64kbps mp3/AAC

1R = .\Samples\001 riteofspring.wav.wav
2R = .\Samples\001 riteofspring_qt_44.wav
3R = .\Samples\001 riteofspring_CEP_dev.wav
4L = .\Samples\001 - riteofspring_Nero.wav
5L = .\Samples\001 riteofspring_qt_32.wav
6L = .\Samples\001 riteofspring_CEP_LP11.wav

---------------------------------------
General Comments:

---------------------------------------
1R File: .\Samples\001 riteofspring.wav.wav
1R Rating: 2.0
1R Comment: Lowpass sounds too low most of the time
annoying ringing sometimes
watery sound
warbeling of higher frequency background sounds/noise


---------------------------------------
2R File: .\Samples\001 riteofspring_qt_44.wav
2R Rating: 3.8
2R Comment: Lowpass high enough: Lack of sharpness/brighness at a few points
ringing/warbeling hard to notice.
---------------------------------------
3R File: .\Samples\001 riteofspring_CEP_dev.wav
3R Rating: 2.5
3R Comment: Practically identical to 1 in:
"Lowpass sounds too low most of the time
watery sound
warbeling of higher frequency background sounds/noise"
but almost no ringing
the background warbeling is a little bit stronger sometimes.
---------------------------------------
4L File: .\Samples\001 - riteofspring_Nero.wav
4L Rating: 2.8
4L Comment: Lowpass noticable but not annoying (similar to 2)
but much ringing/warbeling/watery sound similar to 3
---------------------------------------
5L File: .\Samples\001 riteofspring_qt_32.wav
5L Rating: 3.8
5L Comment: No difference to 2:
Lowpass high enough: Lack of sharpness/brighness at a few points
ringing/warbeling hard to notice.
---------------------------------------
6L File: .\Samples\001 riteofspring_CEP_LP11.wav
6L Rating: 2.4
6L Comment: Same as 3
"Practically identical to 1 in:
'Lowpass sounds too low most of the time
watery sound
warbeling of higher frequency background sounds/noise'
but almost no ringing
the background warbeling is a little bit stronger sometimes."
but a bit more ringing (far less than 1)

---------------------------------------
ABX Results:
Original vs .\Samples\001 riteofspring.wav.wav
   8 out of 8, pval = 0.004
Original vs .\Samples\001 riteofspring_qt_44.wav
   11 out of 12, pval = 0.003
Original vs .\Samples\001 riteofspring_CEP_dev.wav
   8 out of 8, pval = 0.004
Original vs .\Samples\001 - riteofspring_Nero.wav
   8 out of 8, pval = 0.004
Original vs .\Samples\001 riteofspring_qt_32.wav
   8 out of 8, pval = 0.004
Original vs .\Samples\001 riteofspring_CEP_LP11.wav
   8 out of 8, pval = 0.004
.\Samples\001 riteofspring_CEP_dev.wav vs .\Samples\001 riteofspring_CEP_LP11.wav
   8 out of 8, pval = 0.004

_______________________________
ABC/HR Version 0.9b, 30 August 2002
Testname: 002 bigye 64kbps mp3/AAC

1L = .\Samples\002 bigye.wav.wav
2L = .\Samples\002 bigye_CEP_LP11.wav
3R = .\Samples\002 - bigye_Nero.wav
4R = .\Samples\002 bigye_CEP_dev.wav
5L = .\Samples\002 bigye_qt_44.wav
6R = .\Samples\002 bigye_qt_32.wav

---------------------------------------
General Comments:

---------------------------------------
1L File: .\Samples\002 bigye.wav.wav
1L Rating: 1.8
1L Comment: Most annoying:
1. Ringing
2. Lowpass
3. smeared drums
Other problems don't matter in comparison
---------------------------------------
2L File: .\Samples\002 bigye_CEP_LP11.wav
2L Rating: 1.6
2L Comment: Same as 1:
Most annoying:
1. Ringing (slightly worse than 1)
2. Lowpass
3. smeared drums
Other problems don't matter in comparison
---------------------------------------
3R File: .\Samples\002 - bigye_Nero.wav
3R Rating: 3.8
3R Comment: lowpass hard to notice
a few positions with slight chirping/warbling/smearing on cymbals/hihats
---------------------------------------
4R File: .\Samples\002 bigye_CEP_dev.wav
4R Rating: 2.4
4R Comment: Same as 1 + 2, but less ringing
---------------------------------------
5L File: .\Samples\002 bigye_qt_44.wav
5L Rating: 4.0
5L Comment: same as 3 but hihats sound a bit better
---------------------------------------
6R File: .\Samples\002 bigye_qt_32.wav
6R Rating: 4.2
6R Comment: same as 3, 5 but hihats sound best
---------------------------------------
ABX Results:
Original vs .\Samples\002 bigye.wav.wav
   8 out of 8, pval = 0.004

___________________________________
ABC/HR Version 0.9b, 30 August 2002
Testname: 006 experiencia 64kbps mp3/AAC

1L = .\Samples\006 experiencia_CEP_dev.wav
2R = .\Samples\006 experiencia_qt_44.wav
3L = .\Samples\006 experiencia_qt_32.wav
4L = .\Samples\006 experiencia.wav.wav
5R = .\Samples\006 - experiencia_Nero.wav
6R = .\Samples\006 experiencia_CEP_LP11.wav

---------------------------------------
General Comments:

---------------------------------------
1L File: .\Samples\006 experiencia_CEP_dev.wav
1L Rating: 2.2
1L Comment: Lowpass
Ringing
Smeared transients
---------------------------------------
2R File: .\Samples\006 experiencia_qt_44.wav
2R Rating: 3.8
2R Comment: Smeared percussion; almost no ringing, warbling; lowpass sounds noticable but ok
---------------------------------------
3L File: .\Samples\006 experiencia_qt_32.wav
3L Rating: 3.8
3L Comment: Same as 2:
Smeared percussion; almost no ringing, warbling; lowpass sounds noticable but ok
---------------------------------------
4L File: .\Samples\006 experiencia.wav.wav
4L Rating: 1.8
4L Comment: As 1:
Lowpass
Ringing (much more)
Smeared transients
---------------------------------------
5R File: .\Samples\006 - experiencia_Nero.wav
5R Rating: 3.8
5R Comment: More accurate than 2 & 3 (sharper percussion) but sometimes slight ringing; hard to tell which is better
---------------------------------------
6R File: .\Samples\006 experiencia_CEP_LP11.wav
6R Rating: 1.9
6R Comment: As 4:
Lowpass
Ringing (a bit better)
Smeared transients
---------------------------------------
ABX Results:


So far
for AAC quicktime is better (no mentionable difference between sampling rates);
for mp3 FHG Fraunhofer at default 64kbps settings is better. Here things could change because at least the extreme stereo samples seem to sound extremely crappy with FHG's intensity stereo. L8r.
Let's suppose that rain washes out a picnic. Who is feeling negative? The rain? Or YOU? What's causing the negative feeling? The rain or your reaction? - Anthony De Mello

Pre-Test thread

Reply #91
Next result: The Chopin Piano sample. To me surprising compared to previous ones:
- lame performed far better than FHG
- Ahead AAC was worst of all codecs!
Code: [Select]
ABC/HR Version 0.9b, 30 August 2002
Testname: 003 ChopinPolonaiseDMoll 64kbps mp3/AAC

1L = .\Samples\003 ChopinPolonaiseDMoll_qt_32.wav
2L = .\Samples\003 ChopinPolonaiseDMoll.wav.wav
3R = .\Samples\003 ChopinPolonaiseDMoll_qt_44.wav
4L = .\Samples\003 ChopinPolonaiseDMoll_CEP_dev.wav
5R = .\Samples\003 ChopinPolonaiseDMoll_CEP_LP11.wav
6L = .\Samples\003 - ChopinPolonaiseDMoll_Nero.wav

---------------------------------------
General Comments:

---------------------------------------
1L File: .\Samples\003 ChopinPolonaiseDMoll_qt_32.wav
1L Rating: 3.5
1L Comment: Sounds half-good; lowpass detectable but not annoying; ringing, also pre-echo-like (not that much) most annoying artifact; high frequeny noise-like echo reduced on high notes.
---------------------------------------
2L File: .\Samples\003 ChopinPolonaiseDMoll.wav.wav
2L Rating: 4.0
2L Comment: Sounds good (similar to 1); lowpass detectable but not annoying; very small amount of warble on highest notes; high frequeny noise-like echo reduced on high notes.
---------------------------------------
3R File: .\Samples\003 ChopinPolonaiseDMoll_qt_44.wav
3R Rating: 4.4
3R Comment: Without focussing on a few seconds everything sounds fine. In ABX situation high frequency warbeling/ringing detectable.
---------------------------------------
4L File: .\Samples\003 ChopinPolonaiseDMoll_CEP_dev.wav
4L Rating: 3.0
4L Comment: Most noticable: Pre-echo on low notes; compared to this other problems (similar to 1, 2) don't matter.
---------------------------------------
5R File: .\Samples\003 ChopinPolonaiseDMoll_CEP_LP11.wav
5R Rating: 2.8
5R Comment: Almost the same as 4 but a little bit more ringing
---------------------------------------
6L File: .\Samples\003 - ChopinPolonaiseDMoll_Nero.wav
6L Rating: 2.0
6L Comment: Annoying: Warbeling (especially on echos: metallic sound); watery sound. Lowpass no problem here in comparision.
---------------------------------------
ABX Results:
Original vs .\Samples\003 ChopinPolonaiseDMoll_qt_44.wav
   8 out of 8, pval = 0.004


New Results:

Code: [Select]
ABC/HR Version 0.9b, 30 August 2002
Testname: 004 Daft_Punk___Da_Funk mp3/AAC

1R = .\Samples\004 Daft_Punk___Da_Funk.wav.wav
2L = .\Samples\004 Daft_Punk___Da_Funk_CEP_LP11.wav
3L = .\Samples\004 Daft_Punk___Da_Funk_qt_32.wav
4L = .\Samples\004 Daft_Punk___Da_Funk_qt_44.wav
5L = .\Samples\004 Daft_Punk___Da_Funk_CEP_dev.wav
6R = .\Samples\004 - Daft_Punk___Da_Funk_Nero.wav

---------------------------------------
General Comments:

---------------------------------------
1R File: .\Samples\004 Daft_Punk___Da_Funk.wav.wav
1R Rating: 1.5
1R Comment: Smeared transients, high frequency cutoff much too low, ringing
---------------------------------------
2L File: .\Samples\004 Daft_Punk___Da_Funk_CEP_LP11.wav
2L Rating: 1.3
2L Comment: Same as 1: Smeared transients, too low cutoff, ringing; but ringing is slightly more annyoing
---------------------------------------
3L File: .\Samples\004 Daft_Punk___Da_Funk_qt_32.wav
3L Rating: 3.0
3L Comment: Most annoying: lack of brightness, smeared transients
---------------------------------------
4L File: .\Samples\004 Daft_Punk___Da_Funk_qt_44.wav
4L Rating: 3.0
4L Comment: Same as 3
---------------------------------------
5L File: .\Samples\004 Daft_Punk___Da_Funk_CEP_dev.wav
5L Rating: 1.5
5L Comment: similar to 1 + 2, but almost no ringing. Transients even more smeared. Can't tell if 1 or 3 is better/worse
---------------------------------------
6R File: .\Samples\004 - Daft_Punk___Da_Funk_Nero.wav
6R Rating: 2.3
6R Comment: similar to 3+4, but additional warbeling/watery sound. clearly worse.
---------------------------------------
ABX Results:
__________________________________________

ABC/HR Version 0.9b, 30 August 2002
Testname: 005 Enola_Gay 64kbps mp3/AAC

1L = .\Samples\005 Enola_Gay.wav.wav
2L = .\Samples\005 - Enola_Gay_Nero.wav
3R = .\Samples\005 Enola_Gay_CEP_LP11.wav
4L = .\Samples\005 Enola_Gay_qt_44.wav
5R = .\Samples\005 Enola_Gay_CEP_dev.wav
6R = .\Samples\005 Enola_Gay_qt_32.wav

---------------------------------------
General Comments:

---------------------------------------
1L File: .\Samples\005 Enola_Gay.wav.wav
1L Rating: 2.0
1L Comment: most audible/annoying: lowpass, ringing, warbeling, smeared drums
---------------------------------------
2L File: .\Samples\005 - Enola_Gay_Nero.wav
2L Rating: 3.5
2L Comment: Most noticable: lack of brightness, smeared hihats with slight ringing
---------------------------------------
3R File: .\Samples\005 Enola_Gay_CEP_LP11.wav
3R Rating: 2.0
3R Comment: same as 1
---------------------------------------
4L File: .\Samples\005 Enola_Gay_qt_44.wav
4L Rating: 3.3
4L Comment: same as 2 but less bright, less ringing, more smearing; slightly worse for me but a matter of taste I'd say.
---------------------------------------
5R File: .\Samples\005 Enola_Gay_CEP_dev.wav
5R Rating: 2.2
5R Comment: similar to 1 + 3 but almost no ringing. OTH less bright, slightly smearier. All in all a bit better.
---------------------------------------
6R File: .\Samples\005 Enola_Gay_qt_32.wav
6R Rating: 3.5
6R Comment: similar to 2 and 4. least ringing, same smearing as 4, therefore same rating as 2.
---------------------------------------
ABX Results:
_________________________________________

ABC/HR Version 0.9b, 30 August 2002
Testname: 007 gone 64kbps mp3/AAC

1R = .\Samples\007 - gone_Nero.wav
2L = .\Samples\007 gone_qt_44.wav
3L = .\Samples\007 gone_qt_32.wav
4L = .\Samples\007 gone.wav.wav
5L = .\Samples\007 gone_CEP_dev.wav
6L = .\Samples\007 gone_CEP_LP11.wav

---------------------------------------
General Comments:

---------------------------------------
1R File: .\Samples\007 - gone_Nero.wav
1R Rating: 1.9
1R Comment: piano part: 2.0 - warbeling, watery, ringing
guitar part: 1.8 - still annoyingly watery + ringing sound
---------------------------------------
2L File: .\Samples\007 gone_qt_44.wav
2L Rating: 2.8
2L Comment: piano part: 3.4 - background noise gone or bumping pre-echo-like
guitar part: 2.2 - ringing, watery sound almost as bad as 1
---------------------------------------
3L File: .\Samples\007 gone_qt_32.wav
3L Rating: 2.8
3L Comment: piano + guitar part: same as 2
---------------------------------------
4L File: .\Samples\007 gone.wav.wav
4L Rating: 2.9
4L Comment: piano part: 3.8 - same as 2,3 but noise completely removed; sounds like lower cutoff but cleaner, less annoying
guitar part: 2.0 - ringing, warbeling (a little better than 2,3) but low cutoff
---------------------------------------
5L File: .\Samples\007 gone_CEP_dev.wav
5L Rating: 3.0
5L Comment: piano part: same as 4
Guitar part: same as 4, ringing slightly less annoying
---------------------------------------
6L File: .\Samples\007 gone_CEP_LP11.wav
6L Rating: 2.3
6L Comment: piano part: 2.5 - cutoff noticable = lack of brightness, background noise; (slightly) ringing transients
guitar part: 1.9 - similar to 5,6; a bit more annoying.
---------------------------------------
ABX Results:

______________________________________

ABC/HR Version 0.9b, 30 August 2002
Testname: 009 mybloodrusts.sample20sec 64kbps mp3/AAC

1L = .\Samples\009 mybloodrusts.sample20sec_qt_32.wav
2L = .\Samples\009 mybloodrusts.sample20sec_qt_44.wav
3L = .\Samples\009 mybloodrusts.sample20sec_CEP_dev.wav
4R = .\Samples\009 mybloodrusts.sample20sec_CEP_LP11.wav
5L = .\Samples\009 mybloodrusts.sample20sec.wav.wav
6R = .\Samples\009 - mybloodrusts.sample20sec_Nero.wav

---------------------------------------
General Comments:

---------------------------------------
1L File: .\Samples\009 mybloodrusts.sample20sec_qt_32.wav
1L Rating: 3.5
1L Comment: 1st part (left ear mono): good, only slight chirping on right channel, probably not noticable without headphones
2nd part: warbeling cymbals; no obvious lowpass problem
---------------------------------------
2L File: .\Samples\009 mybloodrusts.sample20sec_qt_44.wav
2L Rating: 3.0
2L Comment: similar to 1 but both problems more obvious
---------------------------------------
3L File: .\Samples\009 mybloodrusts.sample20sec_CEP_dev.wav
3L Rating: 1.5
3L Comment: 1st part: severe stereo separation problem
2nd part: high pitched drums totally smeared, replaced by warbeling
---------------------------------------
4R File: .\Samples\009 mybloodrusts.sample20sec_CEP_LP11.wav
4R Rating: 1.3
4R Comment: same as 3; warbeling more "chirpy", therefore more audible; OTH less loss of brightness; slightly worse than 3 I'd say.
---------------------------------------
5L File: .\Samples\009 mybloodrusts.sample20sec.wav.wav
5L Rating: 2.5
5L Comment: stereo separation in 1st part best of all samples
warbeling/cutoff/smearing similar to 3,4 but less annoying
---------------------------------------
6R File: .\Samples\009 - mybloodrusts.sample20sec_Nero.wav
6R Rating: 1.0
6R Comment: 1st part: other kind of stereo separation problem than 3,4 but similar annoying
2nd part: most awful chirping/warbeling
---------------------------------------
ABX Results:


edit: wrong result posted for "007 gone" - corrected.
Let's suppose that rain washes out a picnic. Who is feeling negative? The rain? Or YOU? What's causing the negative feeling? The rain or your reaction? - Anthony De Mello

Pre-Test thread

Reply #92
Hmm.. for plain AAC (not HE AAC)  I think that Ahead's  64 kbps CBR mode is better than 'Tape' preset.  Also, encoder in the latest Nero6 is better.

Pre-Test thread

Reply #93
Weird... for me LAME Polonaise sample is just unlistenable - it is flangy^H^H^H^H^H^Hphasey because of the dropouts!
(It is ~61kbps. I'll try --alt-preset CBR to see if it fixes anything)
FhG samples were just somewhat ringy and preechoey.

/EDIT\ For me, --alt-preset cbr 64 is a bit better on this sample. \EDIT/
ruxvilti'a

Pre-Test thread

Reply #94
Quote
Hmm.. for plain AAC (not HE AAC)  I think that Ahead's  64 kbps CBR mode is better than 'Tape' preset.  Also, encoder in the latest Nero6 is better.

Thanks for the info. If I get this right, this means the Nero Mix demo I've downloaded ~1 week ago doesn't contain the latest Ahead AAC encoder - correct? So ... I'll give Nero 6 a try. (Hopefully I get it uninstalled afterwards without any trouble.)

EDIT: I've just downloaded the latest Nero6 and had a look at the extracted files after starting installation: aacenc32.dll is the same version (2.5.5.1) that I had already in my Ahead/Shared folder (HE AAC encoding also works with NeroMix). Is there a newer one I've missed? - I guess not, so I'll start encoding 64kbps CBR ...

____________________

I won't test mp3 anymore. Enough time spent on it. The decision lame vs. FHG is still hard IMO.
Lame:
- more problems with ringing, warbeling, etc.
- sometimes quite good, sometimes quite annoying
+ much better for extreme stereo (everything on 1 channel)
+ preserves more details on transients

FHG:
- smeared transients
- problems with extreme stereo (Overrepresented in the test samples?)
+ ringing, warbeling not that bad most of the time
+ somehow constant performance over a broad variety of music

I'd probably choose FHG at CEP's 64kbps default setting (22.5kHz sampling rate).
Let's suppose that rain washes out a picnic. Who is feeling negative? The rain? Or YOU? What's causing the negative feeling? The rain or your reaction? - Anthony De Mello

Pre-Test thread

Reply #95
I guess the biggest problem of lame vs. FhG is for stereo imaging sensibility of listeners. Lame will probably have a quite better stereo image since it doesn't use IS at all.

Have you tried encoding with FhG with IS disabled?

Pre-Test thread

Reply #96
Not yet. good idea ...
Let's suppose that rain washes out a picnic. Who is feeling negative? The rain? Or YOU? What's causing the negative feeling? The rain or your reaction? - Anthony De Mello

Pre-Test thread

Reply #97
rjamorim: Didn't you say that you are looking for quality of the encoders? So i think PNS should be on when encodin AAC-LC with nero at this bitrate. Compatibility with portables should not be noticed in this test.

Pre-Test thread

Reply #98
Results

Settings used:

<samplename>.wav.wav
lame 3.90.3
--preset 64

<samplename>_CEP_is_narrow.wav
CEP2.1 fhg mp3 encoder
64kbps CBR, HQ, 22050Hz Samplingrate, default lowpass (10349Hz), joint stereo, intensity stereo, narrowing enabled

<samplename>_CEP_narrow.wav
CEP2.1 fhg mp3 encoder
64kbps CBR, HQ, 22050Hz Samplingrate, default lowpass (10349Hz), joint stereo, narrowing enabled, intensity stereo disabled

<samplename>_Nero.wav
NeroMix 1.4.0.4: Ahead AAC (aacenc32.dll v. 2.5.5.1)
VBR Tape; HQ; LC

<samplename>_Nero_cbr.wav
NeroMix 1.4.0.4: Ahead AAC (aacenc32.dll v. 2.5.5.1)
CBR 64kbps; HQ; LC

<samplename>_qt_44.wav
Quicktime Pro 6.3. AAC
64kbps CBR, Audio Track: "Music"; Stereo; Best Quality; Sample rate 44100Hz

Results:

Code: [Select]
---------------------------------------
General Comments:

---------------------------------------
1R File: .\Samples\001 riteofspring_qt_44.wav
1R Rating: 3.8
1R Comment: small lack of brightness, sharpness in a few places
---------------------------------------
2L File: .\Samples\001 riteofspring_CEP_is_narrow.wav
2L Rating: 2.2
2L Comment: low cutoff, warbeling, watery, ringing, mostly on transients
---------------------------------------
3R File: .\Samples\001 riteofspring.wav.wav
3R Rating: 2.0
3R Comment: similar to 2, ringing sometimes more annoying
---------------------------------------
4L File: .\Samples\001 - riteofspring_Nero.wav
4L Rating: 3.0
4L Comment: warbeling, ringing comparable to 2, 3 but bright as 1, more sharpness
---------------------------------------
5L File: .\Samples\001 riteofspring_CEP_narrow.wav
5L Rating: 2.2
5L Comment: same as 2
---------------------------------------
6L File: .\Samples\001 - riteofspring_Nero_cbr.wav
6L Rating: 3.2
6L Comment: same as 4 but a little less obvious
---------------------------------------

---------------------------------------
General Comments:

---------------------------------------
1L File: .\Samples\002 - bigye_Nero_cbr.wav
1L Rating: 3.0
1L Comment: flanging, ringing, smeared transients (hihats)
---------------------------------------
2R File: .\Samples\002 bigye.wav.wav
2R Rating: 1.5
2R Comment: lowpass, ringing, warbeling
---------------------------------------
3L File: .\Samples\002 bigye_CEP_narrow.wav
3L Rating: 2.0
3L Comment: similar to 2 but less annoying
---------------------------------------
4R File: .\Samples\002 bigye_qt_44.wav
4R Rating: 3.8
4R Comment: similar to 1 but much better (-> ringing), OTOH less sharp
---------------------------------------
5L File: .\Samples\002 - bigye_Nero.wav
5L Rating: 3.6
5L Comment: similar to 1 but less annoying
---------------------------------------
6R File: .\Samples\002 bigye_CEP_is_narrow.wav
6R Rating: 2.0
6R Comment: same as 3
---------------------------------------

---------------------------------------
General Comments:
small annoyances with warbeling/flanging of different kinds with all samples; hard to judge which is worse (exept for 2)
---------------------------------------
1R File: .\Samples\003 ChopinPolonaiseDMoll_CEP_narrow.wav
1R Rating: 2.5
1R Comment: Pre-echo, cut-off sometimes too low
---------------------------------------
2R File: .\Samples\003 - ChopinPolonaiseDMoll_Nero.wav
2R Rating: 1.5
2R Comment: warbeling, ringing, watery; metallic echos
---------------------------------------
3L File: .\Samples\003 - ChopinPolonaiseDMoll_Nero_cbr.wav
3L Rating: 3.5
3L Comment: same problems as 2 but much better; sometimes bumbping low-pitched noise like sound added (15.8-19)
---------------------------------------
4L File: .\Samples\003 ChopinPolonaiseDMoll_CEP_is_narrow.wav
4L Rating: 2.5
4L Comment: same as 1
---------------------------------------
5R File: .\Samples\003 ChopinPolonaiseDMoll_qt_44.wav
5R Rating: 4.0
5R Comment: acceptable cutoff; not much ringing, pre-echo
---------------------------------------
6R File: .\Samples\003 ChopinPolonaiseDMoll.wav.wav
6R Rating: 3.0
6R Comment: similar to 1,4; less pre-echo
---------------------------------------

---------------------------------------
General Comments:

---------------------------------------
1R File: .\Samples\004 Daft_Punk___Da_Funk_qt_44.wav
1R Rating: 3.8
1R Comment: lack of brightness, high-pitched drums smeared, some ringing/warbeling problems
---------------------------------------
2R File: .\Samples\004 Daft_Punk___Da_Funk_CEP_is_narrow.wav
2R Rating: 2.0
2R Comment: cutoff, smeared, warbeling, flanging, ringing
---------------------------------------
3R File: .\Samples\004 - Daft_Punk___Da_Funk_Nero.wav
3R Rating: 3.0
3R Comment: similar to 1, more ringing/flanging, highpitched drums + brightness a bit better; all in all worse
---------------------------------------
4L File: .\Samples\004 Daft_Punk___Da_Funk.wav.wav
4L Rating: 1.5
4L Comment: similar to 2, ringing more annoying
---------------------------------------
5L File: .\Samples\004 Daft_Punk___Da_Funk_CEP_narrow.wav
5L Rating: 2.0
5L Comment: same as 2
---------------------------------------
6L File: .\Samples\004 - Daft_Punk___Da_Funk_Nero_cbr.wav
6L Rating: 3.3
6L Comment: similar to 4 but more warbeling/dropouts; audible clipping at 5 sec.
---------------------------------------

---------------------------------------
General Comments:

---------------------------------------
1L File: .\Samples\007 gone_CEP_is_narrow.wav
1L Rating: 3.0
1L Comment: piano: cutoff, pre-echo, but not much annoying
guitar: cutoff, ringing , warbeling
---------------------------------------
2R File: .\Samples\007 gone_qt_44.wav
2R Rating: 2.5
2R Comment: piano: bright, but pumping noise, chirping, pre-echo
guitar: ringing, watery, metallic, brighter but more annoying than 1
---------------------------------------
3L File: .\Samples\007 - gone_Nero_cbr.wav
3L Rating: 2.3
3L Comment: piano: slightly warbeling noise, flanging
guitar: as 2 but more flanging
---------------------------------------
4L File: .\Samples\007 - gone_Nero.wav
4L Rating: 1.5
4L Comment: piano: 2.0 - warbeling, ringing, watery; lowpass
guitar: as 3 but even more flanging, ringing
---------------------------------------
5L File: .\Samples\007 gone.wav.wav
5L Rating: 2.8
5L Comment: same as 1, more ringing in guitar part
---------------------------------------
6R File: .\Samples\007 gone_CEP_narrow.wav
6R Rating: 3.0
6R Comment: same as 1
---------------------------------------

---------------------------------------
General Comments:

---------------------------------------
1R File: .\Samples\009 mybloodrusts.sample20sec_CEP_is_narrow.wav
1R Rating: 1.5
1R Comment: bad stereo problem in 1st part
2nd part: warbeling cymbals
---------------------------------------
2L File: .\Samples\009 mybloodrusts.sample20sec_qt_44.wav
2L Rating: 2.5
2L Comment: slight chirping in 1st part
2nd part strong chirping on cymbals
---------------------------------------
3L File: .\Samples\009 mybloodrusts.sample20sec.wav.wav
3L Rating: 2.5
3L Comment: lowpass, but best stereo reproduction
2nd: cutoff -> loss of details but flanging + ringing not that annoying
---------------------------------------
4R File: .\Samples\009 - mybloodrusts.sample20sec_Nero_cbr.wav
4R Rating: 1.5
4R Comment: worst stereo problem in 1st part
2nd part not as annoying as 6 but similar
---------------------------------------
5R File: .\Samples\009 mybloodrusts.sample20sec_CEP_narrow.wav
5R Rating: 2.4
5R Comment: 1st part: same as 3 but more flanging
2nd part similar to 3
---------------------------------------
6L File: .\Samples\009 - mybloodrusts.sample20sec_Nero.wav
6L Rating: 1.5
6L Comment: 1st part: bumping stereo problem, 2nd awful warbeling, ringing
---------------------------------------


Conclusions:

In the case where FHG with intensity stereo failed badly before it's improved very much without IS. No general differences detectable, so FHG without IS seems to be the best choice for 64kbps mp3.

Most of the time using CBR improves the performance of Nero AAC but Quicktime is still better.
Let's suppose that rain washes out a picnic. Who is feeling negative? The rain? Or YOU? What's causing the negative feeling? The rain or your reaction? - Anthony De Mello

Pre-Test thread

Reply #99
Quote
Conclusions:

In the case where FHG with intensity stereo failed badly before it's improved very much without IS. No general differences detectable, so FHG without IS seems to be the best choice for 64kbps mp3.

Most of the time using CBR improves the performance of Nero AAC but Quicktime is still better.

Great! Thanks a lot for your invaluable help.

So, here's an updated list of codecs to be featured:

-Nero HE AAC
-Ogg Vorbis
-MP3pro
-WMAv9 std
-Apple AAC LC
-Higher anchor: Lame --alt-preset 128
-FhG MP3 64kbps w/o IS

Real is on hold due to reasons out of my league. I'll know if it'll be featured or not by the weekend.