Skip to main content

Topic: The highest quality wma aac vorbis musepack mp3, which is the best (Read 7969 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • xxy
  • [*]
The highest quality wma aac vorbis musepack mp3, which is the best
wma: vbr 98
aac: q1.00
vorbis: ogg q10.0
musepack: q10
mp3: cbr 320k

Which is the best, or what are the differences if it is hard to say which is the best?
If I want to hear the differences, how much is the audio system that I need?(if necessary, with which genre music?)

Thanks.
(here with the test result from soundexpert.org, http://soundexpert.org/encoders-320-kbps)

PS: If it feels that the words is hard to read, I'm sorry for my poor English 
  • Last Edit: 04 October, 2011, 11:49:46 AM by xxy

  • viktor
  • [*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #1
"If I want to hear the differences, how much is the audio system that I need?(if necessary, with which genre music?)"

you'd need ears which are better than what humans have.

  • dhromed
  • [*][*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #2
First; are you familiar with ABX tests?

  • testyou
  • [*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #3
Use the search function, and read the forums.
You will find lots of information on this topic.

  • pdq
  • [*][*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #4
I think what viktor is trying to say is that you should not hear problems with any of these codecs at those bitrates.

SoundExpert results are somewhat controversial in that you can take an encoded file that is audibly transparent, i.e. people are not able to distinguish it from the original in ABX tests, and extrapolate as to how much "better" than transparent they are. Ordinarily we think of transparent as being an upper bound and there is nothing better than transparent.

So, I would take comparative ratings of these codecs at SoundExperts with a grain of salt. They may or may not mean anything.

  • xxy
  • [*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #5
no audible differences in the system which is over $10000?
though i wouldn't spend so much money on my system.
  • Last Edit: 04 October, 2011, 10:32:03 PM by xxy

  • saratoga
  • [*][*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #6
no audible differences in the system which is over $10000?


A higher quality system does not usually make it any easier to recognize flaws in audio.  If the a file is transparent to inexpensive equipment, its probably transparent to better equipment as well.  The better equipment will (hopefully!) simply introduce less distortion after decoding. 

  • IgorC
  • [*][*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #7
A higher quality system does not usually make it any easier to recognize flaws in

audio.

Let me disagree here. In my experience the reference class headphones helped me a lot to spot artifacts. Expensive headphones have wider frequency response and less harmonic distortion. Previously I had some middle class headphones and couldn't hear the artifacts on high frequencies or other issues which are related with mixing of material.

The organizations of standardization (recommend to) use  high quality headphones for evaluation of audiocodecs (even at low bitrates).

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #8
Either way, the increase definitely is not linear, and spending $10000 is likely to be hilariously excessive (but joyous to ‘audiophile’ marketers). And if perceived quality did scale linearly with cost, what would be the point of spending $10000 on a system and then using a lossy format?

That aside, no one here has your ears, so you will have to conduct your own tests on a representative sample of tracks from your library in order to determine what your personal threshold of transparency is for any given format. Or if you don’t feel like doing that and want to take the gamble that the accumulated results of other users are predictive in your case, there are plenty of previous threads comparing different lossy formats, as has been said. I imagine many users would probably conclude that such high quality settings are excessive, anyway.

  • greynol
  • [*][*][*][*][*]
  • Global Moderator
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #9
The flatter the frequency response, the better artifacts are hidden.  The only factor of which I'm aware about higher-end systems that makes distinguishing artifacts potentially easier is an increased SNR of the listening environment.
13 February 2016: The world was blessed with the passing of a truly vile and wretched person.

Your eyes cannot hear.

  • saratoga
  • [*][*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #10
Let me disagree here. In my experience the reference class headphones helped me a lot to spot artifacts. Expensive headphones have wider frequency response and less harmonic distortion.


I think this is nonsense.  I've never seen a pair of even 20 dollar headphones that cut out in the upper part of the audible spectrum.  I've even EQed cheap mp3 player headphones using pink noise and test tones.  Their high frequency response is not much different then my Sennheisers. 

It sounds to me like you found a pair of headphones you liked the sound of better and are just assuming you can now ABX better.  But its silly to try and generalize based on your preferences like that. 

The organizations of standardization (recommend to) use  high quality headphones for evaluation of audiocodecs (even at low bitrates).


Using high quality equipment makes sense since if you have some cheap headphones with uneven frequency response you might reveal artifacts that wouldn't be audible with higher quality equipment.

  • ExUser
  • [*][*][*][*][*]
  • Read-only
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #11
Using high quality equipment makes sense since if you have some cheap headphones with uneven frequency response you might reveal artifacts that wouldn't be audible with higher quality equipment.
A thousand times, this. Cheap equipment can make artefacts more apparent.

Better equipment does not necessarily improve your chances to differentiate via double-blind testing. Cheap equipment, however, occasionally does.

  • IgorC
  • [*][*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #12
I think this is nonsense.  I've never seen a pair of even 20 dollar headphones that cut out in the upper part of the audible spectrum.  I've even EQed cheap mp3 player headphones using pink noise and test tones.  Their high frequency response is not much different then my Sennheisers.

It's impossible improve bandwidth without worsening distortion (THD) for the same equipment.
The frequency response by itself doesn't define the quality of headphones.  The relationship (frequency response)/(THD) does.

But its silly to try and generalize based on your preferences like that.

The game of "prove it!" is well known. "Prove the contrary" that your statement about no connection between quality of the hardware and ability of hearing artifacts is less vulnerable.

Also previous public tests reveal that there is certain relationship between the ability of hearing the artifacts and quality of the equipment.  Of course, there are a lot of other variables.

  • Dakeryas
  • [*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #13
The flatter the frequency response, the better artifacts are hidden.

This is biased to me since boomy bass can outweigh a bad cymbal tone reproduction for instance. This same goes as regards the opposite. A flatter frequency response may not emphasize one particular artifact but you're more likely to spot odd reproductions across the whole spectrum.


Let me disagree here. In my experience the reference class headphones helped me a lot to spot artifacts.

I can't but utterly agree. I ABX consistently with speakers by the way, not headphones.

I used to have a crappy JBL Creature 2.1 setup and after having done some ABX I switched from WMA Pro VBR 75 (VBR 90 produced too big files for me at that time) to Nero 1.3.3 q.50/Nero 1.5.4 q.50 when it got out.

Then I got the Razer Mako (I know, but I got it half the price and I'm still deeply convinced this is a very nice piece of audio gear) and a few tracks of my own library sounded that bad on casual listening that I started some ABX testing with the APE's from my external HDD (unfortunately I don't have enough place on my laptop at the moment to keep everything in lossless) and I was able to distinguish from the lossless up to q.65 some of the aformentionned tracks. Obviously, most files were very fine even at q.50 when I gave them an ABX try, but some dragged my attention while I wasn't doing ABX at all!

I eventually got an entry level studio system worth 1000€, and exactly the same thing occured. I also tried again to ABX my "killer" samples and successfully did so up to Nero 1.5.4 q.85. I ABXed none at q.90 (I was tired too, constantly taking breaks since nothing was obvious at that bitrate anyway) but thought I had done enough with Nero to give a try to the well-regarded QT encoder (I used qaac with QT 7.7) and on different samples I was able to ABX TVBR127.

In a nutshell, it does make a significant difference. A friend of mine could concur when he switched from entry level Sony headphones to the Audio-Technica ATH-A900, even after having spotted the artificats on the AT and ABXed them, he could not ABX any of the files with the Sony.

The system you're using is relevant regarding mastering issues too, save on Death Magnetic, I had never been bothered with clipping on the Mako's. With the Tannoy's studio monitors I heard clipping even on Load and instantly refrained myself: damn, this album is 1996, it can't be so clipped, then oppened the APE in Audacity and cried...

Here are a few logs so that I'm not hanged, drawn nor quartered:

AAC QT 7.7 TVBR 127 (363kbps)
Code: [Select]
foo_abx 1.3.4 report
foobar2000 v1.1.1
2011/08/30 16:59:49

File A: D:\Users\Recup\To Listen To\TEMP\02 Ground.flac
File B: D:\Users\Recup\To Listen To\TEMP\02 Ground.m4a

16:59:49 : Test started.
17:00:29 : 01/01  50.0%
17:00:35 : 02/02  25.0%
17:00:41 : 03/03  12.5%
17:00:51 : 04/04  6.3%
17:03:07 : 05/05  3.1%
17:03:17 : 06/06  1.6%
17:04:30 : 07/07  0.8%
17:04:43 : 08/08  0.4%
17:04:57 : 09/09  0.2%
17:05:49 : 10/10  0.1%
17:08:38 : 11/11  0.0%
17:08:54 : 12/12  0.0%
17:09:07 : 13/13  0.0%
17:09:09 : Test finished.

 ----------
Total: 13/13 (0.0%)


AAC LC .85 (342kbps)
Code: [Select]
foo_abx 1.3.4 report
foobar2000 v1.1.1
2011/09/05 22:22:23

File A: D:\Users\Recup\To Listen To\TEMP\mmasq.m4a
File B: D:\Users\Recup\To Listen To\TEMP\mmasq.wav

22:22:23 : Test started.
22:22:58 : 01/01  50.0%
22:23:14 : 02/02  25.0%
22:24:11 : 03/03  12.5%
22:26:45 : 04/04  6.3%
22:28:20 : 05/05  3.1%
22:29:32 : 06/06  1.6%
22:30:59 : 07/07  0.8%
22:36:31 : 08/08  0.4%
22:36:59 : 09/09  0.2%
22:38:10 : 10/10  0.1%
22:38:56 : 11/11  0.0%
22:41:09 : 12/12  0.0%
22:42:05 : 13/13  0.0%
22:42:18 : Test finished.

 ----------
Total: 13/13 (0.0%)
  • Last Edit: 06 October, 2011, 11:19:12 AM by Dakeryas

  • greynol
  • [*][*][*][*][*]
  • Global Moderator
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #14
The flatter the frequency response, the better artifacts are hidden.
This is biased to me since boomy bass can outweigh a bad cymbal tone reproduction for instance. This same goes as regards the opposite. A flatter frequency response may not emphasize one particular artifact but you're more likely to spot odd reproductions across the whole spectrum.

As an extreme case, if you low-pass at 600 Hz, then you probably won't hear any artifacts at all, unless the quality of the encoding is so poor that even the bass isn't encoded transparently.  Other than that, if the response is so uneven that you can't manage to pay attention to what's going on at the top end, then I suppose you're right in that regard.

Anyhow, I'm not particularly interested in these extreme cases, rather I'm simply parroting that which has been stated by the developers.  If you have an uneven frequency response, you run the risk of unmasking artifacts.  I will go farther to qualify that we're taking about the frequency bands where artifacts and masking are generally more critical which do not typically occur in the lower range of our hearing.

Regarding headphone differences:
You're going to have to do more to demonstrate differences between headphones and the quality of those differences than by simply quoting prices.  Providing actual test data would be a good start.
  • Last Edit: 06 October, 2011, 12:40:37 PM by greynol
13 February 2016: The world was blessed with the passing of a truly vile and wretched person.

Your eyes cannot hear.

  • Dakeryas
  • [*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #15
Regarding headphone differences:
You're going to have to do more to demonstrate differences between headphones and the quality of those differences than by simply quoting prices.  Providing actual test data would be a good start.


These are no headphones by the by, furthermore I haven't quoted but one price, which was quicker than posting the references of both the speakers and the sub, not to mention the DAC. What's actual test data for you?

The only thing that I am stating is that things I can't spot with my older speakers are either distinguishable or obvious on my "new" system. That's all, and it has nothing or very little to do with the frequency reponse of my speakers being uneven or not regarding the artifacts since I have yet to encounter an artificat I had spotted on my old ones that can't find on my new ones anymore, period.
  • Last Edit: 06 October, 2011, 02:57:27 PM by Dakeryas

  • greynol
  • [*][*][*][*][*]
  • Global Moderator
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #16
That's all, and it has nothing or very little to do with the frequency reponse of my speakers being uneven or not regarding the artifacts since I have yet to encounter an artificat I had spotted on my old ones that can't find on my new ones anymore, period.

What does it have to do with then?
13 February 2016: The world was blessed with the passing of a truly vile and wretched person.

Your eyes cannot hear.

  • krafty
  • [*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #17
I'm gonna be honest... every now and then, constantly, there are topics like this - "which is better MP3 or AAC". I think there should be "sticky" topics on some sub-forums to avoid more questions like that, kind of a sticky FAQ.

To the OP I just have to say that anything over 128 kbps these days is acceptable enough. Unless you can produce a lot of data why X is better than Y, then it's useless. I gave up on lossy just because I found more peace of mind with lossless. Is it much better quality than LAME -V0? To be honest, I never heard a difference.

I would just worry if I would buy MP3 that I did not encode, mentioning 320kbps, with doubtful lowpass and unidentified encoder. But I am mostly buying FLAC and it's good some artists are making FLAC download as a available option in their official stores.

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #18
I think there should be "sticky" topics on some sub-forums to avoid more questions like that, kind of a sticky FAQ.
Not only do I like this idea, but I’ve thought the same thing several times in the past. Honest! If others are anything like me (i.e. criminally lazy), the idea of such an effort isn’t hugely appealing; however, it might be a case of making a bit of an effort now to spare more in the future.

  • Dakeryas
  • [*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #19
That's all, and it has nothing or very little to do with the frequency reponse of my speakers being uneven or not regarding the artifacts since I have yet to encounter an artificat I had spotted on my old ones that can't find on my new ones anymore, period.

What does it have to do with then?


Does the frequency reponse of a speaker is the only thing that speaks for quality?






  • pdq
  • [*][*][*][*][*]
The highest quality wma aac vorbis musepack mp3, which is the best
Reply #20
Don't dodge the question. If you don't know then say that you don't know.
  • Last Edit: 06 October, 2011, 04:02:07 PM by db1989