[Audax]

I know its french and most people won't understand, but bandwidth graphics are self explanatory.

The biggest surprise for me was that bass do not suffer as much as meds and treble when you lower the bitrate.

Anyways , i found this comparison uber helpful and it diserves and a spot in your bookmarks.

I might translate it in near future.

--> Comparatif des formats audio FLAC - APE - MP3 - MP3PRO - MPC - OGG - WMA

 

[sejarzo]

Shouldn't this post be placed in the Computer Audio section?

 

[Audax]

Quote:

Originally Posted by sejarzo /img/forum/go_quote.gif Shouldn't this post be placed in the Computer Audio section?

oops, well, a mod will fix it. Sorry about that.

 

[bigshot]

Compression artifacting almost always occurs right in the middle of the sound. Some people seem to think that lack of frequency extension is the problem, but it isn't. It's artifacting creating distortions in the sound. Once you raise the bitrate to a certain point (which can be different for different kinds of music) the artifacts fall away, and there's no audible difference between compressed and lossless.

See ya
Steve

 

[cconnaker]

Translated

Translated version of http://cf2mweb.phpnet.org/comparatif_audio/

 

[Alleyman]

Please no more graphical tests of lossy audio files. An ABX test is required to tell you if it is missing any audible information in bass/mids/highs compared to the original file.

 

[lordmozilla]

that french review was quite interesting, true i can read french, so it's probably abit more usefull to me, but did prove that mp3 is bad compared to ogg and that wma cheats by creating larger files, but even then looses to ogg.

 

[wanderman]

Quote:

Originally Posted by lordmozilla /img/forum/go_quote.gif that french review was quite interesting, true i can read french, so it's probably abit more usefull to me, but did prove that mp3 is bad compared to ogg and that wma cheats by creating larger files, but even then looses to ogg.

because bandwidth tells how the files sound?

 

[Audax]

Quote:

Originally Posted by Alleyman /img/forum/go_quote.gif Please no more graphical tests of lossy audio files. An ABX test is required to tell you if it is missing any audible information in bass/mids/highs compared to the original file.

I did not know what an ABX test was , but after some reading it appears to be some simple blindtest with two sample files but from different formats or bitrates. ABX test are BIASED by nature, one will hear artifacting and someone else won't. The graphics presented here are measures proving lower bitrate does damage sound quality. An ABX tests would only confirm the losses shown in those grapics. So what's the point in adding them ?


PC ABX Getting Started

 

[Febs]

Quote:

Originally Posted by Audax /img/forum/go_quote.gif I did not know what an ABX test was , but after some reading it appears to be some simple blindtest with two sample files but from different formats or bitrates. ABX test are BIASED by nature, one will hear artifacting and someone else won't. The graphics presented here are measures proving that a format compression does damage sound quality. An ABX tests would only confirm the losses shown in those grapics. So what's the point in adding those ABX test results ?

Because if the losses shown in your graphs are not audible, they are irrelevant. ABX testing tests whether those losses are audible and therefore provides a meaningful way to compare codecs. The graphs do not, or at least, they do not provide a meaningful comparison in the absence of a listening test to determine whether the losses matter.

 

[lordmozilla]

Quote:

Originally Posted by wanderman /img/forum/go_quote.gif because bandwidth tells how the files sound?

no, of course not, but you can see that mp3 has the tendency to cut off the sound after 16-17k, and that will definately be audible, whereas ogg keeps the spectrum alot fuller.

It's also interesting to see that ogg is able to compress alot harder while still keeping more bandwith than mp3.

Now between mp3 at 320kb/s and ogg at 320 you'd be hard pressed to find differences, but when you o lower at 128 or 96kb you can hear the difference, i can't remember who posted a blind test (however reliable those are) with this conclusion, i've done it myself and have to say that at lower bitrates ogg is much better than mp3.

I said mp3 is bad, i didn't say it didnt sound as good, i meant the compression algorithm is not true and doesnt retain the original waveform completely, even at reasonably high bitrates it cuts off higher frequencies

 

[Audax]

Quote:

Originally Posted by wanderman /img/forum/go_quote.gif because bandwidth tells how the files sound?

Of course it does, for example, if you have 0 sound pressure under 120hz you won't hear any bass.

 

[Audax]

Quote:

Originally Posted by Febs /img/forum/go_quote.gif Because if the losses shown in your graphs are not audible, they are irrelevant. ABX testing tests whether those losses are audible and therefore provides a meaningful way to compare codecs. The graphs do not, or at least, they do not provide a meaningful comparison in the absence of a listening test to determine whether the losses matter.

i agree that your first reference should always be your ears. But a Bandwidth test is a useful tool when setting up a sound system. There is a reason to that, a microphone will always hear more than you do ...

 

[ADD]

Quote:

Originally Posted by bigshot /img/forum/go_quote.gif and there's no audible difference between compressed and lossless.

It's about time you did the planet a big favour and mention in your posts that everything you seem to state as fact is actually only your own opinion. It's pretty clear from the rediculous balony you've been posting over the years that you are pretty hard of hearing. You can't tell any difference between a CD copy of an LP and you can't hear the difference between lossy and uncompressed.

I don't object to your rediculous opinions - I just object to the fact that you state them as if they are fact and apply to all of mankind (more like deaf mankind).

There are obviously people around who can hear far, far FAR more than you can, so stop acting like you have the best ears in the galaxy, because I reckon my 68 year old mother would have it all over you in any listening test.

 

[EnOYiN]

Quote:

Originally Posted by ADD /img/forum/go_quote.gif It's about time you did the planet a big favour and mention in your posts that everything you seem to state as fact is actually only your own opinion. It's pretty clear from the rediculous balony you've been posting over the years that you are pretty hard of hearing. You can't tell any difference between a CD copy of an LP and you can't hear the difference between lossy and uncompressed. I don't object to your rediculous opinions - I just object to the fact that you state them as if they are fact and apply to all of mankind (more like deaf mankind). There are obviously people around who can hear far, far FAR more than you can, so stop acting like you have the best ears in the galaxy, because I reckon my 68 year old mother would have it all over you in any listening test.

Don't start attacking people please. Aside from that I think you should at least reread bigshots post (or at least the sentence of which you quoted a part) before replying to it the way you did.

Bigshot is making a valid point. Nothing wrong with it at all.

I quote:

Quote:

Originally Posted by bigshot Once you raise the bitrate to a certain point the artifacts fall away, and there's no audible difference between compressed and lossless.

Try and explain to me why that would not be correct. Try it unless you want to start a lossless vs. lossless debate. In that case don't try to explain it please.