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: How do I perform an ABX test to compare mp3 vs wav (Read 7107 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

How do I perform an ABX test to compare mp3 vs wav

Hello.  I am new to abx testing, and would like to know if I can tell the difference between my lame encoded V1 vbr mp3s and the original wave files. However, I attempted to use WinAbx and it only lets me compare two wave files.  This is strange imo, what is the point? I tried to compare my wav version of Stinkfist by tool to the mp3 but it wouldnt let me select an mp3 in my library, only waves. So, I am sorry if this has already been posted. I tried searching and found only more specific stuff. I also looked around the foobar2000 control panel and found no opiton for abx testing. Thanks for the help

Joe

How do I perform an ABX test to compare mp3 vs wav

Reply #1 –
Foobars ABX component is not installed by default, to install it re-run the installer and manually select it. Then simply select two tracks and you can start the ABX test from the context menu.

How do I perform an ABX test to compare mp3 vs wav

Reply #2 –
Quote from: smkk on
Foobars ABX component is not installed by default, to install it re-run the installer and manually select it. Then simply select two tracks and you can start the ABX test from the context menu.


Ok thanks, I selected the two tracks but I dont know where or what the context menu is, sorry.

How do I perform an ABX test to compare mp3 vs wav

Reply #3 –
Assuming you are not trolling - context menu is standard windows convention of having appropriate commands available via Right Click.

In your case, select the two appropriate files > right click > Utils > abx will get you to the foobar abx command dialog.

this may also help

http://www.hydrogenaudio.org/forums/index....opic=16295&

terry

How do I perform an ABX test to compare mp3 vs wav

Reply #4 –
Quote from: tpijag on
Assuming you are not trolling - context menu is standard windows convention of having appropriate commands available via Right Click.

In your case, select the two appropriate files > right click > Utils > abx will get you to the foobar abx command dialog.

this may also help

http://www.hydrogenaudio.org/forums/index....opic=16295&

terry


Never heard right click called context menu before.  Anyways, thanks for the help.  I'll post the results if I can figure out how.

How do I perform an ABX test to compare mp3 vs wav

Reply #5 –
Quote from: Sound-of-Muzak on
Quote from: tpijag on
Assuming you are not trolling - context menu is standard windows convention of having appropriate commands available via Right Click.

In your case, select the two appropriate files > right click > Utils > abx will get you to the foobar abx command dialog.

this may also help

http://www.hydrogenaudio.org/forums/index....opic=16295&

terry


Never heard right click called context menu before.  Anyways, thanks for the help.  I'll post the results if I can figure out how.


To make a correction: my mp3s are actually v1 not v2. Out of 13 trials I scored 5/13. I'm guessing this is average? I ended up comparing Harvest by Opeth, an acoustic strumming piece with a full band.

How do I perform an ABX test to compare mp3 vs wav

Reply #6 –
I am not sure you understand what you are trying to achieve. Or it easily could be me.

Average? average for what? The point is to draw a conclusion for you and your ears. You tried to determine which file was which, you got it right 5 times out of 13. Wrong 8 times. What does that tell you about the difference between the two files as to your ears?

terry - trying to be helpful.

How do I perform an ABX test to compare mp3 vs wav

Reply #7 –
WinABX is fairly basic, and assumed you'll decode whatever lossy file back to WAV before testing. Foobar2000 already has wide-ranging playback capabilities (including gapless playback - i.e. correct timing) so its integrated ABX comparator can take full advantage and make things very painless.

Scoring 5 out of 13 means you were almost certainly guessing and could not hear a difference between the V1 MP3 and the original CD. A score like 11 out of 13 up to 13 out of 13 would mean that -V1 is very probably distinguishable from the original and you weren't guessing. A result like yours is to be expected with V1, as it's supposed to be a transparent setting (in fact most of the tuning historically was done to ensure that V2 is transparent, so V1 is even more on the safe side). Only "problem samples" are likely to allow you to tell the difference when you use a setting of, perhaps, -V3 or better (lower number), and perhaps then only if you've trained yourself what to listen for.

The tuning done on LAME over the years has frequently aimed to render more and more of the problem samples transparent (usually at -V2 or its predecessor, --alt-preset standard) or if that's not possible thanks to limitations of the MP3 specification, unannoying. Problems are also harder to detect with loudspeakers than with headphones.

A lot of people are hard pressed to spot the differences even at -V5, and rarely find the sound annoying if they can spot the difference. I'm one of them.

[edit: clarity]
Dynamic – the artist formerly known as DickD

How do I perform an ABX test to compare mp3 vs wav

Reply #8 –
Quote from: Dynamic on
WinABX is fairly basic, and assumed you'll decode whatever lossy file back to WAV before testing. Foobar2000 already has wide-ranging playback capabilities (including gapless playback - i.e. correct timing) so its integrated ABX comparator can take full advantage and make things very painless.

Scoring 5 out of 13 means you were almost certainly guessing and could not hear a difference between the V1 MP3 and the original CD. A score like 11 out of 13 up to 13 out of 13 would mean that -V1 is very probably distinguishable from the original and you weren't guessing. A result like yours is to be expected with V1, as it's supposed to be a transparent setting (in fact most of the tuning historically was done to ensure that V2 is transparent, so V1 is even more on the safe side). Only "problem samples" are likely to allow you to tell the difference when you use a setting of, perhaps, -V3 or better (lower number), and perhaps then only if you've trained yourself what to listen for.

The tuning done on LAME over the years has frequently aimed to render more and more of the problem samples transparent (usually at -V2 or its predecessor, --alt-preset standard) or if that's not possible thanks to limitations of the MP3 specification, unannoying. Problems are also harder to detect with loudspeakers than with headphones.

A lot of people are hard pressed to spot the differences even at -V5, and rarely find the sound annoying if they can spot the difference. I'm one of them.

[edit: clarity]


Good to know.  Basically what I meant by average, was if my result is typical, and I now know it is.  I couldn't tell the difference. I initially just tried comparing the very first crash cymbal in the song, thinking it would be easy, but I was wrong.  It's a pretty humbling experience, at least as a musician.  We've got that whole ear superiority complex, haha.

How do I perform an ABX test to compare mp3 vs wav

Reply #9 –
Dynamic

Thanks for the extra detail. I would not of been able to provide OP with as much clarity.

terry

How do I perform an ABX test to compare mp3 vs wav

Reply #10 –
Quote from: Sound-of-Muzak on
It's a pretty humbling experience, at least as a musician.  We've got that whole ear superiority complex, haha.


Well, I wouldn't necessarily give up the ear superiority thing, just realise that maybe what you hear better than the average munter is NOT the hyperfine details of audio reproduction, but a whole lot of things that are more musically relevant, and that I can't guess at, being a sub-average munter.

How do I perform an ABX test to compare mp3 vs wav

Reply #11 –
Quote from: Sound-of-Muzak on
Good to know.  Basically what I meant by average, was if my result is typical, and I now know it is.  I couldn't tell the difference. I initially just tried comparing the very first crash cymbal in the song, thinking it would be easy, but I was wrong.  It's a pretty humbling experience, at least as a musician.  We've got that whole ear superiority complex, haha.


IME musicians are often too interested in the music, to be the best detectors of audible differences related to problems with reproducing sound. It is obvious to me that this would be a favorable predisposition for a musican to have.

Don't let any audiophile make you feel bad for being into the music as opposed to the details of how the sound is being reproduced.

I'm thinking about two cave men out in the bush, and a tiger is sneaking up on them. The one caveman says to the other: "Isn't that a tiger?" The other caveman who was the first audiophile says: "No, that isn't how a tiger sounds on my home system".  So much for the first audiophile - eaten by a tiger at an early age. ;-)

How do I perform an ABX test to compare mp3 vs wav

Reply #12 –
Quote from: Arnold B. Krueger on
Quote from: Sound-of-Muzak on
Good to know.  Basically what I meant by average, was if my result is typical, and I now know it is.  I couldn't tell the difference. I initially just tried comparing the very first crash cymbal in the song, thinking it would be easy, but I was wrong.  It's a pretty humbling experience, at least as a musician.  We've got that whole ear superiority complex, haha.


IME musicians are often too interested in the music, to be the best detectors of audible differences related to problems with reproducing sound. It is obvious to me that this would be a favorable predisposition for a musican to have.

Don't let any audiophile make you feel bad for being into the music as opposed to the details of how the sound is being reproduced.

I'm thinking about two cave men out in the bush, and a tiger is sneaking up on them. The one caveman says to the other: "Isn't that a tiger?" The other caveman who was the first audiophile says: "No, that isn't how a tiger sounds on my home system".  So much for the first audiophile - eaten by a tiger at an early age. ;-)


That's actually great insight. You've stated what I couldn't quite put my finger on. I think bands like Pink Floyd, who created some of the best sounding albums of all time (DSOTM), and at the same time are very interested in the music (chord changes, harmonies, etc.) are few and far between. I think that's what makes a great producer/engineer, someone who is "in" the music and at the same time immersed in the details of the sound. I think someone like Steven Wilson from Porcupine Tree is another great example.