This topic is locked from further discussion.

#1 Posted by kraken2109 (13007 posts) -

It's now an aging debate:

Can you tell the difference between lossless and lossy digital music and what impact does playback equipment have on this?

I guess this is of interest to the people on this board, and I need help with a project I'm doing on this topic. Part of it is gathering results for a test I've created.

Basically I took a friend into the studio and recorded her playing the Cello. I then did a small amount of mixing and encoded it as a CD quality WAV (16-bit 44.1khz). I then created various mp3 versions. Finally I converted the Mp3s back to uncompressed wavs, so that the files look the same to prevent cheating, but underneath they are mp3 quality with varying bit rates.

I ask of you a simple task, load the 5 files as a playlist in the audio software of your choice and fill in my form rating them, then send it to the email address at the bottom of the page.

I'm sorry if anyone considers this spam, if so just ignore it, but I think some people on this board will be interested in taking part, seeing the results and testing their equipment and ears!

Here is a link to the form, it contains links to the 5 files (stored in dropbox) (Try this one if it doesn't work: https://dl.dropbox.com/u/16808054/EPQ/TestForm.docx )

Sorry about the file sizes, but it's the best way to avoid cheating.

Thanks for your help, maybe don't post what you thought in this thread as it may affect the judgement of others, but feel free to say whether you noticed large differences or not, without meantioning the names of the files.

Thanks so much guys (if you want to do the test another time on different equipment just ask and i'll send you links to the files with different names to make it a fair test).

tl;dr version: Download form, download music, listen, fill in form, email me (or PM if you want)

Even if you don't want to take part in the test, feel free to use this thread to dicuss your opinion on the matter of compressing digital audio.

 EDIT: Fixed link

#2 Posted by Wolfetan (7522 posts) -

I've listened to Lossless vs. Lossy, Lossless is technically superior. 256 and 320 there is a difference. 320 and up there isn't.

#3 Posted by Riadon2 (1609 posts) -

I can tell a large difference between 320 kbps MP3 and FLAC.  I can't really tell much a difference between 16/24 bit and 44 khz/192 khz, though.

It all depends on your audio equipment.  Hell, someone with a STAX might be able to tell a difference between FLAC and WAV.

#4 Posted by Wolfetan (7522 posts) -

I can tell a large difference between 320 kbps MP3 and FLAC.  I can't really tell much a difference between 16/24 bit and 44 khz/192 khz, though.

It all depends on your audio equipment.  Hell, someone with a STAX might be able to tell a difference between FLAC and WAV.

Riadon2
Are you sure its 320 kbps? Not saying your wrong, but even some of the biggest audiophiles with STAX SR-009's claim they can't tell a difference between FLAC and 320.
#5 Posted by YoshiYogurt (5973 posts) -
Can't tell the difference between lossless and 320 or 256, but there is a noticeable difference between 128 and lossless.
#6 Posted by Wolfetan (7522 posts) -

It's now an aging debate:

Can you tell the difference between lossless and lossy digital music and what impact does playback equipment have on this?

I guess this is of interest to the people on this board, and I need help with a project I'm doing on this topic. Part of it is gathering results for a test I've created.

Basically I took a friend into the studio and recorded her playing the Cello. I then did a small amount of mixing and encoded it as a CD quality WAV (16-bit 44.1khz). I then created various mp3 versions. Finally I converted the Mp3s back to uncompressed wavs, so that the files look the same to prevent cheating, but underneath they are mp3 quality with varying bit rates.

I ask of you a simple task, load the 5 files as a playlist in the audio software of your choice and fill in my form rating them, then send it to the email address at the bottom of the page.

I'm sorry if anyone considers this spam, if so just ignore it, but I think some people on this board will be interested in taking part, seeing the results and testing their equipment and ears!

Here is a link to the form, it contains links to the 5 files (stored in dropbox)

Sorry about the file sizes, but it's the best way to avoid cheating.

Thanks for your help, maybe don't post what you thought in this thread as it may affect the judgement of others, but feel free to say whether you noticed large differences or not, without meantioning the names of the files.

Thanks so much guys (if you want to do the test another time on different equipment just ask and i'll send you links to the files with different names to make it a fair test).

tl;dr version: Download form, download music, listen, fill in form, email me (or PM if you want)

Even if you don't want to take part in the test, feel free to use this thread to dicuss your opinion on the matter of compressing digital audio.

 

kraken2109
The link isn't working, I wanna try now..
#7 Posted by Riadon2 (1609 posts) -

[QUOTE="Riadon2"]

I can tell a large difference between 320 kbps MP3 and FLAC.  I can't really tell much a difference between 16/24 bit and 44 khz/192 khz, though.

It all depends on your audio equipment.  Hell, someone with a STAX might be able to tell a difference between FLAC and WAV.

Wolfetan

Are you sure its 320 kbps? Not saying your wrong, but even some of the biggest audiophiles with STAX SR-009's claim they can't tell a difference between FLAC and 320.

It really depends on the song.  I was exaggerating when I said there was a "large" difference (in most cases), but it is noticeable depending on what you are listening to.

#8 Posted by Nick3306 (2562 posts) -
This test doesn't really work because it also depends on your audio setup.
#9 Posted by Wolfetan (7522 posts) -

[QUOTE="Wolfetan"][QUOTE="Riadon2"]

I can tell a large difference between 320 kbps MP3 and FLAC.  I can't really tell much a difference between 16/24 bit and 44 khz/192 khz, though.

It all depends on your audio equipment.  Hell, someone with a STAX might be able to tell a difference between FLAC and WAV.

Riadon2

Are you sure its 320 kbps? Not saying your wrong, but even some of the biggest audiophiles with STAX SR-009's claim they can't tell a difference between FLAC and 320.

It really depends on the song.  I was exaggerating when I said there was a "large" difference (in most cases), but it is noticeable depending on what you are listening to.

Ah, exaggerating..
#10 Posted by Mcspanky37 (1695 posts) -
This test doesn't really work because it also depends on your audio setup.Nick3306
#11 Posted by Wolfetan (7522 posts) -
[QUOTE="Nick3306"]This test doesn't really work because it also depends on your audio setup.Mcspanky37

Get some better audio equipment then! Haha.
#12 Posted by Mcspanky37 (1695 posts) -
[QUOTE="Mcspanky37"][QUOTE="Nick3306"]This test doesn't really work because it also depends on your audio setup.Wolfetan

Get some better audio equipment then! Haha.

Why?
#13 Posted by Wolfetan (7522 posts) -
[QUOTE="Wolfetan"][QUOTE="Mcspanky37"]Mcspanky37
Get some better audio equipment then! Haha.

Why?

To listen to music, movies, and game more deeply.
#14 Posted by dxmcat (1121 posts) -

I can def tell a diff between 320k & Flac. I avoid having mp3s under 320k tho. Difference is less noticeable on my speakers / sound in room etc, but easy with headphones.

#15 Posted by Heirren (16533 posts) -
There's a big difference, especially once you recognize the traits of lossless.
#17 Posted by KHAndAnime (13415 posts) -
I've done double-blind AB listening tests between lossy and lossless using a plugin for foobar (http://www.foobar2000.org/components/view/foo_abx). Depending on what your gear is, I'd say it's possible to tell the difference between 320kbps and FLAC. I have a very neutral-sounding DAC with studio monitors that are made to reveal low-quality sources. Some speakers/headphones are made to minimize the difference in sound between FLAC and MP3, which others are made to maximize that difference.
#18 Posted by KHAndAnime (13415 posts) -

[QUOTE="Mcspanky37"][QUOTE="Wolfetan"] Get some better audio equipment then! Haha.Wolfetan
Why?

To listen to music, movies, and game more deeply.

I think his point is that getting better audio gear doesn't fix the OP's experiment. There are no useful statistics to be gained from this because you have two mutually exclusive variables effecting the outcome of the results.

#19 Posted by kraken2109 (13007 posts) -

[QUOTE="Wolfetan"][QUOTE="Mcspanky37"] Why?KHAndAnime

To listen to music, movies, and game more deeply.

I think his point is that getting better audio gear doesn't fix the OP's experiment. There are no useful statistics to be gained from this because you have two mutually exclusive variables effecting the outcome of the results.

Actually I am taking into account equipment, basically I have 4 sets of results, high end headphones, low end headphones, high end speakers and low end speakers. The idea is to compare them and make 2 conclusions, can people tell a difference, and if so does the equipment matter, IE did people with high end equipment do better.
#20 Posted by kraken2109 (13007 posts) -

Guys I fixed the link, please give it a go, it should take 15 mins max.

Some people are making claims and i'd like to see if they're right ;)

#21 Posted by dramaybaz (6020 posts) -
Yo need to have a good setup in the first place to notice the difference.
#22 Posted by kraken2109 (13007 posts) -

Yo need to have a good setup in the first place to notice the difference.dramaybaz
Do my test and find out! :x

#23 Posted by APiranhaAteMyVa (2869 posts) -
Filled in the form and sent, I won't talk about what I mentioned in the email here. I have done ABX test with files before and it is usually around V3 (about 175kb/s) when I can no longer tell a difference with MP3. AAC I haven't really tested but I am happy using 135KB/s VBR AAC files if I need to preserve space.
#24 Posted by KHAndAnime (13415 posts) -
[QUOTE="KHAndAnime"]

[QUOTE="Wolfetan"] To listen to music, movies, and game more deeply. kraken2109

I think his point is that getting better audio gear doesn't fix the OP's experiment. There are no useful statistics to be gained from this because you have two mutually exclusive variables effecting the outcome of the results.

Actually I am taking into account equipment, basically I have 4 sets of results, high end headphones, low end headphones, high end speakers and low end speakers. The idea is to compare them and make 2 conclusions, can people tell a difference, and if so does the equipment matter, IE did people with high end equipment do better.

Still not good enough. You need to create categories to differentiate between hi-fi and studio speakers/headphones. One high-end speaker will make 192kbps sound dangerously close to FLAC whereas another will make 192kbps sound like garbage.
#25 Posted by kraken2109 (13007 posts) -
[QUOTE="kraken2109"][QUOTE="KHAndAnime"] I think his point is that getting better audio gear doesn't fix the OP's experiment. There are no useful statistics to be gained from this because you have two mutually exclusive variables effecting the outcome of the results.KHAndAnime
Actually I am taking into account equipment, basically I have 4 sets of results, high end headphones, low end headphones, high end speakers and low end speakers. The idea is to compare them and make 2 conclusions, can people tell a difference, and if so does the equipment matter, IE did people with high end equipment do better.

Still not good enough. You need to create categories to differentiate between hi-fi and studio speakers/headphones. One high-end speaker will make 192kbps sound dangerously close to FLAC whereas another will make 192kbps sound like garbage.

I am keeping note of the exact equipment linked to every result, at the moment I am just trying to collect as many results as I can.
#26 Posted by dramaybaz (6020 posts) -

[QUOTE="dramaybaz"]Yo need to have a good setup in the first place to notice the difference.kraken2109

Do my test and find out! :x

Link doesn't even work for me!
#27 Posted by kraken2109 (13007 posts) -

[QUOTE="kraken2109"]

[QUOTE="dramaybaz"]Yo need to have a good setup in the first place to notice the difference.dramaybaz

Do my test and find out! :x

Link doesn't even work for me!

Odd, try this

(I think the space in the file name caused issues)

EDIT: Gamespot linking is breaking it, try this one:

https://dl.dropbox.com/u/16808054/EPQ/TestForm.docx

#28 Posted by dramaybaz (6020 posts) -

[QUOTE="dramaybaz"][QUOTE="kraken2109"] Do my test and find out! :x

kraken2109

Link doesn't even work for me!

Odd, try this

(I think the space in the file name caused issues)

EDIT: Gamespot linking is breaking it, try this one:

https://dl.dropbox.com/u/16808054/EPQ/TestForm.docx

Ok, the last one worked. I'll try my speakers, and headphones, and get back in a day.
#29 Posted by kraken2109 (13007 posts) -
[QUOTE="kraken2109"]

[QUOTE="dramaybaz"] Link doesn't even work for me!dramaybaz

Odd, try this

(I think the space in the file name caused issues)

EDIT: Gamespot linking is breaking it, try this one:

https://dl.dropbox.com/u/16808054/EPQ/TestForm.docx

Ok, the last one worked. I'll try my speakers, and headphones, and get back in a day.

Thanks very much
#30 Posted by kraken2109 (13007 posts) -

In case anyone wants to do the test a second time without being influenced by their first time, here is a set of links to the same files in a different order:

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/A.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/B.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/C.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/D.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/E.wav

#31 Posted by sukraj (22131 posts) -

[QUOTE="Nick3306"]This test doesn't really work because it also depends on your audio setup.Mcspanky37

i totally agree.

#32 Posted by kraken2109 (13007 posts) -

[QUOTE="Mcspanky37"][QUOTE="Nick3306"]This test doesn't really work because it also depends on your audio setup.sukraj

i totally agree.

As I've said, that's kinda the point...
#33 Posted by Wolfetan (7522 posts) -

In case anyone wants to do the test a second time without being influenced by their first time, here is a set of links to the same files in a different order:

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/A.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/B.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/C.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/D.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/E.wav

kraken2109
Whats the order for this one?
#34 Posted by kraken2109 (13007 posts) -
[QUOTE="kraken2109"]

In case anyone wants to do the test a second time without being influenced by their first time, here is a set of links to the same files in a different order:

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/A.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/B.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/C.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/D.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/E.wav

Wolfetan
Whats the order for this one?

What do you mean?
#35 Posted by Lach0121 (9793 posts) -

I notice a difference between mp3 and FLAC (or other lossless).  I think it also matters what you are listening to. What you are listening to it on. IF you are use to listening to more than just Mp3.  I can tell the difference though 320kps is getting up there, and it is hard to tell the difference a lot of the time. The lower you go from 320, the more notice(able) it becomes.

However I do not notice a difference between FLAC 16-bit and FLAC 24-bit.  (if there is one I don't notice it) however 16 bit is standard CD quality.

 

I must also say that I didn't take your test, I just used my own. From other artists, and my own music that I make.  I render to many formats, various Mp3(lossy) as well as Wavpack, wav, and FLAC (all lossless)

#36 Posted by Zaral_1 (186 posts) -
I honestly find this claim hard to believe, that someone can tell the difference between lossless and losy. At the my old university I used to develop a new 3D video codec to improve 3D video and learned a lot about audio and video compression. First if you look at the basis of MP3 the sampling rate follows the NyquistShannon sampling theorem here http://en.wikipedia.org/wiki/Sampling_theorem Which proves you can create a bandlimited function can be perfectly reconstructed from an infinite sequence of samples if the bandlimit, B, is no greater than ½ the sampling rate (samples per second). aka 40 khz which is rounded up to 44.1 khz Second when creating MP3's you use 4 formulas for removing audio from a source Limits of perception - throw away audio that is above or below human range of hearing Sound localization - crazy formulas that identify a sound source in X,Y,Z coordinates, Masking effects - Throw away audio that can not be heard since it is masked by another sound, think there is a bomb exploding and someone is whispering 20 feet behind the bomb, the bomb mask the whisper Missing fundamentals - when its overtones suggest a fundamental frequency but the sound lacks a component at the fundamental frequency itself. All these things that are thrown away you can not perceive in real life since our hearing is not advanced enough since we are weak humans. This is all proven by the scientific community. Unless you guys are dogs or mutants I can not see how anyone can tell the difference unless your audio hardware is not efficient. To back up my claim here are some resources http://en.wikipedia.org/wiki/MP3 http://en.wikipedia.org/wiki/Psychoacoustic http://en.wikipedia.org/wiki/Sound_localization http://en.wikipedia.org/wiki/Auditory_masking http://en.wikipedia.org/wiki/Missing_fundamental http://en.wikipedia.org/wiki/44,100_Hz and my research paper if anyone cares http://dl.acm.org/citation.cfm?id=1989240.1989265
#37 Posted by Mcspanky37 (1695 posts) -
[QUOTE="Zaral_1"]I honestly find this claim hard to believe, that someone can tell the difference between lossless and losy. At the my old university I used to develop a new 3D video codec to improve 3D video and learned a lot about audio and video compression. First if you look at the basis of MP3 the sampling rate follows the NyquistShannon sampling theorem here http://en.wikipedia.org/wiki/Sampling_theorem Which proves you can create a bandlimited function can be perfectly reconstructed from an infinite sequence of samples if the bandlimit, B, is no greater than ½ the sampling rate (samples per second). aka 40 khz which is rounded up to 44.1 khz Second when creating MP3's you use 4 formulas for removing audio from a source Limits of perception - throw away audio that is above or below human range of hearing Sound localization - crazy formulas that identify a sound source in X,Y,Z coordinates, Masking effects - Throw away audio that can not be heard since it is masked by another sound, think there is a bomb exploding and someone is whispering 20 feet behind the bomb, the bomb mask the whisper Missing fundamentals - when its overtones suggest a fundamental frequency but the sound lacks a component at the fundamental frequency itself. All these things that are thrown away you can not perceive in real life since our hearing is not advanced enough since we are weak humans. This is all proven by the scientific community. Unless you guys are dogs or mutants I can not see how anyone can tell the difference unless your audio hardware is not efficient. To back up my claim here are some resources http://en.wikipedia.org/wiki/MP3 http://en.wikipedia.org/wiki/Psychoacoustic http://en.wikipedia.org/wiki/Sound_localization http://en.wikipedia.org/wiki/Auditory_masking http://en.wikipedia.org/wiki/Missing_fundamental http://en.wikipedia.org/wiki/44,100_Hz and my research paper if anyone cares http://dl.acm.org/citation.cfm?id=1989240.1989265

This has been argued to death! What's so hard to believe? Compress a song to 96kbps and you'd have to be deaf not to hear the difference between the lossy file and the lossless one. MP3 files only contain a very small fraction of the information found in the source. It would be one thing if the effects of compression were inaudible, but no matter how you slice it, my 192kbps MP3 files are a far cry away from my FLAC music. Some audiophiles are convinced that changing the electricity in your house will effect your sound. The quality of the source is certainly important. Sounds like you need to do less research and more listening. :P Or maybe you aren't doing proper research, because less expensive audio-gear would be less capable in discerning the difference in FLAC and MP3. Being able to tell the difference between 320kbps and FLAC could perhaps be considered a stretch, but some people performing ABX tests claim can consistently tell the difference.
#38 Posted by Wolfetan (7522 posts) -
[QUOTE="Wolfetan"][QUOTE="kraken2109"]

In case anyone wants to do the test a second time without being influenced by their first time, here is a set of links to the same files in a different order:

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/A.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/B.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/C.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/D.wav

https://dl.dropbox.com/u/16808054/EPQ/Cello/Speaker%20Test/E.wav

kraken2109
Whats the order for this one?

What do you mean?

Like which is FLAC? Which is MP3? Stuff like that.
#39 Posted by kraken2109 (13007 posts) -
[QUOTE="kraken2109"][QUOTE="Wolfetan"] Whats the order for this one?Wolfetan
What do you mean?

Like which is FLAC? Which is MP3? Stuff like that.

I can't tell you that until you do the test and email me your results.
#40 Posted by kraken2109 (13007 posts) -

I honestly find this claim hard to believe, that someone can tell the difference between lossless and losy. At the my old university I used to develop a new 3D video codec to improve 3D video and learned a lot about audio and video compression. First if you look at the basis of MP3 the sampling rate follows the NyquistShannon sampling theorem here http://en.wikipedia.org/wiki/Sampling_theorem Which proves you can create a bandlimited function can be perfectly reconstructed from an infinite sequence of samples if the bandlimit, B, is no greater than ½ the sampling rate (samples per second). aka 40 khz which is rounded up to 44.1 khz Second when creating MP3's you use 4 formulas for removing audio from a source Limits of perception - throw away audio that is above or below human range of hearing Sound localization - crazy formulas that identify a sound source in X,Y,Z coordinates, Masking effects - Throw away audio that can not be heard since it is masked by another sound, think there is a bomb exploding and someone is whispering 20 feet behind the bomb, the bomb mask the whisper Missing fundamentals - when its overtones suggest a fundamental frequency but the sound lacks a component at the fundamental frequency itself. All these things that are thrown away you can not perceive in real life since our hearing is not advanced enough since we are weak humans. This is all proven by the scientific community. Unless you guys are dogs or mutants I can not see how anyone can tell the difference unless your audio hardware is not efficient. To back up my claim here are some resources http://en.wikipedia.org/wiki/MP3 http://en.wikipedia.org/wiki/Psychoacoustic http://en.wikipedia.org/wiki/Sound_localization http://en.wikipedia.org/wiki/Auditory_masking http://en.wikipedia.org/wiki/Missing_fundamental http://en.wikipedia.org/wiki/44,100_Hz and my research paper if anyone cares http://dl.acm.org/citation.cfm?id=1989240.1989265Zaral_1

I'm researching it too, hence this thread.

Take part in the test and help me find out.

#41 Posted by Microsoft1234 (7674 posts) -
I can't tell the difference between mp3 320 and lossless but i can usually tell the differece (which one is better) betweeen those horrible 128 kbps itunes used to sell and 320.
#42 Posted by kraken2109 (13007 posts) -
[QUOTE="Microsoft1234"]I can't tell the difference between mp3 320 and lossless but i can usually tell the differece (which one is better) betweeen those horrible 128 kbps itunes used to sell and 320.

Take my test and we'll see.