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: Differentiation of music playback quality (Read 6748 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Differentiation of music playback quality

First of all I just wanted to say that I'm not a regular user, so here's to hoping that I'm not screwing up in putting this topic here. It involves several subjects that have their specific sub-forums, but what I'm trying to inquire here is broader rather than specific.

So, maybe it's the just the tiredness and self-induced paranoia, but I think I'm having different listening experiences with the same media. For instance, the same audio CD played through EAC seems to have a different output when played through iTunes. I've tried searching for sources to back this up but I can't find anything. So basically my question number one is, does EAC provide different CD playback quality than other software? I could swear the same CD sounds "thiner" and less "boomy" on EAC than it does on iTunes, for instance.

Following that stream of consciousness, after properly ripping a CD to MP3@320k with EAC and playing it through different software (iTunes vs WMP in this case) is it supposed to sound different or the same? Because again I could swear that iTunes (iteration 11) makes the rip sound "boomier" and "fuller", although in this case "fuller" seems to correlate with sounding dynamically more compressed. Is this me tripping or does it have any kind of basis? Is it just skewed perception?

So basically what I'm asking is if with the same MP3 file played through different software the sound is equal or different. I'm also asking the same for audio CD reproduction, and in this specific case between playing it through EAC and any other software. I know these are probably just dumb questions anyway, but given the reputation this site has when it comes to expertise in terms of music quality I though asking here would be the right thing to do.

Feel free to drop some feedback on this matter, even if these are just dumb questions.

Differentiation of music playback quality

Reply #1 –
The MP3 spec defines how an MP3 file must be decoded, so if a piece of software differs from the spec it is broken. 

MP3 is so old at this point though that basically all software works correctly.

Differentiation of music playback quality

Reply #2 –
Quote from: saratoga on
The MP3 spec defines how an MP3 file must be decoded, so if a piece of software differs from the spec it is broken. 

MP3 is so old at this point though that basically all software works correctly.

So you're saying that all software should output the same MP3 file similarly, right?

How about audio CD playback, which is the second part of my question, is it the same? Universally used specs make for equal results?

I'm actually curious as to where each person's perception comes into play in this matter. Meaning that you actually witness a lot of people saying that X is better sounding than Y and all that. Is it just wanting to see (or in this case hear) something that's just not there?

Differentiation of music playback quality

Reply #3 –
Well I'm not sure if iTunes is using some sound effects (equalizer,  ...) by default - I guess not - but to be certain I would suggest to use some other player for comparison; foobar2000, boom, Winamp, ...

You should also check if one of your players is using WASAPI or ASIO output and the other not. Usually this doesn't make any difference but some machines have some (stupid) effects in the driver software enabled (like Dolby Logic, THX, ...) which get circumvented by WASAPI and ASIO. You probably want to disable those.

Differentiation of music playback quality

Reply #4 –
I'd do a search for the CD question.  The condition of the disk and how hard a ripper tries to reread damaged sectors can have a large impact on its accuracy, at least for CDs in poor condition.

Differentiation of music playback quality

Reply #5 –
Quote from: Wander on
Well I'm not sure if iTunes is using some sound effects (equalizer,  ...) by default - I guess not - but to be certain I would suggest to use some other player for comparison; foobar2000, boom, Winamp, ...

You should also check if one of your players is using WASAPI or ASIO output and the other not. Usually this doesn't make any difference but some machines have some (stupid) effects in the driver software enabled (like Dolby Logic, THX, ...) which get circumvented by WASAPI and ASIO. You probably want to disable those.


I was also thinking something along those lines.

For CD playback, for instance, as the OP mentioned EAC, we know rippers usually are plain vanilla when it comes to disc playback - no DSP, EQ or whatever; wouldn't the differences mentioned by the OP be more down to that than anything else?

As for encoded files, doesn't the mileage vary slightly from one software player's lossy decoder/plugin to another? (the old days of Winamp with MAD plugin spring to mind). I mean, doesn't that add up to the equation also?

Edit: added last sentence.
Listen to the music, not the media it's on.
União e reconstrução

Differentiation of music playback quality

Reply #6 –
Thanks for the replies.

I use everything either "flat" or turned off in terms of effects when it comes to output software. I have the sound configured as 44Khz/16bit/2 speakers, to have everything at standard levels. I have some slight EQ settings configured on the Audio Center of my Xonar Essence STX, as well as a setting called "Flexbass" turned on at a certain threshold (it's a LFE crossover cut-out that supposedly redirects bass frequencies from the speakers to the subwoofer). I also have the main system volume at 50% to reduce distortion and use the speakers' volume knob to change it up and down. But those are "global settings" that affect the entire environment, i.e. every piece of software I run on my system.

@ Wander - You'll to spell it out for me, since I don't know how to check for WASAPI/ASIO usage... I know they're audio norms (please correct me if I'm wrong) but other than that I'm pretty much in the dark.

@ saratoga - Just so we're clear, I'm talking about pristine CDs, not stuff that has to do several runs to be able to be ripped. So rip accuracy isn't a factor here.

What's funky to my ears is that I can ear a noticeable difference between EAC and iTunes while playing an audio CD. Today I received a new CD in the mail and the mids (mainly the guitars) seem to be more buried while being played in iTunes than on EAC. The bass is also significantly clearer on EAC (less "boomy"), which then again makes me wonder why. In fact I've had this feeling about EAC for a while now, although I never tried to dig up some reasons why it would be acting like that. But, you know, people have been complaining about iTunes 11 taking a nosedive in terms of audio quality, which I have no idea if it's true or just hearsay that gets repeated so often that it become almost factual. I like using iTunes for the sake of organizing, but at heart I'm all about sound quality. And more often than not I feel like iTunes 11 is presenting some albums with a sound that's louder and more distorted (compressed) than it should be.

I'm sorry for all the repetition but I'm really trying to understand if this is just my perception or if there's actually any difference.

Differentiation of music playback quality

Reply #7 –
Quote from: includemeout on
For CD playback, for instance, as the OP mentioned EAC, we know rippers usually are plain vanilla when it comes to disc playback - no DSP, EQ or whatever; wouldn't the differences mentioned by the OP be more down to that than anything else?

As for encoded files, doesn't the mileage vary slightly from one software player's lossy decoder/plugin to another? (the old days of Winamp with MAD plugin spring to mind). I mean, doesn't that add up to the equation also?

Thanks for pitching in. Love your avatar! 

Regarding audio CD playback on EAC, that's what I thought initially, but as I say in my last post I don't use any kind of environmental effects on my soundboard's settings, nor on third party media players. That's why it's bugging me.

Picking up on your second sentence, would each media player's coding/programming make a difference in terms of audio output when it comes to playing an audio CD? I'm thinking along the lines of making it sound "punchier" by default, if you know what I mean. Because that's mostly how it sounds on iTunes, "punchier". I know that this shouldn't happen because CD audio tracks are uncompressed (but again, please correct me if I'm wrong) and would only apply to lossy files that need decoding. But could this be a factor? That each media player presents default global settings that make the output slightly different, even though it's an uncompressed source?

Differentiation of music playback quality

Reply #8 –
Quote
For instance, the same audio CD played through EAC seems to have a different output when played through iTunes. I've tried searching for sources to back this up but I can't find anything. So basically my question number one is, does EAC provide different CD playback quality than other software?
It shouldn't make any difference, but that doesn't mean it's your imagination.    It could just be a volume difference, or iTunes might be doing something you're not aware of.  If you have SoundCheck enabled in iTunes, that will (potentially) automatically adjust the volume, although I don't know if SoundCheck does anything when playing a CD.

You can make a test-tone with Audacity and burn it to a CD, or just make a WAV file to compare different player applications.  It would be easier to nail-down a volume difference with a test-tone.  Ideally, you'd have a multimeter to measure the voltage output, because you can't instantly switch back-and-forth between CD player applications to listen for volume differences.  I generally use 400 or 500Hz, but it might also be helpful to make a pink noise file. 

If you have a stand alone CD player (or DVD player), that would probably be a good "reference standard", because you can be pretty sure there is no EQ or other "funny" processing.  But again, you really need to level-match if you are going to compare the sound of your CD player to the sound of your computer, and that may not be easy without a meter.

Differentiation of music playback quality

Reply #9 –
SoundCheck is off, as I hate volume normalization of any kind. Each album has its own sound, you know, so normalize them to have everything at the same volume is something I just don't do.

Interesting though, that configuration page on iTunes has one thing I hadn't tried touching before. It allows for the audio being played to use either Direct Sound or Windows Audio Session (which I'm assuming it's DS vs WASAPI). I tried changing between the two and checked a small tune with a fretless bass soloing on top of an accompanying acoustic guitar and just a little bit of percussion. Now, I'm using earphones, but still I believe there's a noticeable difference. Using DS makes the song have more "attack", especially in the cymbals which sound more "shimmering". If I understood it correctly though, if iTunes (or any other media player) is using WASAPI is it bypassing the configurations I have on the Xonar Audio Center (EQ/FlexBass)? Or in other words, does using WASAPI bypasses Xonar's Audio Center altogether?

As for the tests you mention DVDdoug, I get the point but I don't think I can do them in practical terms. It seems a whole lot like doing things I'm barely aware of, which would just result in me not being able to properly interpret the results. I thanks you though.

Differentiation of music playback quality

Reply #10 –
Quote from: androdion on
SoundCheck is off, as I hate volume normalization of any kind. Each album has its own sound, you know, so normalize them to have everything at the same volume is something I just don't do.

Interesting though, that configuration page on iTunes has one thing I hadn't tried touching before. It allows for the audio being played to use either Direct Sound or Windows Audio Session (which I'm assuming it's DS vs WASAPI). I tried changing between the two and checked a small tune with a fretless bass soloing on top of an accompanying acoustic guitar and just a little bit of percussion. Now, I'm using earphones, but still I believe there's a noticeable difference. Using DS makes the song have more "attack", especially in the cymbals which sound more "shimmering". If I understood it correctly though, if iTunes (or any other media player) is using WASAPI is it bypassing the configurations I have on the Xonar Audio Center (EQ/FlexBass)? Or in other words, does using WASAPI bypasses Xonar's Audio Center altogether?

As for the tests you mention DVDdoug, I get the point but I don't think I can do them in practical terms. It seems a whole lot like doing things I'm barely aware of, which would just result in me not being able to properly interpret the results. I thanks you though.

Just wanted to leave an addendum regarding WASAPI on iTunes.

It uses shared mode, so Xonar's Audio Center settings are still able to affect the sound being played. However, and again I was using earphones so I can't really tell for sure, it seems like there's an improvement in sound definition. I'll have to verify this with my speakers, but for the time being I can't make any noise since I have people sleeping at the house. I shall do it tomorrow and report back here with my findings.

If I'm lucky enough maybe changing this setting is enough to dismiss the compression/distortion I was experiencing with iTunes. Fingers crossed!

Differentiation of music playback quality

Reply #11 –
Record the output.  RMAA will be able to easily tell you if there is any processing going on.

Differentiation of music playback quality

Reply #12 –
Quote from: androdion on
For instance, the same audio CD played through EAC seems to have a different output when played through iTunes.

First, in order to eliminate one source of possible confusion: You mean the very same CD rip? If you merely mean the same title - EAC plays it, and you compare to a file you have purchased - then they might be different masterings, and those do sound different as they are not the same signal.

Apart from that, yes, volume matching is crucial. Volumes need not be more than slightly off, and the louder will sound different. You say you don't do volume normalization, but it is a bit unclear what you - or the applications - do. There are actually a couple of possible effects that would normally lead to, subjectively, the opposite result. For example, lossy-encoding a PCM with full peak could lead to a file with peaks above 1, and a ReplayGain-aware (or whatever the iTunes analogue is) player could then reduce volume to avoid clipping, but if that sounds subjectively different it would be likely to be perceived as less boomy.

(And don't get me started at pre-emphasis, which is fortunately not common.)

Differentiation of music playback quality

Reply #13 –
Quote from: androdion on
First of all I just wanted to say that I'm not a regular user, so here's to hoping that I'm not screwing up in putting this topic here. It involves several subjects that have their specific sub-forums, but what I'm trying to inquire here is broader rather than specific.

So, maybe it's the just the tiredness and self-induced paranoia, but I think I'm having different listening experiences with the same media. For instance, the same audio CD played through EAC seems to have a different output when played through iTunes. I've tried searching for sources to back this up but I can't find anything. So basically my question number one is, does EAC provide different CD playback quality than other software? I could swear the same CD sounds "thiner" and less "boomy" on EAC than it does on iTunes, for instance.

Following that stream of consciousness, after properly ripping a CD to MP3@320k with EAC and playing it through different software (iTunes vs WMP in this case) is it supposed to sound different or the same? Because again I could swear that iTunes (iteration 11) makes the rip sound "boomier" and "fuller", although in this case "fuller" seems to correlate with sounding dynamically more compressed. Is this me tripping or does it have any kind of basis? Is it just skewed perception?


In principle both high bitrate MP3s and CDs should be sonically transparent or very close to it, in all technically competent implementations.

With all due respect the weakest part of your post appears to be the technology which you appear to be using for the comparisons. I don't see where it is level-matched, time-synched, and adequately bias controlled.

Doing good listening tests involving comparisons of MP3s to the original .wav files is pretty easy:

(1) Select a .wav file which should be BTW audibly indistinguishable from a well-played CD

(2) Convert the selected .wav file or the corresponding source CD track in a  MP3 file at the desired bitrate.

(3) Convert the MP3 file back to .wav format.

(4) Examine the before and after .wav files using an audio editor such as Audacity to ensure consistent levels and timings.  We need 1-2% tolerance on level and 5 milliseconds or better on timing.

(5) Compare using FOBAR2000 and its ABX plug in. A PC with a really good PC audio interface and onitoring loudspeakers, headphones or earphones is highly recommended.

Differentiation of music playback quality

Reply #14 –
Thanks for the replies!

Quote from: Porcus on
First, in order to eliminate one source of possible confusion: You mean the very same CD rip? If you merely mean the same title - EAC plays it, and you compare to a file you have purchased - then they might be different masterings, and those do sound different as they are not the same signal.

Apart from that, yes, volume matching is crucial. Volumes need not be more than slightly off, and the louder will sound different. You say you don't do volume normalization, but it is a bit unclear what you - or the applications - do. There are actually a couple of possible effects that would normally lead to, subjectively, the opposite result. For example, lossy-encoding a PCM with full peak could lead to a file with peaks above 1, and a ReplayGain-aware (or whatever the iTunes analogue is) player could then reduce volume to avoid clipping, but if that sounds subjectively different it would be likely to be perceived as less boomy.

(And don't get me started at pre-emphasis, which is fortunately not common.)

No, I mean an actual audio CD. Original, silver coloured,  factory pressed disc. It sounds more "natural" and less "boomy" on EAC than it does on iTunes, almost as if iTunes was doing some kind of processing on top of reading it, which I know sounds totally weird. It's just how it sounds to me.

I don't use any kind of effects turned on in either the Xonar Audio Center or any third party software, apart from the already described EQ and FlexBass settings on Xonar's software. As I said before those are "global settings" that affect the sound of the entire system and all its applications. On media players I have EQ turned off, as well as SVN (Smart Volume Normalization) effects or any kind of enhancements or cut-outs. Basically my system is "vanilla" except for global EQ/FlexBass settings. And with that being said I can notice some slight differences between iTunes 11 and the remaining software I use. If iTunes is to blame or not is still what I'm trying to understand here.

I only do CD rips with EAC configured with the best 320k options, as chosen and advertised by the specialists and guides here, so it's not a matter of bad encoding. I'm also only comparing rips I made from original CDs I own, so all in all the encoding factor is pretty much off the table here.

Quote from: Arnold B. Krueger on
In principle both high bitrate MP3s and CDs should be sonically transparent or very close to it, in all technically competent implementations.

With all due respect the weakest part of your post appears to be the technology which you appear to be using for the comparisons. I don't see where it is level-matched, time-synched, and adequately bias controlled.

Doing good listening tests involving comparisons of MP3s to the original .wav files is pretty easy:

(1) Select a .wav file which should be BTW audibly indistinguishable from a well-played CD

(2) Convert the selected .wav file or the corresponding source CD track in a  MP3 file at the desired bitrate.

(3) Convert the MP3 file back to .wav format.

(4) Examine the before and after .wav files using an audio editor such as Audacity to ensure consistent levels and timings.  We need 1-2% tolerance on level and 5 milliseconds or better on timing.

(5) Compare using FOBAR2000 and its ABX plug in. A PC with a really good PC audio interface and onitoring loudspeakers, headphones or earphones is highly recommended.

I would need to have really thin skin to take your statement the wrong way.

It's the weakest part as you say because I'm mainly functioning by ear (although I have above average hearing), and based on my perception of about 18 months using this hardware/software. Recently I updated iTunes to version 11, and honestly speaking I think I've always noticed a slightly higher compression rate in the MP3 rips being played since then. It's a matter of only doing something about it now. Last few months have been a bit hectic, so now I can approach the matter with a clear head, you know?!

I just came home now, so I'm on to do some tests and listening comparisons about the things posted above. Mainly, trying out WASAPI on iTunes and if it actually makes a difference in sound quality.

Speaking of WASAPI, I do have a technical question. Even though it's using shared mode, it still means that the sound reproduction is bit perfect (I think that's the term) as long as there are no other sources of sound coming from the software (Windows sounds, IM/AV notifications sounds, etc.), right? So, in theory, if no other application is producing sound the music reproduction through iTunes will be optimal. Is that correct or did I get it wrong?

Differentiation of music playback quality

Reply #15 –
Quote from: Arnold B. Krueger on
(4) Examine the before and after .wav files using an audio editor such as Audacity to ensure consistent levels and timings.  We need 1-2% tolerance on level and 5 milliseconds or better on timing.

(5) Compare using FOBAR2000 and its ABX plug in.


foo_abx takes care of the level matching, actually.  (And ... do mp3 encoders really not manage to get the beginning within one frame tolerance?)

Differentiation of music playback quality

Reply #16 –
Quote from: androdion on
No, I mean an actual audio CD. Original, silver coloured,  factory pressed disc.


If it is one particular CD, I would actually look for this:
Quote from: Porcus on
(And don't get me started at pre-emphasis, which is fortunately not common.)


I don't know if EAC applies de-emphasis when you play. But recent versions scan only TOC, not subcode. AFAIK iTunes scans also the subcode. (Do not ask me how it indicates it, I don't use that software.) CUETools also scans subcode.

Look at http://blowfish.be/eac/Rip/Images/preemph_column.png ; how does that column in EAC look with that particular CD?

Differentiation of music playback quality

Reply #17 –
@ Porcus - "Pre-Emphasis" appears as no. But I'm not sure what that means to be honest.

I've recorded the "what you hear" output of the PC, using Audacity and a segment of a song. For testing purposes I switched off everything in the Xonar Audio Center, and recorded a track with Direct Sound and another with WASAPI selected on iTunes. Then I inverted one of them and exported as WAV and 320k MP3. Both files had no sound, which would mean that having DS or WASAPI selected in iTunes' preferences is pretty much indifferent right? It they null one another then they're the same, right?

Differentiation of music playback quality

Reply #18 –
Quote from: androdion on
It sounds more "natural" and less "boomy" on EAC than it does on iTunes

You don't get to make this claim without objective proof compliant with the rules of this forum which you agreed to follow when registering.

http://www.hydrogenaudio.org/forums/index.php?showtopic=3974
Specifically, #8.

Differentiation of music playback quality

Reply #19 –
I didn't mean to break any rules, I'm not that kind of guy. Nor am I here with the intent of causing any kind of harm to the forum users.

With that being said, all I was trying to do was understand if it was my skewed perception or an actual acoustic difference. You'll notice that I posted about a test I did with recording the output and inverting the waveform of one of the audio tracks to to see if it would reproduce any sound, but if that's outside of the rules please let me know. As I said, I was just looking for answers provided by people who are clearly more knowledgeable about sound than I am, and even more, people who know what they're talking about. That was the whole point of me posting here, having the reassurance that what is being said to me is technically correct.

PS: You'll also notice how every subjective term is within "", which is my way of saying that it's what I perceive and not a fact.

Differentiation of music playback quality

Reply #20 –
Quote from: androdion on
"Pre-Emphasis" appears as no. But I'm not sure what that means to be honest.


If CUETools detects pre-emphasis then you have a case. Although I still put my money on output volume ...

(Pre-emphasis is an EQ applied to the master, meant to be corrected after conversion to analog by the inverse curve. Same principle as the RIAA curve on vinyl. The idea was to raise the treble volume so one would't need so steep filters to get rid of the quantization noise one octave over the hearing threshold. Oversampling eliminated the need for it, so it is fortunately rare nowadays, but back in the past it did appear on releases by names like Black Sabbath, Deep Purple, Miles Davis, Leonard Bernstein, Pink Floyd and Roger Waters (solo). And on lots of classical music, which is often licensed out to budget labels and repackaged, spreading the disease.)