Anyone seen a proper MP3 review?

[Mark van der Eynden]

Has anyone ever seen a 'proper' review of the MP3 medium?

By 'proper' I mean, say the ripping of a audio 'test' CD and then a
plotting of the Frequency response, distortion and other 'traditional'
quality metrics.

It's one of my pet beefs that MP3 is such a low quality medium that
the record companies should be touting it as such and then it would
die all of its own accord. Litigating against 11 year olds is not the
way to go.

Anyway, I guess I shouldn't make too much noise before I've seen some
figures, kind of surprised there is nothing out there though.

Cheers,

Mark

 

[Tony Pearce]

"Mark van der Eynden" <mvandere@iprimus.com.au> wrote in message
news:48ed3358.0309171631.66d0ec34@posting.google.com...

Has anyone ever seen a 'proper' review of the MP3 medium?

By 'proper' I mean, say the ripping of a audio 'test' CD and then a
plotting of the Frequency response, distortion and other 'traditional'
quality metrics.

Steady state tests have limited value for MP3, but there have been a few
reasonable attempts to quantify the losses produced by various encoders. Do
a search and you will turn up some examples I'm sure.

It's one of my pet beefs that MP3 is such a low quality medium that
the record companies should be touting it as such and then it would
die all of its own accord. Litigating against 11 year olds is not the
way to go.

The quality is dependent on bit rate and encoder. There are not too many
people who can pick (or care) the difference between a good 320 kbs MP3 and
CD using their own sound system. Remember the number of people who actually
care about Hi-Fi is a small percentage of listeners. Most are quite happy
that even 128 kbs MP3 is better than the cassettes they were previously
satisfied with.

Anyway, I guess I shouldn't make too much noise before I've seen some
figures, kind of surprised there is nothing out there though.

There is, try looking harder.

TonyP.

 

[Trevor Wilson]

"Mark van der Eynden" <mvandere@iprimus.com.au> wrote in message
news:48ed3358.0309171631.66d0ec34@posting.google.com...

Has anyone ever seen a 'proper' review of the MP3 medium?

**Not me, but doing such a test would be simple enough.

By 'proper' I mean, say the ripping of a audio 'test' CD and then a
plotting of the Frequency response, distortion and other 'traditional'
quality metrics.

**Easy enough.

It's one of my pet beefs that MP3 is such a low quality medium that
the record companies should be touting it as such and then it would
die all of its own accord. Litigating against 11 year olds is not the
way to go.

Anyway, I guess I shouldn't make too much noise before I've seen some
figures, kind of surprised there is nothing out there though.

**I purchased a CD/MP3 player for my car. Even through the crappy, standard
Commodore speakers, it is possible to hear a significant quality improvement
over FM radio. At 320k/s, _I_ can't hear the difference from a CD.


--
Trevor Wilson
www.rageaudio.com.au

 

[Clifford Heath]

Trevor Wilson wrote:

At 320k/s, _I_ can't hear the difference from a CD.

Ditto. 192k/s seems to be enough for most material, but I'm
really over 128k, that is noticably worse than CD.

The trouble with a scientific test is that though the waveforms
reproduced may differ, the perceptual value is much less
affected. It's very hard to quantify the perceptual differences.

 

[Arpit]

USing an ABX test, I can't tell the difference between the standard
lame preset and a cd.

On 17 Sep 2003 17:31:34 -0700, mvandere@iprimus.com.au (Mark van der
Eynden) wrote:

Has anyone ever seen a 'proper' review of the MP3 medium?

By 'proper' I mean, say the ripping of a audio 'test' CD and then a
plotting of the Frequency response, distortion and other 'traditional'
quality metrics.

It's one of my pet beefs that MP3 is such a low quality medium that
the record companies should be touting it as such and then it would
die all of its own accord. Litigating against 11 year olds is not the
way to go.

Anyway, I guess I shouldn't make too much noise before I've seen some
figures, kind of surprised there is nothing out there though.

Cheers,

Mark

 

[GM]

hi all,
there used to be a site called r3mix that had some very interesting
material relavent to this thread. The encoding method for the LAME encoder
discussed is capable of being controlled by a large number of factors such
as the psycho acoustic model used, the quality of the mp3 (not just the
bitrate), absolute threshold of hearing etc etc. I personally think if you
can hear the difference between a cd and an mp3 which uses 320kbps, either
you are fooling yourself, or you decoder is a POS.


Regards
Giles


"Arpit" <DONTSPAMMEF00Lneko4@dodo.com.au> wrote in message
news:tq4jmvchm2injlos90v9u4pscmrlobeetv@4ax.com...

USing an ABX test, I can't tell the difference between the standard
lame preset and a cd.

On 17 Sep 2003 17:31:34 -0700, mvandere@iprimus.com.au (Mark van der
Eynden) wrote:

Has anyone ever seen a 'proper' review of the MP3 medium?

By 'proper' I mean, say the ripping of a audio 'test' CD and then a
plotting of the Frequency response, distortion and other 'traditional'
quality metrics.

It's one of my pet beefs that MP3 is such a low quality medium that
the record companies should be touting it as such and then it would
die all of its own accord. Litigating against 11 year olds is not the
way to go.

Anyway, I guess I shouldn't make too much noise before I've seen some
figures, kind of surprised there is nothing out there though.

Cheers,

Mark

hi all,
there used to be a site called r3mix that had some very interesting
material relavent to this thread. The encoding method for the LAME encoder
discussed is capable of being controlled by a large number of factors such
as the psycho acoustic model used, the quality of the mp3 (not just the
bitrate), absolute threshold of hearing etc etc. I personally think if you
can hear the difference between a cd and an mp3 which uses 320kbps fixed (or
variable with a 128kbps minimum) , either you are fooling yourself, or you
decoder is a POS.


Regards
Giles

 

[Mark van der Eynden]

Lots of Snips

With all due respects to the posters so far, it would appear you are

all saying (in not these words) that 'I can't hear the difference
between MP3 and CD so there isn't any'.

Now I haven't been in the Audio industry for 30 years now, so I don't
know what the latest trends are, but back in the '70s it worked this
way.

If the distortion was greater than 0.01% the device was crap and
didn't bear listening to, after that the lower the better, although
there was a point of no return after about 0.001%.

It was understood that just because, with a given music source, there
was no audible distortion, this didn't mean that the distorion wasn't
there. That is why technical measurements were made. I have heard
plenty of amplifiers that appeared to be distortion free of certain
types of music, yet on others they were virtually unlistenable.

I would estimate that, depending upon sampling rate and the content of
the medium that MP3 could have distortion rates up to 20%. Most of
this could probably be called 'valve' distortion, by those into 'that'
mumbo jumbo.

Using a high data rate, and a more 'compliant' medium would probably
see distortion fall to about 1-2%. Even this is in the 'garbage bin'
range for anyone above the 'low end' audiophile.

So what gives? Are we moving so quickly in this day and age that we
can't find the time to sit back and seriously listen to the 'music'
that is pumping into our heads? Are our ears so badly damaged that we
all have 'tin' ears these days? Are we so much into 'majority rulez'
that we don't want to hear the minority (i.e. those subtle things that
truly go missing with MP3)?

Any chance of a reality check?

Mark

 

[Phil Allison]

"Mark van der Eynden" <mvandere@iprimus.com.au> wrote in message

Now I haven't been in the Audio industry for 30 years now, so I don't
know what the latest trends are, but back in the '70s it worked this
way.

If the distortion was greater than 0.01% the device was crap

** Well, that requirement knocks out all tape recorders, vinyl, FM radio
and all valve gear and all speakers too.

So what was it you listened to in the '70s ????




............. Phil

 

[Oldus Fartus]

On 19/09/2003 6:33 AM, Mark van der Eynden wrote:

Lots of Snips


With all due respects to the posters so far, it would appear you are
all saying (in not these words) that 'I can't hear the difference
between MP3 and CD so there isn't any'.

Now I haven't been in the Audio industry for 30 years now, so I don't
know what the latest trends are, but back in the '70s it worked this
way.

If the distortion was greater than 0.01% the device was crap and
didn't bear listening to, after that the lower the better, although
there was a point of no return after about 0.001%.

It was understood that just because, with a given music source, there
was no audible distortion, this didn't mean that the distorion wasn't
there. That is why technical measurements were made. I have heard
plenty of amplifiers that appeared to be distortion free of certain
types of music, yet on others they were virtually unlistenable.

I would estimate that, depending upon sampling rate and the content of
the medium that MP3 could have distortion rates up to 20%. Most of
this could probably be called 'valve' distortion, by those into 'that'
mumbo jumbo.

Using a high data rate, and a more 'compliant' medium would probably
see distortion fall to about 1-2%. Even this is in the 'garbage bin'
range for anyone above the 'low end' audiophile.

So what gives? Are we moving so quickly in this day and age that we
can't find the time to sit back and seriously listen to the 'music'
that is pumping into our heads? Are our ears so badly damaged that we
all have 'tin' ears these days? Are we so much into 'majority rulez'
that we don't want to hear the minority (i.e. those subtle things that
truly go missing with MP3)?

Any chance of a reality check?

Mark

I must be misunderstanding previous posters Mark, because I have not
noticed anybody saying there was no difference, only that some people
said they could not hear the difference.

If you look at how MP3 compression is carried out, and the fact it is a
lossy compression system, then obviously the MP3 can not be considered
identical to the original, and does not pretend to be. I assume you
know how it works, if not then
http://entertainment.howstuffworks.com/mp31.htm has a reasonable
description. Definately not hi-fi, but designed so the "average"
listener, if there is such an animal, can not pick the difference.

You seem to have gone off track though. You mention "devices" with
unacceptable distortion, but MP3 is a format, not a device, unless you
are talking about portable MP3 players, or some other MP3 player.
Perhaps you should be comparing the vinyl medium with MP3, or the tape
medium with MP3 or FM radio with MP3 and so on? Perhaps the testing
you would like to see done should use (say) a mid range DVD player with
MP3 playback capability attached to a hi-fi amplifier along with
turntable and tape deck. What I, and others in this thread have said
is that in blind test, using the setup above, many listeners will not be
able to pick the difference between an original cd and an mp3 copy,
providing the mp3 was encoded using a decent bitrate. I believe even
my "old" ears would pick a tape or vinyl or radio with the same source
material.

I am reminded of a discussion I had recently with an avid DVD collector
friend of mine. He was discussing his new disc purchase, and was
raving about the technical quality of the sound track, the clever use of
colour, lighting and the transfer quality. When I asked him what the
story was about, he did not really know - he was too interested in the
technical aspects to have really watched the movie. To each their own
I suppose, but sometimes many of us get so bogged down in technical
details we fail to enjoy the performance.

Please note, I am not suggesting the differences are not there between
MP3 and other formats, and I am not suggesting those differences could
not be measured in some way, only that by design the MP3 format
endeavours to make those differences inaudible to the "average" listener.


--
Cheers
Oldus Fartus

 

[Chris Hoffmann]

In article <48ed3358.0309181433.790d8908@posting.google.com>,
mvandere@iprimus.com.au says...

Lots of Snips

more snips

If the distortion was greater than 0.01% the device was crap and
didn't bear listening to, after that the lower the better, although
there was a point of no return after about 0.001%.

When this stuff last mattered to me (>15 years ago) the weakest link in
any (affordable) audio chain was the speakers.

snip

Using a high data rate, and a more 'compliant' medium would probably
see distortion fall to about 1-2%. Even this is in the 'garbage bin'
range for anyone above the 'low end' audiophile.

Notwithstanding that audiophiles were not really part of the discussion
(as I read it), and that the listeners under discussion would most
likely be using commodity-type, all-in-one, <$1000 sound systems I
would think that just the speakers in these setups would easily be
contributing >2% distortion. Particularly since the price of this type
of equipment is ridiculously low and that the electronics cost can
hardly be shaved further the savings must be coming out of the speaker
quality.

Sounds like a suitable platform for MP3 to be played on in terms of bang
for buck ;-)

So what gives? Are we moving so quickly in this day and age that we
can't find the time to sit back and seriously listen to the 'music'
that is pumping into our heads? Are our ears so badly damaged that we
all have 'tin' ears these days? Are we so much into 'majority rulez'
that we don't want to hear the minority (i.e. those subtle things that
truly go missing with MP3)?

Unfortunately it looks more & more like the answer is yes to most of
your questions, there...

Chris.

 

[Trevor Wilson]

"Mark van der Eynden" <mvandere@iprimus.com.au> wrote in message
news:48ed3358.0309181433.790d8908@posting.google.com...

Lots of Snips

With all due respects to the posters so far, it would appear you are
all saying (in not these words) that 'I can't hear the difference
between MP3 and CD so there isn't any'.

**Well, for me, I did not say that. I qualified my words very carefully.
Through my bog-standard Commodore car speakers, I can't hear the difference
between MP3 and regular CD, at 320k/s bit rate. I went further to state that
MP3 was better than FM radio (in my car). Outside my car and in a decent
home stereo, things may not be the same.

Now I haven't been in the Audio industry for 30 years now, so I don't
know what the latest trends are, but back in the '70s it worked this
way.

If the distortion was greater than 0.01% the device was crap and
didn't bear listening to, after that the lower the better, although
there was a point of no return after about 0.001%.

**Nope. Some, really dumb number chasers felt that way. I was testing lots
of gear on some fine test equipment, back in the 70s. Often-times the test
results bore no relation to the sound quality. I even ended up with a
(highly modified) Dynaco tube preamp in my own system. I'd hate to think how
it measured, but it sure sounded a whole bunch better than most of the other
crap I'd used.

It was understood that just because, with a given music source, there
was no audible distortion, this didn't mean that the distorion wasn't
there. That is why technical measurements were made. I have heard
plenty of amplifiers that appeared to be distortion free of certain
types of music, yet on others they were virtually unlistenable.

I would estimate that, depending upon sampling rate and the content of
the medium that MP3 could have distortion rates up to 20%. Most of
this could probably be called 'valve' distortion, by those into 'that'
mumbo jumbo.

**You'd be wrong. Let Google be your friend and learn about MP3 files.
They're pretty darned clever. THD is just part of the story, but I'll betcha
they measure just fine.

Using a high data rate, and a more 'compliant' medium would probably
see distortion fall to about 1-2%. Even this is in the 'garbage bin'
range for anyone above the 'low end' audiophile.

**Marginal. It depends on the type and spectra of the distortion products.
Don't forget: At reasonable listening levels, most speakers create more
distortion than 1 or 2%. Anyway, I'll betcha MP3 measures better than that,
too. Hell, I'll do some tests this afternoon. I'll get back to you.

So what gives? Are we moving so quickly in this day and age that we
can't find the time to sit back and seriously listen to the 'music'
that is pumping into our heads? Are our ears so badly damaged that we
all have 'tin' ears these days? Are we so much into 'majority rulez'
that we don't want to hear the minority (i.e. those subtle things that
truly go missing with MP3)?

**Let's get real here: For my car, MP3 is just fine and dandy. Even when I
get around to upgrading the Crappydore speakers, I reckon MP3 (at 320k/s)
will be just fine. For critical listening - I don't think so. For wandering
around the house, or at a party -sure, MP3 will work just fine.


--
Trevor Wilson
www.rageaudio.com.au

 

[RMD]

Hello Mark,

Well, OK you have defined yourself as a big swinging dick re sound, as
in nothing is good enough for your picky, picky golden ears. Or your
fat, fat wallet.

Well, gosh, poor little me can tell the difference betwen CD's and
cassettes and LP's and FM and AM radio and all that sort of dreary
stuff.

But I'm obviously too piss-poor to hear if there are any obvious
hearable differences between any of the above sources and the same
source when encoded as an MP3 at a reasonable Kbps level. Because I
can't hear any such difference.

What an absolute crock as in your claim that anything above 0 .001%
distortion is crap, which only eliminates everything.

Ross

On 17 Sep 2003 17:31:34 -0700, mvandere@iprimus.com.au (Mark van der
Eynden) wrote:

Has anyone ever seen a 'proper' review of the MP3 medium?

By 'proper' I mean, say the ripping of a audio 'test' CD and then a
plotting of the Frequency response, distortion and other 'traditional'
quality metrics.

It's one of my pet beefs that MP3 is such a low quality medium that
the record companies should be touting it as such and then it would
die all of its own accord. Litigating against 11 year olds is not the
way to go.

Anyway, I guess I shouldn't make too much noise before I've seen some
figures, kind of surprised there is nothing out there though.

Cheers,

Mark

 

[Albm&ctd]

On 17 Sep 2003 17:31:34 -0700, mvandere@iprimus.com.au (Mark van der
Eynden) wrote:

Has anyone ever seen a 'proper' review of the MP3 medium?

Why MP3? Found our CD collection was OK ripped with cdda2wav and

encoded at Q 6 with ogg vorbis. Any higher quality was just a waste of
disk space for my olde..ish ears. Even found an ogg player for DOS and
added lfn support, sounds quite good and would be good to make an old
computer into a jukebox thingo.

Al

2003 insult page awaits your contribution
http://kwakakid.cjb.net/insult.html

 

[Mark van der Eynden]

ross_dalyNS@yahoo.com (RMD) wrote in message news:<3f6ab1ac.1343634@news.iprimus.com.au>...

Hello Mark,

But I'm obviously too piss-poor to hear if there are any obvious
hearable differences between any of the above sources and the same
source when encoded as an MP3 at a reasonable Kbps level. Because I
can't hear any such difference.

What an absolute crock as in your claim that anything above 0 .001%
distortion is crap, which only eliminates everything.

Many years ago, a man came into the shop to upgrade his speakers. He

spent about 10 times as much as he did on his previous speakers, and
thus went home with a very nice pair of speakers.

He came back the next day complaining that they didn't sound as good
as his old ones. To cut a long story short he was comparing his old
speakers to his new ones and because his new ones sounded 'different'
that made them inferior.

And therein lies the crux of the problem, ears are verrrry subjective,
a double blind test will only enable you to identify a 'true'
difference, not which is 'better'. This is the whole purpose of test
gear, to remove the subjectiveness.

I should have qualified my 0.001% as purely pertaining to the
amplifier. Vinyl records don't cut it at all, Speakers aren't that
great either. The idea of having the minimum distortion at every point
is that it 'multiples'.

Anyone ever noticed that listening to crappy quality music for
extended time periods gives you a headache, whereas good quality music
doesn't? I'm sure ther red wine people will be onto me for that, but
anyway, it shows that you really can hear the difference (well, at
least measure it by the size of your headache).

Mark