Quality of modern recordings

[Marvel]

Well said, Mike! I couldn't agree more, and it's the music that is important, and the emotional impact ot brings to the table.

Bruce

[DrWho]

They are listening to the music. And I also submit that they are so "in tune" with the music that their brain actually fills in all the details that "audiophiles" hold in such high regard.

Is that even a problem?

So, you think all this "elite audiophile stuff is what's
killing this hobby for the younger generations", do you? I've got news
for you Mike, it's always been that way. Nothing has changed. And
nothing has been killed off. It was this way 20 years ago, 30 years ago,
50 years ago, and even in Edison's day. There were those who pushed the
limits of what was possible, and the rest mostly never even knew about
it.

Well I can't exactly take credit for the thought. It
comes from working with youth groups for the last couple years and
actually interacting with that generation. They're the ones telling me
not to be so elitist, and they're the ones rolling their eyes when
reading websites/forums/magazines on "high end" audio.

I'm also pretty sure that if we look at the numbers, high end
speaker sales have been steadily declining. At least I certainly don't
believe the khorn is seeing record sales in the last few years compared
to the PWK days. Perhaps I'm mistaken?

Personally, I think just because I said MP3
sound like crap, you've somehow construed or confused my interpretation
of the issue of accuracy of sound reproduction with the enjoyment of
music.

It's that whole "accuracy" that I'm hung up on
because I'm pretty sure you would defend that a stereo pair mic array in
an ambient hall played back on your system would be an example in
"accuracy". I would contest that it is not always the case, and might
even go far as to claim that an mp3 played back through a bose
acoustimass system could in fact be more accurate. []

It all
comes down to the intent of the artist, and I see no reason to limit
"musical accuracy" to approaches that documentaries and scientists would
be more apt to use.

And if you'd like, I would be more than
willing to explain the basic physical fallacy behind stereo mic'ing and
2-channel (or 3-channel) reproduction. It doesn't take much to realize
that it's physically impossible to recreate the actual live event. The
geometry simply cannot work. Is it wrong to say that you like those
recording techniques because they are sufficient to fool you into
experiencing the actual live thing (while your brain fills in the
details)? I know I was faked out major style at your place however many
years ago that was. Your system is very impressive in that regard and
continues to be a benchmark for me when it comes to those very specific
genres of music.

My question (and that of my younger peers) is how come what fools
your (my) perception must somehow be the criteria for someone else that
is seeking after a different kind of accuracy? Why are they suddenly
ignorant because their minds can fill in different kinds of gaps that
you (I) can't? Or what happens when they don't get faked out by the same
thing that you (I) are able to fill in the gaps on?

And then why
do we fault music that captures the emotions in different ways? A lot of
modern music isn't trying to be a "documentary of a live event", but
rather trying to paint a new soundscape and offer sounds that could
otherwise never be experienced. It has often been the case in my
experience that those kinds of genres require very different playback
systems in order to maintain "accuracy". When played back on a system
good for stereo-mic'ed classical, these genres are gonna sound horrible.
Of course the knee-jerk reaction is to assume accuracy of the playback
system while bemoaning the crap of modern recordings. On the flip side
though, there is no shortage of measurements and good science to show
that different approaches would in fact be more accurate to the intent
of the artist. And when those approaches are carried out, the effect is
just as powerful at transforming you into another place.

I'm reminded of this one art exhibit that involves some vivid colors in what must otherwise be a pitch black room
in order to capture the full essence. However, when viewing classical
paintings it's supposed to be the holy grail to have the most natural as possible lighting conditions. However, if that were
to be applied to the light color exhibit thing in the dark, then the entire
artistic effect would be completely ruined. The same paradigm happens with our playback systems.

[DrWho]

I can't explain this!

I think I'd be perfectly comfortable referring to it as nostalgia...

[tube fanatic]

I can't explain this!

I think I'd be perfectly comfortable referring to it as nostalgia...

Yes, indeed Mike! And as far as "accurate" reproduction goes, Langford-Smith said this decades ago, and it's still quite valid imo: "If we apply tone correction in the form of bass and treble boost, it is difficult to gauge the correct amount, and most correcting circuits only give a rough approximaton of the ideal. If we leave the adjustment of tone controls to the listener, he will adjust them to please himself and the result is usually far far from a true reproduction of the original." It doesn't matter if one is using the controls on a vintage tube amp, or a modern digital room correction device. One is still going to do what sounds "best" which is rarely the same for 2 people. My daughter grew up with state of the art sound and yet is perfectly content to listen to her music (mp3s) on her portable devices with crappy buds. Her husband, a video nut, set up a system using my AR-2s and Large Advents and she could care less about listening to music through them. While she says that it sounds very impressive she, and all of her friends, couldn't even conceive of spending money on an excellent system.

[Mrtie]

I also think mp3's get demonized a bit more so then they deserve, the compression was painfully obvious in the early days when 192kb/s was the norm, but in all honesty I get a huge kick out of threads on places like head-fi where they ABX test 320kb/s versus WAV/FLAC, very very very few people can consistently tell the difference and this is mostly through mid to very high end headphone gear, I wouldn't stand a snowballs change in hell ABX on my jubs.I still back up in flac becuase hey space is cheap, but I'm sure I would never be able to tell.

[Chris A]

...very very very few people can consistently tell the difference...

I've found that the ability to hear a difference is strongly correlated to WHAT you listen to. If your listening palette is restricted mostly to pop-rock recordings made decades ago as master tapes for LPs (or even the LPs themselves), and recently recorded-but-compressed recordings, i.e., "loudness war" recordings, then I'd agree that it is very difficult to hear any difference.

However, if you put on a more revealing recordings (BTW, the type that I listen to most of the time) then it isn't very difficult to hear the difference: flattened soundstage, strange changes in air and feeling of room/environment, some loss of detail, and strange sounding musical attacks and sustains are very audible. Many of these revealing recordings are classical and jazz recordings but also some "new age-ish" recordings.

The most revealing recordings that I can think of are recently recorded solo violin, string ensembles, percussion ensembles, or pianoforte. Even the highest bit rate mp3's sound odd or artificial under these circumstances on the main rig.

Moreover, CD recordings from the late 80s and early 90s can typically be distinguished from later recordings of higher quality. Because of this, I'm slowly updating my music library with more recent recordings wherever possible. I've found that most recently recorded Linn recordings are outstanding (digital disks). Remastered recordings from the 80s or before are almost always audible--and in odd ways.

One test that I cannot tell the difference: two channel SACD or PCM (i.e., red book "CD") recordings on the same disk in "A-B" mode. However, multichannel SACDs are usually much more realistic than their embedded two-channel CD tracks off the same hybrid disk.

Chris

[annt]

Back in 2005, while I was in school for a project we performed double blind listening comparions between Lossy (ABX l.a.m.e. encoded mp3) and wav files (perfect bit for bit cd rips via eac). I dont remember exactly how many people we tested but it was around 20 people who were all required to listen to both versions of the same 30 second snippet of a song randomly. Each person had about 15 songs to go through and they did this twice. They then were asked if there was any difference and if so, which sounded better. Music was played via laptop to a fairly highend USB Dac that fed a headroom headphone amp and nice set of higher end Sennheiser headphones. People ranged from your average ipod listener, sound engineer students and self proclaimed audiophiles. In the end the results showed that noone could consistently tell the files apart.

One thing we did realize by accident while preparing the tests was that any difference in volume levels, however small, we tended to favor the slightly louder version as sounding better. Another thing was the drastic difference in mp3 encoders. At the time, the LAME encoder was (and still is I think) considered the best mp3 encoder, but we tried others and there was definitely quality differences at the same bitrate.

Anyhow, I think lossy music files gained a terrible reputation in the early days because most cd rippers were terrible, encoders sucked and your average computer had crappy sound cards.

[garyrc]

no one could consistently tell the files apart.

Even though no one could consistently tell the files apart, could some (few?) inconsitently discriminate between them at a statistically significant level (say, at or beyond the .05 level)? Only partly kidding.

AES came to a conclusion similar to yours about SACD v.s. Redbook CD. I'm still not convinced. I'd like to see a similar double blind experiment done using ONLY files considered to be excellent recordings. So, using hybrids, CD layer v.s. SACD layer (same mix, put through two channels only), and MP3s v.s. bit for bit. With an N big enough to provide sufficient power, the experimenter could buld an attribute variable into the design, such as "type of listener," i.e., " highly experienced" (self proclaimed audiophiles, musician, sound engineers) v.s. "people gathered randomly" from campus with any happening to fit the 'experienced" category reassigned to the other group, etc., then look for interactions as well as main effects.

[Quiet_Hollow]

A lot of
modern music isn't trying to be a "documentary of a live event", but
rather trying to paint a new soundscape and offer sounds that could
otherwise never be experienced.

This is an excellent point with roots spanning as far back as the 50's[][Y]

I've found that the ability to hear a difference is strongly correlated to WHAT you listen to.

Bingo...and I think this is where the crux of differential testing lies. Trying to A-B something when the item is foreign is an exceptionally difficult task, for anyone, regardless of what's under scrutiny.

Are the songs (or sounds) used when comparing source file types familiar to the listeners? Are the appointed critics keenly aware of the audible character of the playback system / device? If not, the odds of them picking out differences, with any degree of confidence, are going to be slim at best. I'd wager anyone's first thought would be, "What am I supposed to be listening for?"

[Chris A]

Back in 2005, while I was in school for a project we performed double blind listening comparisons between lossy (ABX l.a.m.e. encoded MP3) and WAV files (perfect bit-for-bit CD rips via EAC). I don't remember exactly how many people we tested but it was around 20 people who were all required to listen to both versions of the same 30 second snippet of a song randomly. Each person had about 15 songs to go through and they did this twice. They then were asked if there was any difference and if so, which sounded better. Music was played via laptop to a fairly high-end USB DAC that fed a headroom headphone amp and nice set of higher end Sennheiser headphones

My tests included ripping using iTunes into an an iPod Touch (latest generation), then out using the headphone analog port. into my preamp without further processing. So anything inside of that "black box" is what I'm referring to.

Additionally, my tests included MP3s bought directly from Amazon of a solo pianist playing the Goldberg Variations (BWV 988) out on a HDMI from a computer to the preamp's DACs. The comparison setup was direct from the CD and converted to analog via the DAC in an Onkyo preamp (Burr-Brown 192/24) as another "black box".

While the MP3s worked nicely on the iPod Touch using reasonably good headphones, they don't sound very good on the main rig.

I suspect what we are talking about is quality control and implementation, but my question would be: why alter the original recording at all to put it into MP3 format? The answer to this is the crux of the argument. Backing store (i.e., hard disk) storage is dirt cheap, so the issue is portability on a very small, low power hand-held device. If you are listening using headphones, then I see no issues. However, if you at home using a computer feeding a good hi-fi, then the need for MP3 seems to disappear.

For me, the bottom line is that most people are listening to not-very-good recordings (bought or pirated MP3s in the wild) on variable quality earbuds from very small and portable devices (NOT relatively cheap, however). That should be the end of the argument, however, since when you try to hook that small device to a relatively good multi-channel reproduction system in-room, the sound that emerges makes one want to turn it off immediately. [+o(]

One more point: why buy a system that intentially and involuntarily alters anything from the original signal at all? When you DO want to hear the performance as close a possible to the original (assuming that is ever a priority for you...) and if you've already made compromises in the quality or fidelity of your playback equipment, you've lost the ability to faithfully reproduce the information on the original media. To me, this seems an obvious reality, but apparently this is not what is being argued: rather I see discussions involving economics and small-size/looks being argued instead of hi-fi. Economics are important, but as a tradeoff decision variable, not as an objective, IMHO.

Just my $0.02

Chris

[Mrtie]

As I stated in my previous post I agree there is no reason not to store in lossless given how cheaply storage is for a home media solution, and I do. I do however believe the differences in high bitrate well encoded mp3 are not as large an elephant that some make it out to be, but again thats my ears, my music, my system, this obviously is not true for everyone.If I really wanted to open a can of wormsI also think my digital front end sounds better then my vinyl system, but I still enjoy both for what they do.

[fenderbender]

[Jim:] "I don't care

Shut up

Play the record."

[Chris A]

As I stated in my previous post I agree there is no reason not to store in lossless given how cheaply storage is for a home media solution, and I do.

I reinterated your point above. I responded to "annt" so that the test conditions were available for my comments. No comments were directed your way in my response.

I do however believe the differences in high bitrate well encoded mp3 are not as large an elephant that some make it out to be

My point is that this isn't really an argument about the best of MP3 vs. the best of WAV format, but rather an argument trying to compare typical vs. typical cases: the issues seem to be in the devices used and in the exact type of formatting used by commercial vendors selling MP3-formatted music files. This is where the loss of fidelity seems to appear, not the "best case" data formatting scheme. I don't see the typical or most likely situation in the arguments using MP3s.

I don't know many hi-fi enthusiasts that sink thousands of dollars into their main rigs, then turn around and use an iPod as their main input device, however, you might. I also don't know many hi-fi enthusiasts that use their ripped iTunes files in MP3 or other lossy format from their computers to drive their systems. They instead tend to use lossless formats that do not damage the original data format stream after reconstruction (de-packing) of the files at playback time.

Chris

[Mrtie]

I agree Chris, honestly I do, my idea isn't even really for you, if anything I just get a bit of kick out of friends who praise the significant sound difference in lossless and listen to it on cheap hardware, in the end I think it's mostly in their heads, being a younger guy most of friends are not listening to high quality recording to begin with anyway. Either way in the end I love my share of well and poor recorded music, and it's all about enjoying the experience, something we are all blessed with in this hobby no matter our budgets.

[garyrc]

While the MP3s worked nicely on the iPod Touch using reasonably good headphones, they don't sound very good on the main rig.

That's exactly what happened repeatedly at a local medium and high-ish end store. People found that MP3s didn't sound very good on the store's good equipment, compared to how they sounded on the iPod. A CD or, sometimes, a SACD from the same original recording sounded better.. Sometimes the good system revealed more warts (distortion of various named and un-named kinds).

For me, the bottom line is that most people are listening to not-very-good recordings (bought or pirated MP3s in the wild) on variable quality earbuds from very small and portable devices (NOT relatively cheap, however). That should be the end of the argument, however, since when you try to hook that small device to a relatively good multi-channel reproduction system in-room, the sound that emerges makes one want to turn it off immediately.

Exactly!

[Bubo]

My bad, reposted below for readability using IE, not chrome

[Bubo]

My bad, I keep forgetting that Chrome posts get packed making reading more difficult, reposted using Explorer

Found the information that is used to design compression algorithms (mp3 and others) and music designed for MP3 players.

http://en.wikipedia.org/wiki/Psychoacoustics

http://en.wikipedia.org/wiki/Cognitive_neuroscience_of_music

My summary, all of the tones, buzzing and hissing that are common in the top 40 are put there for a reason. To trick the Brain into either ignoring sonic components or getting the brain to estimate what is missing and create it for you.

There is no free lunch.

For the record, I find the noise generated for the top 40 to be irritating in its own right and probably masks some really mediocre performers that the studios are pushing.

[sunnysal]

Reviewing mentally as much time and resources as I have put into trying to improve the sound quality in my systems over the years I must admit I am pleased (yes pleased) that so many people are ignoring sound quality as an important factor in music playback and seem ignorant or uninterested in listening to the differences in the same music played back via a hi rez source and high quality sound system. why? simply this, it means to me that people are probably enjoying the music not the sound, as simple as that. I remember well, in the late 60's and early 70's when my first "stereo" was a portable mono all-in-one record player and later a "boom box" and most of my listening was to, LPs, FM radio and cassettes. was the audio quality of those sources and system hi fidelity? nope. did I enjoy the music? OMG yes! I remember laying there at night listening to a Santana concert broadcast over FM radio and being completely engrossed by the music, I can still remember it. As I got older and grew in my financial capacity I acquired equipment that allowed me to reproduce music with much more accuracy but honestly it never really increased my enjoyment of the music itself to any significant degree. Can I listen louder? clearer? less hiss? no pops and clicks? yes, but when James Brown forces me out of my chair to dance around the living room it is not the sound quality doing it, it is the music speaking to me and frankly when it speaks to me, IF it speaks to me, it has little to do with accurate reproduction. I am not making a religious statement here so no flames please, just pointing out that for me the balance of importance between sound quality and the musical content lies far into the music content side, with the emotional reaction, always has in my case and thus I find myself obsessing far less about my equipment. do I purposely look for low quality source and play back equipment to listen to my music? nope. But MP3s seem to do the job just as FM radio and cassettes used to. This probably explains why first my turntable was retired, then my cassette player and now my tuner and CD players. I am quite happy listening to music on my squeezebox fed by my ripped CD library and internet radio stations. I am perhaps just missing a few LPs I have not ripped to digital format but that will com eventually/ YMMV, IMHO, etc., etc. warm regards from sunny El Salvador, Tony