Comments on: Why mp3s suck, and how to hear it

By: Phredreeke

Phredreeke — Sun, 18 Mar 2012 22:31:46 +0000

Well iPod suggests portable listening and most people have lower demands for that than more “serious” forms of listening. It doesn’t have to be “true to the original” just “good enough to listen to”.

As for AAC it stays “good enough to listen to” for very low bitrates.

I think however that Ian’s problem with this isn’t what people listen privately to on their iPods or Sansas or whatever music player you use, it’s your choice what codec to encode with and at which bitrate. The problem is that music are being sold already encoded in MP3 formats which means that someone else already made the choice for you.

By: chris

chris — Thu, 08 Mar 2012 21:25:32 +0000

Your statements conerning quality mean nothing since you haven’t ABX’d FLAC and v0/320. You haven’t even posted what equipment you’re listening on. Sure, you can say, “If it’s going to take up space, I’m going lossless!” if you have the storage capacity to spare (in which case it’s a moot point – just use whatever has the best quality. However, you mentioned that your iPod is 2/3 full using 128kbps, which I find intolerable and leads me to believe that you ARE willing to SEVERELY compromise quality due to limited space), but for most people, v0 mp3 is indistinguishable from the source and takes up significantly less space (3x) then FLAC.

By: Piercejeans

Piercejeans — Wed, 07 Mar 2012 07:07:09 +0000

To add a little education about quality… back in the 1970′s… speaker manufacturers went back and forth on basically this very same issue… should a speaker be built for quantity of sound… volume… and sound that would impress (distort) the true sound… or should a speaker be built for accuracy… quality in parameters of frequency response, total harmonic distortion, phasing shift distortion, time distortion, etc. At the two ends were manufacturers like JBL… the California sound… and IMF… four foot tall studio speakers that Deutchegrammophone uses to mix down classical orchestra music.

Of course… I love JBL… but they did not butcher sound like some manufacturers just going for LOUD… like Klipsch or Bose or Cerwin Vega.

The amplifier / receiver manufacturers took sides. Harmon Kardon and Yamaha were on the accuracy end… reproducing sound frequencies electronically up to 250,000 cps… 250khz… because they found that if they just made stereo systems to reproduce the audible frequencies of the human ear… 20-20,000Hz… that the distortion affected the audible range of the reproduction. That is… the sounds would both be there on both amplifiers… but the Yamaha and Harmon Kardon amplifiers… sounded… live / real / airy.

As far as I remember… the manufacturers that lined up on the two sides of the THD war was roughly Yamaha / Harmon Kardon / Kenwood VS companies like Pioneer / Sony / Sansui… who didn’t give a rat’s ass about high frequencies. They just wanted to shake your windows for the lowest cost possible.

By: Piercejeans

Piercejeans — Wed, 07 Mar 2012 06:52:28 +0000

Funny, through the years listening to music, you always assumed things would continue to get better. They haven’t. Even CD’s CAN sound lousy depending on the recording engineer and what he did with the high end… the treble… the top of the sounds and timbre frequencies that add reality… because it’s what we really hear.

Case in point that came through recently… even though I am a classic rock fan, I bought / listened to two different Frank Sinatra CD’s. One of the CD’s was the Best of the Best… and recorded at Capitol studios. The compressed, chopped high end sound was terrible… even on my thirty dollar Altec speakers on my computer. The other Frank Sinatra CD… Nothing But the Best… by WEA… Warner Electra… had all of the high end harmonics. The difference is startling. Amazon has both albums on sample for you… amaze yourself at how good or bad sound can be if they ditch the high end frequencies… for whatever reason.

By: Marcin

Marcin — Sun, 04 Mar 2012 17:15:42 +0000

It’s all setup thing: on my setup I can easily distinguish 320cbr mp3 and flac.

I’m using flac for computer use and ogg-aotuv vbr ~320 for mobile phone

By: daynurd

daynurd — Tue, 28 Feb 2012 17:40:55 +0000

With my current audio setup, I can’t tell the difference between ~245 V0 LAME MP3 and FLAC using the same song unless I REALLY focus on analyzing specific parts of a song. I keep the FLAC file for archival purposes, LAME MP3 for my Rockboxed Sansa Clip+. I never use iTunes, but do they still use 128kbps? I’ve downloaded some songs off of Amazon and they have 256kbps MP3′s…though they sound significantly worse than if I had ripped the song from a CD and converted to LAME MP3 for some unknown reason.

By: Taariq Hassan

Taariq Hassan — Tue, 28 Feb 2012 07:15:02 +0000

The great unwashed cannot hear the difference between
digital and analog.

By: Ian Shepherd

Ian Shepherd — Sun, 26 Feb 2012 20:56:11 +0000

Hi Matthew,

Thanks for commenting. I’m a mastering engineer with 15 years experience, as well as a blogger – and, I’ve done the blind tests, not that any are needed at 128 Kbps.

You’re right, this post is a rant, and not intended to be scientifically objective – and it says so. But I disagree that both the examples are exaggerated. The Deep Purple clip is a particularly poor example, yes – but the jazz clip is a pretty decent encode, and still comes up lacking, in my opinion.

In this day and age, with superior codecs available, I don’t see any reason to choose mp3.

Ian

By: Matthew

Matthew — Sat, 25 Feb 2012 18:28:52 +0000

Also, Chris’s comments above are very on point. He is someone who gets it. He mentions that he isn’t always sure (could be placebo effect) when he can tell the difference (again double blind tests are necessary because the psychoacoustic effect is so strong). He also talks about the impact of the bit rate.

By: Matthew

Matthew — Sat, 25 Feb 2012 18:21:52 +0000

ABX double blind tests are required for this stuff. Any audiophile or audio engineer worth his or her salt knows that without it, you are likely to be more influenced by psychoacoustic effects and auditory illusions than by actual differences in the music. Unfortunately I’ve become convinced that most do not.

The examples given here are over exaggerated and don’t really prove any point, however, I do appreciate the blogger’s inclusion of some scientific based papers.

End of the day, here is the more nuanced story: for low quality bit rates (128kbps) a lot/most people may be able to tell the degradation of quality. Once you get up around 192 or 256, it starts to become more difficult and may require particular source material in addition to a very good ear to be able to tell the difference. From what I’ve seen from studies, probably around 320kbps is about where the ability to differentiate starts to drop.

Is mp3 worse than AAC? I have no idea. I’d have to do deeper research than what is presented here. I don’t know what the background of the paper presented is. I’ve heard claims counter to the ones presented; that apple doesn’t know what they are doing and the AAC format does some really dumb things. End of the day, the over-exaggerated examples here don’t prove anything.