NPR Quiz – can you hear the difference?

[mikehalloran]

I don't see this posted yet.

A new quiz from NPR challenges you to try and tell the difference between uncompressed music and compressed versions of the same samples. The 128kbps-quality clips may have a noticeable if subtle dip in quality, but the difference between 320kbps and uncompressed WAV files is likely to be lost on all but a handful of “As a music producer listening through my standalone DAC and HD Ultrasone headphones, I am appalled at the garbage I’m hearing, etc., etc.” types. Go ahead and take the challenge, or better yet, make your audio snob friend take it in front of you and marvel at the sputtered excuses he comes up with for not getting 100 percent.

http://www.npr.org/sections/therecord/2 ... io-quality

Whole article here:

http://www.avclub.com/article/quiz-call ... ess-220355

I was surprised at my results. I figured that my 61 year old ears wouldn't do so well. Through my Behringer MiniMON and Equator D5s, the differences between 125k and 320k were easy to hear. As for the rest, it took quite a few listens with the volume cranked before I'd say, "It is probably this one...". The top end was just that much nicer to me but I was never positive except with the chamber music piece.

I had to come to the conclusion that 320k was nicer than I thought. Now that I know, I will see if I can hear it any better through my Sony 'phones I have multiple pairs of the V6 and 7506.

[Phil O]

Hey, good post! I missed the Jay Z one. Got the others. The differences are subtle on some, more obvious on others.

Phil

[mikehalloran]

Hey, good post! I missed the Jay Z one. Got the others. The differences are subtle on some, more obvious on others.

Phil

That one was a blind guess for me. It did reinforce the need for a better subwoofer – big time.

[Robert Randolph]

Nailed Wave 6/6, then tried to pick out 128. Got 6/6 again. Tried to pick 320 and went 4/6.

There are some subtle clues if you are listening for encoding artifacts. If someone set them in front of me and said 'Which is better', I doubt I'd score more than 50% between 320/wav.

Some of the music samples though.. ugh. I couldn't listen to more than 5 seconds of the neil young. That was enough thankfully.

[monkey man]

Many years ago, once I'd made the decision to reconstruct my iTunes lib from scratch, I had to make a call between wav, flac and 320k MP3. I chose 320k, not just 'cause the file sizes were on average 4 or 5 times smaller, but because I figured that in a living-room context where one's simply listening to one's hi-fi there'd be little or no discernible difference.

Mike, thank you for posting this. Makes me feel all warm and fuzzy.

Mind you, the inevitable audiophile geek invasion ensued, replete with its compulsory technobabble. Interesting quote from the comments:

The audiophiles I know are no doubt better educated, more successful, more powerful, more important and certainly not as rude and prone to stereotyping others as you clearly are.

What a classic, IMHO... and on so many levels.

[bayswater]

Isn't that second link from The Onion? Better educated audiophiles? They left the planet with James Taylor's "circles around the sun"

[Michael Canavan]

Some of the music samples though.. ugh. I couldn't listen to more than 5 seconds of the neil young. That was enough thankfully.

With Katy Perry and Coldplay in the mix Neil Young stood out as the worst???
Wow?

[Shooshie]

I picked the Wav 4 times, and the 320 twice. On three of the Wavs that I picked, I was certain. On the other one I was not really sure, but just "felt" something different. Couldn't put my finger on it.

On the two in which I picked the 320, I really couldn't tell. I had narrowed it down correctly to one of the two better ones, but I honestly could not tell which one was the Wav.

Here's how it broke down:

Picked the WAV with Certainty:

• Murray Perahia
  Suzanne Vega
  Coldplay

Correctly picked WAV, but wasn't sure:

• Neil Young

Picked the 320kbps:

• Katy Perry
  Jay Z

I think I could have picked any of the three for Katy Perry. It's so compressed in the mastering that I feel out of my element. I just don't listen to music like that. I also don't listen to rap, though I was pretty sure which one was the 128kbps. It could be that we are accustomed to certain cues that give it away, so in music where those cues don't exist, we don't really hear the difference.

By the way, Murray Perahia's Mozart was one of the worst-mastered classical recordings I've heard in a long time, barring those of the pre-LP era. In fact, a lot of the pre-LP recordings are better than THAT! But when I read that it was recorded in 1981, I gave them a pass, because very few people knew how to master digital audio at that time. Not only that, but the piano had a clinker on what sounded like an Eb. Columbia put out a lot of crap back in the 1980s.

Interesting test. I use 320kbps for all my AACs, though when I'm using LAME for MP3, I use VBR (Variable Bit Rate) and the highest setting. I guess i could use the highest setting without VBR these days, since we're really no longer concerned with file size. In taking the test, I had to listen to Katy Perry and JayZ over and over and over and over. Gads! It occurred to me that unless Katy Perry is mixing her own music, creating all the effects, and mastering it, she's not really the star of her own songs. In fact, I think the same could be said for most of those.

Regarding MP3 quality: When I use 320kb in iTunes, I really can't tell most of them apart from the original unless I really study it. They're awfully good. When I hear someone go on and on about MP3 sucking so bad that it's destroying the fabric of society, I just roll my eyes. It hasn't been like that since the 1990s. Then again, my ears haven't been like that since the 1990s, either!

Shooshie

[Robert Randolph]

Some of the music samples though.. ugh. I couldn't listen to more than 5 seconds of the neil young. That was enough thankfully.

With Katy Perry and Coldplay in the mix Neil Young stood out as the worst???
Wow?

You probably hear a song or an artist that you are familiar with and a song you enjoy.

I just hear tape noise, distortion (vocals are quite distorted) and off-pitch vocals. It sounds very unpleasant to me when trying to evaluate the quality of the playback media.

I shared this last week with a bunch of friends and nearly everyone had similar comments about the Neil Young sample. One person wouldn't even finish the test because they hated it and the Jay-Z sample.

That's not to say that all of the samples were fantastic recordings, but the worst sounding to me was the Neil Young. Quite frankly, I didn't like most of the samples. They all were fairly poor examples for their genre except for Suzzane Vega.

[Michael Canavan]

Not at all my favorite Neil Young track and it was a strange choice to use that song since it always sounded to me intentionally lo fi.

As far as out of pitch vocals, yes please, so completely tired of everything perfect and sanitized.


I have no dog in this race, I couldn't care if I wanted to about tape hiss etc. So many classic songs were recorded badly or with inferior equipment compared to today, and conversely so much of what get recorded these days is clean but crap.

[Robert Randolph]

Not at all my favorite Neil Young track and it was a strange choice to use that song since it always sounded to me intentionally lo fi.

As far as out of pitch vocals, yes please, so completely tired of everything perfect and sanitized.

I have no dog in this race, I couldn't care if I wanted to about tape hiss etc. So many classic songs were recorded badly or with inferior equipment compared to today, and conversely so much of what get recorded these days is clean but crap.

Well, as I said, it was the intention of the quiz to evaluate audio quality. I think we can at least agree that it was a rather poor choice for that goal.

In the context of the quiz I feel like it's dishonest for the article to include a lo-fi song when you're being asked to differentiate between high-quality and low-quality encoding. Somewhat ironically though, that's the easiest song for me to pick out.

[Michael Canavan]

In the context of the quiz I feel like it's dishonest for the article to include a lo-fi song when you're being asked to differentiate between high-quality and low-quality encoding. Somewhat ironically though, that's the easiest song for me to pick out.

Well as you noticed intentional lo fi VS aliasing artifacts are pretty obvious. also there's a pretty much single note high end string section on that track which is pretty much the perfect thing besides cymbals for hearing aliasing. <-- Notice that both those instruments are pretty dammed noisy to begin with, well far away from sine waves.

I sincerely can't do these tests though because I would take a badly recorded Neil Young track over Coldplay any day.

[Shooshie]

Some of the music samples though.. ugh. I couldn't listen to more than 5 seconds of the neil young. That was enough thankfully.

With Katy Perry and Coldplay in the mix Neil Young stood out as the worst???
Wow?

You probably hear a song or an artist that you are familiar with and a song you enjoy.

I just hear tape noise, distortion (vocals are quite distorted) and off-pitch vocals. It sounds very unpleasant to me when trying to evaluate the quality of the playback media.

I shared this last week with a bunch of friends and nearly everyone had similar comments about the Neil Young sample. One person wouldn't even finish the test because they hated it and the Jay-Z sample.

That's not to say that all of the samples were fantastic recordings, but the worst sounding to me was the Neil Young. Quite frankly, I didn't like most of the samples. They all were fairly poor examples for their genre except for Suzzane Vega.

The Neil Young sample was truly awful, and awful in many ways. He sounded like he was phoning it in, the arrangements sounded totally disconnected from the actual song or Neil Young, or the lyrics; and the mixing sounded like it was done by two people in two places and times, with two different sets of tracks that they just went ahead and combined while mastering them. It was a little like Neil got dropped accidentally on the David Foster assembly line.

But several of them sucked almost as bad for various reasons. Katy Perry may have been exquisite for Katy Perry; I just have no point of reference. It was not something I would ever hope to encounter again. JayZ's was also outside my realm, and likewise, I was thankful I won't have to listen to it ever again.

Coldplay's sample sounded pretty good, actuallyv. And i loved Suzanne Vega's little "found" poem. Not because of the words or the music, but the way they came together in Suzanne's voice. Little mixing to discuss there; it just worked.

The thing that bothered me the most about the four heavily-produced tracks was the way that the arrangers seem to have not even the slightest connection with the artists. I know it doesn't matter to anyone what I think, but before you even get to mixing and mastering, the song has to be self-consistent. That arrangement needs to fit the artist, the mood, the lyrics, and so forth. In that sense, Katy Perry may have be one of the better ones, if her intent is to be a prop in her own song while her arranger goes nuts with mix tricks.

[Michael Canavan]

Harvest is a great album, maybe you guys can judge a song by 30 seconds or less, but the piece fits in with the rest of the record. In my opinion of course, Young has a good amount of great songs, but the only two records that are great are After the Gold Rush and Harvest.

This pretty much cements why to me the whole jumping after hi fi is a pipe dream, and yeah I do think it's ironic that old man Young is pushing ridiculously hi fi portable music players, and we're discussing the recoding quality of one of his most critically acclaimed records as bad.

There is something sort of depressing about the fact that you can get so wrapped up in the technicality of the recording or song that it can ruin something. I remember hearing the first Iggy and the Stooges record and thinking it was cro-mag but genius, now when I hear I Wanna Be Your Dog all I can hear is the annoying one note piano line I never paid attention to before.

[bayswater]

This pretty much cements why to me the whole jumping after hi fi is a pipe dream, and yeah I do think it's ironic that old man Young is pushing ridiculously hi fi portable music players, and we're discussing the recoding quality of one of his most critically acclaimed records as bad.

Maybe that's why.