How well can you hear audio quality?

[IHateRap]

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


The files get switched so you can do the test multiple times without knowing which is which. So I did it twice. Two different pairs of headphones plugged directly into my Mac Tower.....I didn't go thru my good audio interface. The first was with a pair of Shure SRH840 headphones. I missed all of them although on 3 of the 6 I picked the 320. I used Shure SRH1840 headphones for the second try and got 4 out of 6 correct. 128 and 320 for the wrongs. Clarity and realism were more evident with the better headphones. For the record I didn't think any of them sounded bad.

 

[MrDowntown]

I tried it once with earbuds that one gets with an iPad/iPhone. I got 320 on most, and got the "higher quality" only twice. Had to turn it up way too loud, for what I'd consider safe and smart for ear bud use, just to discern what my opinion of higher quality is. Like you, none sounded terrible.

So For me...no...it wouldn't be worth the extra scratch for the uncompressed stuff....JMO.

ETA. I had to listen to each sample track multiple times to pick apart what I considered to be quAlity nuances.

 

[scottosan]

It's hard to tell on most studio produced recordings since the tracks are isolated. It's much easier with audiophile type recording in which the entire room is mic'd and there is an actual sound stage and depth.

 

[rcm78]

I've always found rock/metal or anything with live drums and cymbals very susceptible to low quality audio. Dense mixes are easy to tell. Spoken word is very difficult.

I was able to select the 320 or uncompressed files for all of them. But, the Jay-Z and Suzanne Vega clips were extremely difficult to tell the difference. I had to turn it up pretty loud and listen to each clip several times and even then I still think I got lucky with the Suzanne Vega clip...

I listened on AKG K240 headphones...

 

[moltenmetalburn]

High frequency alias is usually pretty easy to hear in any situation, ill check this out.

 

[Inearthed]

Fun test. Is it me or do the drums in general in the Coldplay song just sound terrible? Don't know if it is the performance or just the tone of the drums but, ugh...

 

[JamesPeters]

I only tested these at the volume I feel comfortable listening to music for any significant length of time (moderate to loud 2-person conversation volume). I see no point in testing any louder since I can't hear fine (quieter) details compared to loud sounds smashing my ear drums after a relatively short period of time (and I don't like hearing loss either). I was using Sennheiser HD280 Pro, with damping material added on the inside lining the speaker enclosure behind the speakers, to reduce the emphases to a reasonably "closer to flat" response (they now have a slight thickening of mids and lows but that's it), and I used my desktop PC with a good audio device in a quiet room.

I got three of the results "right", guessing which were uncompressed WAV. One I guessed the 320 Kbps MP3 was best, and I was on the fence for that one. The other two, I guessed the 128 Kpbs MP3 was the best file.

I knew right away for one of the songs I wasn't going to be able to guess with any degree of certainty, since some types of content (less dynamic, less broad frequency range, reduced stereo image) compress with less artifacts. That one was totally up in the air, even when I increased the volume a bit. I was straining to notice a difference and although I might have noticed a difference, none of the files sounded better to me.

For another song, I was paying attention to one aspect of the mix that I thought sounded most interesting to me (which was emphasized therefore "sounded more hifi" to me) and not constantly going back and forth between files to hear if I had "missed something"...but then again that's normal since that's not what you do when listening to music anyway.

And for one song I knew immediately which file was which (all three samples of the song), with a high degree of certainty.

For a couple others, I had to go back and forth a few times and my confidence level was maybe 70% that I was guessing right.

This helped illustrate that sometimes I can hear a difference depending on the content, if I'm paying attention and listening under optimal conditions. But beyond that, it's not something I can say with any level of certainty. I don't have a high-end hifi system, but my system is overall quite good and noticeably better than using earbuds with my phone (which I do often enough, or using the HD 280 with my phone which is still noticeably lower in quality than with my desktop PC's audio device). If I were a hifi sort of person who owned a multi-thousand-dollar hifi system, I'd probably pay for the uncompressed WAV files just to be sure that for the ~50% of the time I might notice that it's better than 320 Kbps (going out on a limb with that number), that I'd be getting "the best".

 

[8len8]

I only got 3/6 when playing the samples through studio reference monitors. Guess that mans my hearing is bad .

One thing people need to remember is that different songs respond to mp3 compression differently. This test probably didn't g out of its way to use "worst-case" songs who have a mix that isn't compassion friendly...

 

[SpiderWars]

I used my home stereo speakers and got 5/6. I got confident and rushed through the Jay Z one so I might have gotten it had I taken the time. Once you find a part with lots of high end detail, just go back and forth between each one and it will become evident. You can just click anywhere on any bar at any time and it will start there.

 

[milkchickenbomb]

Weirdly I picked the 320 6/6, and the uncompressed wav sounded the worst to my ear each time.