Telling the difference between mp3 and flac quality

[melancholera]

http://www.npr.org/sections/therecord/2015/06/02/411473508/how-well-can-you-hear-audio-quality

 

A test trying to determine if people can really tell the difference between low quality mp3 (128kbps ), high quality mp3 (320 kbps) and lossless audio (flac).

 

I got 2/6 flac but only chose the lowest quality once. This was through my logitech 2.1 computer setup. Not the best listening station in my house.

[nice]

Yeah I got all of them wrong except the first one. 128kbs vs 320 is noticeable, 320 vs uncompressed is almost identical to my ears. I listened with my Logitech G430.

[WeAreOceans]

Doesn't this depend on your setup? You're probably not going to hear much of a difference with an on-board sound chip.

[xfxf]

onboard sound is pretty decent these days tbh. that's heavily a throwback to a decade plus ago when onboard sound was genuinely awful

[WeAreOceans]

onboard sound is pretty decent these days tbh. that's heavily a throwback to a decade plus ago when onboard sound was genuinely awful

 

yeah I guess so, just got 5/6 on my work Thinkpad laptop and Koss Porta Pros without an amp/DAC

 

not that I'm much of a FLAC fanatic, my digital collection is almost entirely 320 MP3s.

[xfxf]

ya; I'm currently searching for a good USB DAC/amp for my laptop at work. The onboard sound on my macbook pro is better than one I just bought haha :(

[joshuatxuk]

Yeah I got all of them wrong except the first one. 128kbs vs 320 is noticeable, 320 vs uncompressed is almost identical to my ears. I listened with my Logitech G430.

 

I haven't tried this yet but I've always been able to distinguish 128 vs 320 fairly easily.

[eugene]

on sennheiser hd650 connected to asus xonar stx soundcard i picked 128kbps on the classical piece, but the rest of the choices were either 320 or wav.

i've already did abx tests for myself comparing flac and 320 and couldn't differentiate between them so it's not news to me. still i download solely flac though, because who knows, maybe with some tracks under certain conditions it will make a difference to me so i don't see the point in risking it by downloading mp3.

 

gonna try it with speakers tomorrow.

[Bechuga]

1st time round, got 2 right.

2nd time (the positions change for the answers, so it's just as hard) I got four. Although I had my ear pretty much right next to my speakers and eyes closed, which is perhaps the point: unless you're straining pretty hard next to good speakers or wearing solid headphones, you're not gonna notice it.

 

That said, like Eugene, I'm still gonna do flac anyway, especially if I'm at home (flac is more bassy imo, could just be psychological tho). At work, in a louder environment, I couldn't tell an wav from a ogg files.

[chenGOD]

1 out of 6, but I chose 320 for all the others. That's through my macbook air with 10$ panasonic earbuds.

 

Not bad for my old, heavily abused ears - but it also shows that the encoding algorithms have gotten much better.

[modey]

Is there a graphable comparison somewhere of DACs? I read somewhere that the DAC almost doesn't matter at all.

[eugene]

probably because they got very good and cheap by this time, a top notch dac chip costs about 15$ on ebay. people with proper skills could build a full fledged dac board for less than 50$ i'm sure.

[Goiter Sanchez]

Got 3/6 right. Tougher to tell with the Jay Z and Katy Perry track I find due to how they're mastered.

[chim]

6/6 on iPhone + Koss Portapro, get on my level

 

It's funny how the Coldplay track sounds better in 128kbps as it masks the clipping from the aggressive mastering

 

Is there a graphable comparison somewhere of DACs? I read somewhere that the DAC almost doesn't matter at all.

Not sure when it comes to listening purposes but some high quality dac's can clip a waveform a couple dB's above 0 without generating audible distortion, and are often used in mastering for this purpose

[melancholera]

The vocal only track was by far the most obvious to me. Something about the breathy highs and the room ambience. Maybe it being very simple and naturalistic helped as well.

[joshuatxuk]

The vocal only track was by far the most obvious to me. Something about the breathy highs and the room ambience. Maybe it being very simple and naturalistic helped as well.

 

That's funny, that's the one I picked 128 kps for! (TWICE!)

 

Heard the classical piano one right away as lossless. Couldn't make heads or tails of the others, between 320 and wav anyway

[Adieu]

I refuse to take part in anything that wants me to compare the quality of Coldplay songs against one another

[Kooch]

You can't tell on compueter speakers, but with really good headphones or played on a good stereo it is a big difference in quality.

[may be rude]

depends what you're listening to. autechre is a good example of something flac serves well. some people whine about how flac doesnt matter, im not interested in proving anything to them. 320 takes the edge off, rounds things out, blurs and smudges them. sometimes the edge is key.

[Hugh Mughnus]

I refuse to take part in anything that wants me to compare the quality of Coldplay songs against one another

 

+1

[joshuatxuk]

1 out of 6, but I chose 320 for all the others. That's through my macbook air with 10$ panasonic earbuds.

 

Not bad for my old, heavily abused ears - but it also shows that the encoding algorithms have gotten much better.

[eugene]

was curious to see whether browser does something to the sound so i recorded the samples through stereomix and looked at the spectrals. all looks good, they're very easy to tell apart by the cutoff freq. too bad i can't hear beyond 15.5khz, would have been easy to spot. though it's still weird how 320kbps look, they should have all of the information up to 20khz, but it looks like the cut something in that spectrum.

 

[may be rude]

the frequency cut-offs are not built in to the mp3 encoding, but are just part of the preset of the encoder, which people almost always use. you can have 128s with full spectrum

[eugene]

hm, that's true, didn't know this. just don't tell it to what.cd people though, they'll go insane.

[Guest skibby]

http://www.npr.org/sections/therecord/2015/06/02/411473508/how-well-can-you-hear-audio-quality

 

A test trying to determine if people can really tell the difference between low quality mp3 (128kbps ), high quality mp3 (320 kbps) and lossless audio (flac).

 

I got 2/6 flac but only chose the lowest quality once. This was through my logitech 2.1 computer setup. Not the best listening station in my house.

 

I won't tell if you won't tell