How Well Can You Hear Audio Quality?

[Archsam]

I came across this website with an interesting listening test - 3 different samples of the same song, each one a different sampling rate (128kps / 320kps / WAV) and you have to guess the best sounding one.

https://www.npr.org/sections/therec...ll-can-you-hear-audio-quality?t=1589645030356

I got 3 out of 6.....

 

[Xulonn]

You are probably not getting responses because there are several threads on this topic based on tests of compression/bitrate vs audibility that are more detailed and sophisticated.

Did you do a search before you posted?

 

[Maxicut]

That test has been around for while & it's a good test, but only if you understand how audio works. You can pick the difference with the tracks that have a good DR, but the others you can't, because as far as your ears are concerned, a DR4 sounds the same as an mp3.

 

[Blumlein 88]

Yes, we've had at least a couple of threads about it. That's okay not like I do a search before any thread I start (though maybe I should).
For what it is worth I chose each correctly for 5 of the 6. On the JayZ my choice flipped the 320 MP3 and WAV. As I recall from the other thread a few people did the same missing on that same sample.

While listening I wasn't sure I was hearing a difference other than the 128 sample. But I must have been. I listened over headphones for this one. I don't remember if I tried it over speakers or not.

I just tried to find one of the threads to post a link, and couldn't using search. oops

 

[Jimbob54]

Yes, we've had at least a couple of threads about it. That's okay not like I do a search before any thread I start (though maybe I should).
For what it is worth I chose each corr
ectly for 5 of the 6. On the JayZ my choice flipped the 320 MP3 and WAV. As I recall from the other thread a few people did the same missing on that same sample.

While listening I wasn't sure I was hearing a difference other than the 128 sample. But I must have been. I listened over headphones for this one. I don't remember if I tried it over speakers or not.

I just tried to find one of the threads to post a link, and couldn't using search. oops

Well I know I have golden ears because I have invested money in this hobby so clearly the test is flawed and *anxiety increases* there must be high levels of radiation in the atmosphere today *sweats" they must have used a good encoder on the Suzanne Vega 128 *chews lip*...



(Every time I do it I get the classical right, mix some 320 with wav and probably pick one or two 128)

 

[Maxicut]

It's designed so you can only accurately pick 4 of the 6 correctly.

 

[Jimbob54]

It's designed so you can only accurately pick 4 of the 6 correctly.

REDEMPTION!!!!!

 

[Xulonn]

Yes, we've had at least a couple of threads about it. That's okay not like I do a search before any thread I start (though maybe I should).

LOL! ¡yo tambien! I think you recognized that I wasn't being judgemental - just making an observation...

And I even posted once about not liking the ASR search engine interface.

 

[Robert394]

It's designed so you can only accurately pick 4 of the 6 correctly.

How so? I did pick 4 out of 6 correctly. I guess I'll trust my own judgement more when A/Bing at home.

 

[Maxicut]

How so? I did pick 4 out of 6 correctly. I guess I'll trust my own judgement more when A/Bing at home.

Well done, you picked the 4 that were possible to pick without guessing! ... so I don't understand the question?

 

[Robert394]

Well, I mean, can you elaborate when you say it's designed so you can only pick 4/6 correctly? In what way is it designed that way?

 

[Tks]

Yes, we've had at least a couple of threads about it. That's okay not like I do a search before any thread I start (though maybe I should).
For what it is worth I chose each correctly for 5 of the 6. On the JayZ my choice flipped the 320 MP3 and WAV. As I recall from the other thread a few people did the same missing on that same sample.

While listening I wasn't sure I was hearing a difference other than the 128 sample. But I must have been. I listened over headphones for this one. I don't remember if I tried it over speakers or not.

I just tried to find one of the threads to post a link, and couldn't using search. oops

Same here, the JayZ one is just a mess of noise more than anything. I couldn't AB that at all, tested myself 33 times just for that track (was going to do 40) but yeah, I can't pick out any reliably of that song.

 

[Sal1950]

Same here, the JayZ one is just a mess of noise more than anything.

That's true of all his stuff. LOL

 

[Maxicut]

Well, I mean, can you elaborate when you say it's designed so you can only pick 4/6 correctly? In what way is it designed that way?

Read my 1st post

 

[mansr]

That test has been around for while & it's a good test, but only if you understand how audio works. You can pick the difference with the tracks that have a good DR, but the others you can't, because as far as your ears are concerned, a DR4 sounds the same as an mp3.

What gave you that idea?

 

[miero]

This test is obsolete. Nowadays streaming services are using more modern formats than MP3.

From a recent release of Moby song on Youtube:

249          webm       audio only tiny   59k , opus @ 50k (48000Hz), 2.08MiB
250          webm       audio only tiny   78k , opus @ 70k (48000Hz), 2.79MiB
140          m4a        audio only tiny  130k , m4a_dash container, mp4a.40.2@128k (44100Hz), 5.32MiB
251          webm       audio only tiny  150k , opus @160k (48000Hz), 5.53MiB

 

[Archsam]

Actually the point of the test is more relevant than ever - if some of us are still unable to reliably pick out the 128kbps mp3 from the lossless WAV, what is the point of chasing ever higher sampling rates and higher resolution playback gears?

It's a uncomfortable thought, but the bottleneck in the whole audio reproduction chain might be my own ears, and all the better gears with more accurate frequency response and lower distortion won't make a big difference because my ears can't tell the difference.

When I came across the test I just casually listened with my old B&W P7 plugged into my laptop, so the sound quality is hardly top notch, but maybe I'm just fooling myself by blaming the gears

 

[mansr]

as far as your ears are concerned, a DR4 sounds the same as an mp3.

Regarding that claim, here's a synthetic sample with practically no dynamic range at all, yet it's easy to tell the original from the version that has been through an mp3 encoder.

 

[Jimbob54]

The worst to listen to was Coldplay for me. That just sounded horrid in all formats. Agree about the Jay z, too hard.

 

[Archsam]

The worst to listen to was Coldplay for me. That just sounded horrid in all formats. Agree about the Jay z, too hard.

Same here, the Coldplay track was the only one I picked the 128kbps mp3