Exactly. I posted the test in another thread: How Well Can You Hear Audio Quality? : The Record : NPR
I failed miserably. Couldn't tell 128 from uncompressed WAV more than half the time.
I failed miserably. Couldn't tell 128 from uncompressed WAV more than half the time.
rpo
Member
That was a fun test...never done one before. I got 5 of 6 of the uncompressed songs and the 6th I chose the 320 kbps version. Listening on monitor speakers probably helped a bit.
Wow, that's pretty impressive. Think you could tell without focusing on it and playing them back to back repeatedly? Obviously I know I couldn't.
Reminds me of when I rigged up a high end audio system with two cables, one a $0.99 Radio Shack thin wire and one a $100 top of the line wire to speakers and asked people to tell me which was which. Nobody ever heard a difference.
Reminds me of when I rigged up a high end audio system with two cables, one a $0.99 Radio Shack thin wire and one a $100 top of the line wire to speakers and asked people to tell me which was which. Nobody ever heard a difference.
There is a difference here though. The compressed music is actually "less" sound than uncompressed. Chopping off bits here and there to save space because they are not going to be reproduced by most systems anyway is a true defined change to the original music. To tell the difference you MUST use a system capable of resolving the missing bits, otherwise it will sound exactly the same. These systems are available, and there are quantifiable and objective ways to measure the differences.
If you fire up a standard bose stereo set and try to hear the differences between the uncompressed and compressed, you are going to say they are exactly the same. Try the same thing with a properly setup studio system in a nice quiet room, or a high end pair of headphones... The differences should be obvious.
Electricity, on the other hand, doesn't really care about the wires being used. Especially at the low amounts of power for firing up speakers. If you change the electrical properties of the wires (introduce things that could be pseudo capacitors, increase resistance, for example) then there would be a change to the resulting sound away from the original intent. Using wires, no matter the type/source, as long as they are at least "just big enough" and are not picking up outside signals will work just fine. I've never seen a reliable objective test on wires that proved they affect anything.
(This isn't true on analog signal cables though, they are significantly lower in power and can pick up outside signals or have dropoff or any number of other things which get picked up and magnified, but that's another show.)
If you fire up a standard bose stereo set and try to hear the differences between the uncompressed and compressed, you are going to say they are exactly the same. Try the same thing with a properly setup studio system in a nice quiet room, or a high end pair of headphones... The differences should be obvious.
Electricity, on the other hand, doesn't really care about the wires being used. Especially at the low amounts of power for firing up speakers. If you change the electrical properties of the wires (introduce things that could be pseudo capacitors, increase resistance, for example) then there would be a change to the resulting sound away from the original intent. Using wires, no matter the type/source, as long as they are at least "just big enough" and are not picking up outside signals will work just fine. I've never seen a reliable objective test on wires that proved they affect anything.
(This isn't true on analog signal cables though, they are significantly lower in power and can pick up outside signals or have dropoff or any number of other things which get picked up and magnified, but that's another show.)
I don't think a moving vehicle is this system so it in the context of this discussion, I doubt anyone could hear the difference in a car with any modicum of ambient or road noise.
Some of this is sooo wrong. The MS upgraded stereo is hugely sensitive to source quality. If you listen to compressed, it sounds like crap. If you listen to uncompressed, it comes alive. Yes, car is hostile environment for audio. But some setups expose source better than others, and in this regard the MS system is like grado headphones.
Jeff P
Member
Yes, this. I get mine next week, with the upgraded audio, but I honestly got it because I need XM. Well, I don't need it, but I hate being without it.
I've seen these debates countless times, and the science and related testing has demonstrated over and over again that there is a threshold where you can't tell the difference in terms of data compression bit rates. (Dynamic range compression, is of course a different problem.) I don't recall exactly where that is, and some kinds of music are more sensitive than others (in the same way that some pictures compress better than others), but there is a threshold. I find that a 256k MP3 is adequate most of the time, and the exceptions I found might honestly be due to poor encoding or original mastering.
And in the car, yeah, I don't buy that anyone can tell the difference with wind and tire noise.
I've seen these debates countless times, and the science and related testing has demonstrated over and over again that there is a threshold where you can't tell the difference in terms of data compression bit rates. (Dynamic range compression, is of course a different problem.) I don't recall exactly where that is, and some kinds of music are more sensitive than others (in the same way that some pictures compress better than others), but there is a threshold. I find that a 256k MP3 is adequate most of the time, and the exceptions I found might honestly be due to poor encoding or original mastering.
And in the car, yeah, I don't buy that anyone can tell the difference with wind and tire noise.
While I think I've always been blessed with great frequency range in my hearing (I can always hear a flyback transformer in the same room, and I can hear those HF mosquito apps that blast a 18-21 kHz tone), I find that 320K MP3's are just fine for me. XM and Slacker are too low of a bit rate, IMO.
Jeff P
Member
blind tests are not reliable in my mind in this area. Sound or taste are very tough to make empirical. I'm guided by how I feel after listening, and with MP3 I hear songs I like, with FLAC I hear the song and notice elements of songs I didn't hear in MP3. A cymbal ride here, a piano riff there... Very subtle. Not easy to blind test. But I do not think placebo.Until and unless you're subjected to a blind test in the car, couldn't all of it be the placebo effect?
For playback it doesn't matter if a file is 24bit/192kHz, what matters most is the mastering quality (other than your sound system).
Many Red Book CDs are ruined by the loudness war or otherwise poor mastering.
The only thing 24 bit achieves is a theoretically lower noise floor we can't hear, and cars have notably poor acoustics and are noisy - even in a Tesla.
192 kHz gives you headroom for sample rate conversion during mastering (speed up/slow down, convert to 44.1/48 etc.)
more info: 24/192 Music Downloads are Very Silly Indeed
Many Red Book CDs are ruined by the loudness war or otherwise poor mastering.
The only thing 24 bit achieves is a theoretically lower noise floor we can't hear, and cars have notably poor acoustics and are noisy - even in a Tesla.
192 kHz gives you headroom for sample rate conversion during mastering (speed up/slow down, convert to 44.1/48 etc.)
more info: 24/192 Music Downloads are Very Silly Indeed
Last edited:
Blind tests are not about assigning a numeric value to the quality of mp3 files. They are about whether or not you can differentiate between lossless or lossy formats. You sit down, you put on headphones. The music starts playing. Then you have to decide whether that was mp3 or wav. Then the same music starts playing again, and you decide what it was this time. Let's say you listened to the same 30 second section 10 times in a row and decided it was mp3, wav, wav, mp3, mp3, mp3, wav, wav, mp3, wav. Then they compare your guesses to the actual data. During these blind tests most people can not differentiate between 128 kbps and wav files. However some people can do that. Therefore 128 kbps is considered insufficient quality. However if you increase mp3 quality to 192 kbps, then people can't differentiate it from wav or flac.
This might be a placebo effect or maybe the mp3 files you are listening are less than 192 kbps or poorly created. You shouldn't be able to differentiate mp3 at 192 kbps from FLAC if you didn't know what format you are listening to. However in last few years this is becoming a non issue because increase in memory sizes. I prefer storing music files in wav format.
Last edited:
I can assure you that there are many poorly created recordings. All mine are CD -> FLAC encoded and the difference in how various albums sound is dramatic. (And as my hearing isn't all that good, so if I can tell the difference... Note that I probably couldn't tell the difference in with the same album encoded in various formats, but between albums there is a significant difference)
Similar threads
