[paulr]

I wouldn't suggest mp3 as an archiving format regardless of bit rate. It doesn't matter if there's no audible difference from the original, since there can still be differences that may come out after (say) conversion to some other format. In other words even if wav and 320 sound identical, wav->128 may give a better 128 result than 320->wav->128. So I would do all my archiving with flac which is around 750 kbps. As the other poster says, hard drive space is dirt cheap these days. For listening on a portable, 128 kbps mp3 has been good enough for me most of the time, even if I can tell the difference from wav if I listen carefully. I have limited space on my current portable (4gb flash) so I usually put 128's on it, but keep the flac's on my computer.

 

[bordins]

You need an accurate measurement equipment in order to verify differences do exist.

Untrained human ears (brains) are the least reliable equipments !!

Who care, BTW ?

 

[uhcmos1]

It is not going to change the overall tonal balance.

 

[Riordan]

Quote:

Originally Posted by eightbitpotion /img/forum/go_quote.gif Also this dumb crap about how it sounds better through certain SOFTWARE players and different OSes is simply ignorance- if it's digital...It's either 0 or 1...aka it either works or it doesn't (that's how digital works) ...

this part of your post needs clarification: you would be completely right if bit-perfect reproduction were the norm in home computer audio. it isn't.

resampling is a very real issue, and choice of os/audio player/setup HAS an effect in SOME cases. on windows xp systems (as well as win98 ), digital audio data is routinely converted from 44.1 khz (cd standard) to 48 khz (the legacy pc audio standard). with some soundcards, audio data is even resampled twice. old soundblaster cards were the prime offenders.

whether your computer setup is affected by resampling depends on your combination of hardware (while creative labs cards prior to the x-fi series resample no matter what, there are very cheap sound cards that output bit-perfectly with an open source driver but not with their original drivers) and software (while windows xp kmixer resamples everything, many audio players provide a way around it, by using asio or kernel streaming).

now whether the resampling done by card, driver or os is audible or not is a similar question to the one about 192/320kbs mp3s, but sadly cannot be tested as easily by using foobar abx...

 

[kendal3334]

with my ZVM and MX500, i can find some differences between 320K mp3 and WAV, but they appears only on the high frequency.

with my other stuffs i'm not listening to mp3.......

 

[928GTS]

I can certainly hear the difference between MP3's in various bit rates and FLAC and I can hear a decent difference between,say,a 160kbps file and one ripped at 320kbps. However,when you're getting music from sources besides CD sometimes you just don't have a choice.

If I booted up a filesharing program or bittorrent I'm not going to spend an extraordinary amount of time searching for the highest quality rip I can find. I'll give it my best effort,sure,but in the end I want to listen to music and enjoy myself.

I'm eventually going to buy a drive and devote it entirely to FLAC recordings just so I can have that reference quality there should I desire it.

 

[krmathis]

Quote:

Originally Posted by kendal3334 /img/forum/go_quote.gif with my ZVM and MX500, i can find some differences between 320K mp3 and WAV, but they appears only on the high frequency.

No wonder, since 320Kbps files run through a low-pass filter during encoding.

 

[chadbang]

Quote:

Originally Posted by poo /img/forum/go_quote.gif Not sure if this would work... but bear with me and let me know what you think. What if I were to create 2 files, both the exact same section of music from the same disc; one encoded at 320kbps and the other at 192. I would then convert the files back to .wav and join them into one file so that one would play first, then the other. I can create a poll on Head-Fi asking which section (the first or second part of the track) is which.

I recently tested myself (and my new Kef 103.2 reference speakers). I took Pink Floyds "Wish You were here" and ripped it into Lame at 256Kbs. Then I burned it back to a cd as a wav/aiff file (I'm on a Mac and we use .aif not .wav).

I could easily hear the difference between the original cd and the 256k file. The mp3 lacked air and bass roundness compared to the original.

But I only rip to Mp3 for my ipod, so the difference is probably -- probably -- moot. But comparing the 2 formats on a $2k system, yeah, no problem telling the difference. So many of the guys here who say they can hear the difference are probably the guys with the good portable rigs. They're not being snobby audiophiles, the just have killer portable systems.

 

[milkweg]

I've done ABX in foobar with 192kb/s mp3 and lossless wav on myself and a friend and we both failed to tell the difference. This was on HD595 headphones via an EMU 0404 used as a DAC and amp with a conection to an X-Fi via SPDIF coax. I used both classical and rock music too. Perhaps that equipment is not resolving enough to tell the difference?

 

[Seamless Sounds]

Our hearing threshold goes up to about 20khz. A 192kbps gets chopped off at around 16khz. A 320kbps goes up to 20khz, the human hearing threshold.

I've done some tests on myself. The 4khz that was chopped off is extremely quiet. All you hear in this region is a tweetering sound, like a very high pitch whistle. However I was able to isolate the high frequency (>16khz) from the lower frequency (<16khz) and compared both 320kbps and 192kbps (at frequencies greater than 16khz). Since it's so quiet, I upped the volume. The difference is like night and day. The 192kbps sounded lifeless like a dog limping while the 320kbps sounds fuller like a dog running with spirit. Of course if your cans are more treble oriented or totally flat (no huge drops in the high frequency end) then you could probably hear the difference.

Soundstage is also different. I took snapshots of both songs at different bitrates. The top pic is 320kbps and the bottom is 192kbps.





At first I thought stereo imaging wouldn't be affected much. Apparently when it's encoded, it's altered somewhat.

 

[knights]

i can hear dif between 128, 192 and 320.. but i failed in 320 vs. loseless... im using v0 VBR mp3 by lame3.98 via foobar...

source: miniONE notebook + ibasso boa + sen eh-350

 

[indianbraker]

im pretty sure 98% of people cant hear the difference between 320 and lossless.....i guess the ones with 10k in equipment but i have never been able to tell the difference myself.

 

[Currawong]

How about seeing the difference?

http://www.head-fi.org/forums/f15/lo...-p-7-a-225356/

 

[milkweg]

Quote:

Originally Posted by HeadFi Fanatic /img/forum/go_quote.gif Our hearing threshold goes up to about 20khz.

Most people over 30 can not hear above 17k. That's a fact and not opinion. In the UK they are testing a method to keep teenagers from loitering outside certain stores by playing high frequency tones that only younger people can hear just for that reason.

I just ordered a pair of BeyerDynamic DT-990 so will run the test again with those headphones too. I still expect I will fail to tell the difference because I am 50 years old so my ability to hear really high frequencies is poor unless I really crank the volume. I have a test cd that will play all the frequncy tones and have tested my ears with that and can only hear above 17k if I turn the volume up really loud which is way above the level I listen to music at and even then it is very faint. Headphones are tuned to roll off the high frequencies anyway so they don't hurt our ears because of how close the headphone speaker is to the ears compared to stereo speakers.

Seeing the difference has nothing to do with hearing the difference. It's like the light spectrum, we can only see a small part of the light spectrum with our eyes.

 

[krmathis]

Quote:

Originally Posted by milkweg /img/forum/go_quote.gif I've done ABX in foobar with 192kb/s mp3 and lossless wav on myself and a friend and we both failed to tell the difference. This was on HD595 headphones via an EMU 0404 used as a DAC and amp with a conection to an X-Fi via SPDIF coax. I used both classical and rock music too. Perhaps that equipment is not resolving enough to tell the difference?

It can be the equipment, or something else. Like:
* The music. Since some music are harder to lossy compress and keep transparent than others. Ex. low tempo classical vs. metal.
* Your ability to hear artifacts.
* and probably more...