256 vs. 128 kbps: Can You Tell the Difference?

The difference is more obvious in the 256kbps Mozart piece (the strings don't sound so "mushy"). The R.E.M. 256kbps excerpt has more clarity or brilliance. Now, I'm listening on headphones and I usually don't mind casual (brief) music listening this way or on my PC's speakers. However, it's my personal experience that almost anything encoded at less than 320kbps played on my living room stereo system draws my attention to the low quality and it quickly becomes annoying because of that.

Buying music at 128kbps is like buying a book that only shows every other letter - you don't experience the work as intended by the author. Remember, lossy compression is named that way for a reason.

My day job is building and operating extremely large concert audio systems. Low bit rates MP3's are quite obvious even to the untrained ear. Totally acceptable for my computers multimedia speakers, but in a high quality audio environment, nothing beats a good old C.D.

I don't want to sound impolite, but this 'test' is completely meaningless unless you enlist someone's help and perform it on yourself blind, without knowing which you are looking to.

Using high-end headphones and a variety of classical music, under blind conditions I cannot tell any difference between original CDs and 128K AAC. I may *think* I can--and until it sank in that I was doing no better than chance, I *did* think I could--but I can't.

So there's really no point in encoding in anything higher than 160K AAC, so far as I can tell.

Similarly to peteraudrain, I thought that I could tell a difference until I called somebody to play one or the other randomly. Not only could I not tell which one was higher quality, I consistently thought that the 128K version was the 256K, and the 256K version was the 128K.

I certainly agree that there is no point in enoding high bitrates, I'd rather have the space for my application of digital music BUT take the high quality headphones and classical music out of the test - Use a big audio system, crowded noisy arena or auditorium and a music recorded in the sixties and seventies and you have a situation where the songs just "sound different." And, it's easy for me (and the boss) to distinguish because his 1/2 million dollar P.A. system suddenly sounds like a clock-radio(to us.) The crowd does not hear anything out of the ordinary and when a contemporary recording comes up, everything sounds normal again.
I was just cutting my teeth in the concert biz when C.D.'s overtook vinyl and cassettes and thought the audiophiles of the day that hated CD's were kooky!

I did the test "blind."
My wife chose which version to play without my being able to see which one she chose. I picked the higher bitrate version every single time over multiple plays (3 times for each version for a total of 12 tests). Since turnabout is fair play, I did the same for my wife. She scored 100 percent for Mozart, but R.E.M. didn't do so well -- she couldn't tell the difference at all.

Will the same test work for everyone? Not a chance! The phrase "tin ear" didn't arise without reason, after all (sorry, sweetie).

As for storage purposes on something like an iPod, well yes, lower bitrate is better if you want to tote more music. I don't tote my living room sound system with me, so disk space is irrelevant.

Higher bitrates are always better from a sound quality standpoint. Period.

With a high-end audio system or quality headphones, the difference is apparent. However, the majority of people will be hearing their music through iPods, PC speakers, cheap headphones, or worse yet on their car stereo through an FM modulator (yikes!). These items cannot possibly take full advantage of the higher bit rate.
Audiophiles will continue to rip their own music at the highest bit rate possible, file size notwithstanding. If we need more storage space, we buy another hard drive. To me, the higher bit rate is worth the additional 30 cents.
Regardless of the bit rate, I am glad to see EMI take the step of dropping DRM. I always viewed DRM as the equivalent of not being able to play your new pressed CD in your car because you first played it in your home system. "Oh, you want to play it in your car too? You have to buy another copy of the CD to do that."
Way to go, EMI.

Quality over quantity. Lossless compression audio formats like .shn or .flac are the way to go. With 1 terrabyte hard drives soon to arrive why save on disk space ? There is nothing wrong with listening to uncompressed .wav files either. In 10 years propietary music file formats will be dead (excepting Microsoft's) - duh...

The 256Kbps is nice and all but I have 30 purchased Itunes songs and I riped a lot of my music. I terrabyte drives it's ok but I don't have that. Besides I love my current HP and they'll never get me to throw it away.

192kbs MP3 still the best overall. I think this should be "INDUSTRY STANDARD". I have been ripping and recording at this rate using LAME encoder for years and it seems to have the best sound for my ear as well as a manageable file size. Sound quality is at or above any CD that i have heard.

It takes more than just good speakers to listen to this and tell a difference. I have some very nice Klipsch speakers plugged into my machine, but I still have a hard time hearing the difference. The acoustics in the room aren't great, I can't turn the volume up too high, and the window is open. This is why Apple chose 128kbit in the first place; most people don't have the opportunity to listen on headphones or speakers where it makes a difference. That said, Metric's tune, "Monster Hospital" suffers really badly when encoded. The cymbals and vocals end up mushy and washy, no matter what you do. I've got some Shure e2 headphones coming next week; I'll try this again. With proper sound isolation, I'll probably be able to hear a bit of a difference.

Having a high bit rate is not only about being able to hear the difference. It's also about future-proofing your music for many things you may want to do with it later. Maybe you can't hear the difference on your computer speakers, but you might end up buying a top-notch stereo system where you can hear it. If you need to convert the song to another format to play it on a different medium, it will exacerbate any artifacts present in the audio, possibly making previously inaudible defects audible. If you want to sample the music, perhaps playing around with the time scale or applying some other effects, small defects can become magnified and much more apparent.

Lossy encoding always discards some information. Even though it may seem like that information isn't important, you never know when you might miss it. As many others have pointed out, storage is dirt cheap and getting cheaper. There's no real benefit to squeezing the file down to such a small size.

Having this test not be double blind makes it hard to actually see if there's a difference, or if it's just psychological.

So I made a little program to give you a double blind test for two audio files! Download it from http://iscs.us/AudioDueler.zip

Does anyone know the name of the Mozart clip used? The name or the album on which it can be found will be greatly appreciated.

It is very hard to tell the difference between the two for Mozart because it has A LOT less complication in sound. Therefore, the compression does not reduce the quality nearly as much as it does for music that has all kinds of sounds with all different kinds of frequencies. Cymbals, for an example, show a great deal of distortion when compressed. In the R.E.M. song, the difference between the two is obvious, and the cymbals are the greatest difference.

192kbps is definitely not at CD quality or better. That is a VERY false belief. CD quality is strictly 1411kbps, and if you rip a CD into the FLAC format, it only compresses information that is being wasted, and still keeps the full quality of the sound intact. This is what I have most of my music in because with $70 headphones, MP3s really sound like crap.

A better poll would be to ask if anyone can tell the difference between 128kbps and 256kbps blindly without you labeling it. Have a total of 10 clips, which 5 each. The only thing with this poll is cheaters could verify which version would be larger size, so you would need to mask that some how. Then, right a report about how well everyone did based on your poll above where I see many said they can hear a difference. Now, this report , in my opinion, would carry more weight -- because, I don't think there's that much of a difference. - Evan

Being an audiophile who's dumped a bunch of money in hi-fi I have learned the hard way that the principles of smoke and mirrors do exist. Many audio lovers are equipmentphiles not audiophiles. A true audiophile who love musicality. Plus most people don?t like audiophiles anyway. But of the 2 bit rates 256K has the best sound. And of the 2 clips its more noticeable with the vocal in the REM piece. For reference audio with a human voice, piano, harp, are a real litmus test. And Sony SACD is dead!

The difference in quality between the 128kbps and 256kbps is VERY noticeable to me. I rip all of my cds at around 240kbit VBR - I define that as "listenable" quality. Anything lower than around 224kbps, I can detect lossiness in the audio. So I'm really happy about this higher quality, DRM-free music from iTunes. =) If they go through with this plan, I will be more confident in buying from the iTunes store again. The 2 things keeping me from the iTunes Store in the past were: 1) The horrible sound quality, and 2) The DRM. By removing these two obstacles, I will be more than happy to legally buy music online! Thank you EMI!

I tried this DOUBLE BLIND. I put the files in a play list in my iTunes, set the mode to shuffle (with repeats allowed), hit play, and wrote down answers for the first 32 clips I heard. Next I went to "recently played" and checked my answers. I scored 11/16 on the Mozart piece and the same with R.E.M. I agree that the human voice is a good reference, it was much richer in 256kbps. Also, the violin sounded dull to me in the Mozart piece when played at 128kbps.

I have a very high-end car stereo (all Xtant components) and I can impress the masses with SQ even at 128 kbps. If you can't tell the difference, dont feel bad, it just means you can save tons of money on your stereo.

My test system consisted of two Marantz Imperial 7s and a Pioneer SA-850, all vintage. My window was open, and my room is not ideal for listening. I imagine I could tell better in a purpose-made sound room.

Apple: Try $0.89 per song for 128kbps, $1.09 for 256kbps. $1.29 puts the price of a CD at $16!!!!!

I have been paranoid about acquiring the best sound at the different bit rates in ripping music for some time now. I have a vast music collection and started ripping at the itunes default rate of 128kps without knowing I could rip at higher rates a few years ago. After hundreds of CDs later I realized I could rip at higher rates. If I decided to go back and re-rip at 256kps or higher it would take me months and months to re-rip. However, being as paranoid at having the best in ear quality I have compared time and time again with various songs to distinguish clarity and quality between 128kps and higher bit rates. Needless to say I cannot tell the difference for the most part and I listen to most of my music with my Shure E4 headphones. Obviously this means that I am no audiophile even though I love my music and have tons of it. Guess I will be saving myself some time and money, which is a huge relief.

I've read the article and through some of the comments and I have to wonder what everyone is encoding from that would give them any better encoding than using 128 kbps. CD's wouldn't do it; CD sample rate is 44100 kHz with 16-bit sample accuracy. Doing the math, that's 1411.2 kbps of raw audio signal. After making the adjustment for MP3 or AAC compression (both being about 11:1), you end up with 128.29 kbps. Regardless of what is believed, any higher bit-rate MP3 or AAC encoded from a CD cannot sound any different.

Are you encoding from an analogue medium such as vinyl? Higher sample-rate medium such as DVD-A? As an audio engineer, I am curious. I've done my own comparisons many times in the past and want to compare apples to apples.

About the tracks provided through iTMS: Since EMI has access to the originals, naturally a higher bit-rate product is possible and does provide a better quality experience. Whether the listener can hear that difference will differ from person to person

256 kbs is better than 128 and I am very happy ITunes has started to release songs in this format. I have a benchmark DAC connected to 300b valve amp and efficient speakers and to me 256 is as good as CD! The problem is that recorded music is inconsistant in its quality so the limiting factor is often the source material anyway. CD itself is by no means perfect and music quality suffers from quantizing to 16bit 44khz. I would ideally like 24bit 192khz (mp4 compressed with v low loss) sold on itunes, I would pay little extra for this and that way we would be getting better than CD quality with convenience of on line download. But life is often comprimise and 256 is step in the right direction and should equal CD quality, the most imprtant thing is access to fantastic new music plus live recorded music as close to how the artist intended. Enjoy and listen to the music not the inadeqacies of the technology reproducing it :)

Surprise!: I often seek out or create **96k** encodings after hearing and rejecting 128K or higher. 96k often sounds much more "real" and the difference is obvious. Try it and see -- but try a variety of songs, as 128K is often superior instead. This gets back to the CD-versus-tape debate: CD can be too "perfect" to feel real, with too "gentle" a sound. Similarly, if I can find an analog radio to use, I will sometimes slightly detune it to add a small amount of distortion to the audio. You might call me a fringe or renegade audiophile. I care about what sounds best to me, not what is most technically perfect. This depends very much on the kind of music involved.

I see that the results are varied in the comments. That is to be expected, because of two variables: 1. your ears. If you have listened to blaring music for a long time, etc, you will lose your ability to distinguish quality. 2. your audio equipment. If you listen to music with integrated sound and $15 speakers, chances are you won't see a big difference. I have a $280 sound card and $115 5.1 speakers (not extremely expensive considering home systems). I can hear a tremendous difference between 192 kbps and 1411 WAV. Yes, I still encode my CD's in lossless WAV. Sometimes the difference is slightly more subtle, but usually it is brought out with many simultaneous instruments or background sound. Until proved mathematically, those who claim 128 is the best sound are full of crap or can't hear the difference.

Higher quality offerings are great, but except for singles they won't interest me until bit rates are WMA lossless or WAV.

By coincidence, I've just today been ripping some CDs to wma format. I started at 64kbps, thinking that wma is more efficient than mp3. Then I noticed that the sound quality was distinctly dull compared to the original CDs. So I tried 128kbps. Still not a big improvement. So I tried 192kbps - that did the trick. On my good (but not super) computer speakers, I can hear no difference between 192kbps wma and CD audio. i think the word I'm looking for is ambiance. 128kbps doesn't have it.

Conclusion is that the move from 128 to 256 is very worthwhile to anyone who has invested in even moderately good hi-fi equipment.

Personally, I never cared about file format, but i found that when I heard someone else play a song, it sounded much much better, like there were subtle sounds that were just nonexistent in my own low quality version. I would say, if you like to listen to music, use a higher quality format, if you just like to have music playing in the background of things, then you'll never notice the difference.