Jump to content
  • 0

NW-S706F ATRAC3+ vs. MP3 audio quality test

Rate this question


DonCan

Question

I spent hours today experimenting and listening on my new NW-S706F. Used ATRAC 3+ 128, 96, and 64kbps to convert three songs. The files converted were native MP3's, two @ 192 and one @ 224kbps. I transferred all the files over onto the player and listened to all the different quality versions. That means for each song, there were four files transferred, one in the native MP3 format and the others in the mentioned ATRAC 3+ bit rates. I can honestly say that I could not tell the difference between the ATRAC3+ 96kbps and the 192 and 224kbps native MP3'S. However, I did notice slightly lower quality on the 64. I also want to say that I'm very pro-quality but if i can't tell the difference, then why transfer my files at 192 when i can almost double the space if I convert to ATRAC3+ at 96kbps. Anyway, those are my tests, and they seem in line with Sony's claims on ATRAC quality. If I did notice lower quality in the ATRAC3+ 96kbps, it was psychological because the listening tests were not blind and I knew I was listening to a lower bit rate file. If I had gotten someone to play the files to me randomly and ask me which sounded better, I don't think I would've noticed a difference.

I think if these tests were done on high fidelity equipment with top quality head phones, then we'd hear a difference, but these are more normal conditions.

(I also want to add that sonicstage made this process a pain in the ass as it doesn't allow you to transfer more than one file to the player that's the same song even if they are at different bitrates, so it took much longer than it needed to find a way around it)

A sidenote, I tried playing the native MP3 from my laptop and then to the same ATRAC3+ converted file @ 96kbps on my NW-S706F using the same headphones and to my surprise, the audio actually sounded BETTER on the MP3 player. A testament to the amazing sound quality of this particular player and Sony DAP's in general as it outperformed the sound card on my laptop!! (It's a decent laptop)

I think if the self professed audiophiles out there (myself included) would put our pride aside just for a bit, we'd double our music capacity. ;) But hey, if you can hear the difference, then stick with the bit rate/CODEC that makes you happy. I originally transferred about 3.5 gigs of native MP3's at mostly 192kbps. After this test, I converted them all to ATRAC3+ @ 96kbps and retransferred them and now I can fit double the number of songs! :)

Edited by DonCan
Link to comment
Share on other sites

11 answers to this question

Recommended Posts

  • 0

I think the proper way of testing is by converting from audio cd to atrac and same audio cd to mp3. if you convert from one compressed format to another, the later is going to suffer. In your case, the atrac is from mp3. the resulting atrac quality will not be the same as getting atrac direct from audio cd(or any uncompressed source). Mp3 is a lossy format, so by no chance resulting atrac can have any better sound. Pls correct me if I am wrong.

Link to comment
Share on other sites

  • 0

I think the proper way of testing is by converting from audio cd to atrac and same audio cd to mp3. if you convert from one compressed format to another, the later is going to suffer. In your case, the atrac is from mp3. the resulting atrac quality will not be the same as getting atrac direct from audio cd(or any uncompressed source). Mp3 is a lossy format, so by no chance resulting atrac can have any better sound. Pls correct me if I am wrong.

you are right. especially reducing the bitrate to half will not yield quality results.

Link to comment
Share on other sites

  • 0

I agree, but this test was more to see if I could notice the difference when I convert MP3's to ATRAC3+ because I was comtemplating whether or not to transfer my files in native MP3 or convert them. Realizing that converting one lossy format to another could possibly affect quality, I wanted to make sure I could not tell the difference before I proceeded.

I'd definitely love to do a real test and rip from a CD like mentioned, but man, these tests take so long, haha. I don't think I'll be doing one in the near future. Plus, sonicstage lags quite a bit which doesn't help. But on a positive note, Sonicstage does not alter your native MP3's when you transfer them in ATRAC format, otherwise, forget about it.

Link to comment
Share on other sites

  • 0

It's obvious that a good listening test has to be doen with files ripped directly from an audio-CD (preferably retail)

to the target codec.

This provides the best results.

greetz

I agree, but this test was more to see if I could notice the difference when I convert MP3's to ATRAC3+ because I was comtemplating whether or not to transfer my files in native MP3 or convert them. Realizing that converting one lossy format to another could possibly affect quality, I wanted to make sure I could not tell the difference before I proceeded.

I'd definitely love to do a real test and rip from a CD like mentioned, but man, these tests take so long, haha. I don't think I'll be doing one in the near future. Plus, sonicstage lags quite a bit which doesn't help. But on a positive note, Sonicstage does not alter your native MP3's when you transfer them in ATRAC format, otherwise, forget about it.

A good test to perform is to listen to a range of full quality audio formats/files etc.

So begin with a Music DVD or SACD, then try a CD (obviously going down in quality very slightly..) and then try atrac and mp3.

For those without a SACD or Music DVD then just try a cd then a ripped version of the track on e.g cd as a mp3 and atrac. trust me there is a considerable difference! Even to the average perosn on the average player.

The song commonly used throughout tests at work are 'mona lisas and mad hatters' re done by 'heart'.

very good example.

Many Thanks

Bueller

Link to comment
Share on other sites

  • 0

Wow, 5 people saying the same thing, and even the first response was redundant :P

It is obvious that transcoding is 'bad'; I'm sure OP understands that general rule too, and is in fact why he did this test (in part) -> it doesn't matter what is theoretically true if your ears don't pick up a difference. This difference is harder to pickup via a portable(* since most people's portable gear gives lower fidelity than their stationary gear), in general, and since most portables have limited space it's obviously a good idea to see where you pick up the difference with your ears. Everything else is dingbats. And if someome wishes to save space and battery power by using a lower-bitrate atrac codec and already has a lot of files as mp3s, chances are it's going to be very messy to find every source and re-encode as that lower-bitrate atrac format.

The topic was a bit misleading though, still it wasn't totally blinding.

Link to comment
Share on other sites

  • 0

Probably having too much time to spend, I performed a similar (completely unscientific) test just a few days ago, where I compared the following:

1) CD converted to MP3 @320

2) CD converted to MP3 @192

3) CD converted to ATRAC3+ @256

4) CD converted to ATRAC3+ @192

5) CD converted to ATRAC3 @132 (LP)

6) MP3 @320 transcoded to ATRAC3+ @256

7) MP3 @192 transcoded to ATRAC3+ @256

I used my Sony NW-HD1 (upgraded to native MP3 playback capability) to play the tracks and I used both my Sony EX71's and my Sony MDR-CD780 headphones to listen. EQ was of course set to off.

ATRAC3+ @256 sounds the best, by far. I was expecting the MP3 @320 to sound better, but for some reason it sounds "duller" than the ATRAC3+ @256.

ATRAC3+ @192 is nearly as good as ATRAC3+ @256. It is not easy to tell the difference between these two bitrates except for some tracks where the 192 displays some slight loss of "edge".

MP3 @192 is slightly inferior: some minor "swishing" can be heard in some places and again, a certain loss of treble.

I could very easily hear annoying audio artefacts at ATRAC3 LP @132, so I decided not to go further down the scale than this.

I next moved over to the transcoded files, expecting to hear quite a sonic degradation. Surprisingly enough, I was completely unable to hear any degradation whatsoever with the transcoded files from MP3 @320 to ATRAC3+ @256. Whenever I heard a difference, the transcoded files sounded better than the original MP3. I know it should in theory not be possible but that is still the case with my equipment.

On the other hand, the files transcoded from MP3 @192 to ATRAC3+ @256 were quite easy to pick out, mainly because of a loss of "sparkle". If listened to by itself, I would probably not hear anything wrong with these tracks, but when comparing them to the original MP3's, the loss of definition is fairly obvious.

My quality scale goes like this:

1) ATRAC3+ @256

2) MP3 @320 transcoded to ATRAC3+ @256

3) MP3 @320

4) ATRAC3+ @192

5) MP3 @192

6) MP3 @192 transcoded to ATRAC3+ @256

7) ATRAC3 @132

My conclusion is that the either ATRAC3+ is more suited to my ears than MP3, or that my NW-HD1 is in some way biased to be able to play ATRAC files better than MP3 files.

Tom

Link to comment
Share on other sites

  • 0

I'd like to mind you guys that early Netwrok walkmans with native MP3 support were not fair to this codec.

Sony muffled the sound of MP3 and EQ etc. was not usable with MP3.

So in that way it's not a very fair comparison.

If I listen too MP3 and atrac3plus on my NW-S700 I barely notice difference. If I'm going to sit in a quiet room all alone, Yes then atrac sounds better.

But in daily use it's really not noticable. Expecially if you use good MP3 encoders (LAME 3.97 for example)

greetz

Link to comment
Share on other sites

  • 0

I'd like to mind you guys that early Netwrok walkmans with native MP3 support were not fair to this codec.

Sony muffled the sound of MP3 and EQ etc. was not usable with MP3.

So in that way it's not a very fair comparison.

If I listen too MP3 and atrac3plus on my NW-S700 I barely notice difference. If I'm going to sit in a quiet room all alone, Yes then atrac sounds better.

But in daily use it's really not noticable. Expecially if you use good MP3 encoders (LAME 3.97 for example)

greetz

Recent listening tests conducted by sound coders @192 kbit/s have shown mp3lame 3.97@vbr2 setting to be very good, but not as good as good aac codecs.

I would not make the mistake and compare mp3s from the days of old encoded with defunct crap encoders to current atrac3+

If you compare lame mp3 to atrac3+ you will be hard pressed to identify any difference at all. DSP´s comments on once subpar mp3 implementation are right on

but fortunately thing have changed at S.

However, I believe the future will belong to aac. Compatibility is already there with Apple, Sony, Zune, Cowon, Archos and virtually all cell phones out there.

Embracing AAC was actually a pretty smart move by Sony.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...