[james444]

   
With the amount of heavy heads around here, you all would make a high level sound engineer jealous! (compliment)
 
With that say, from a layman's point of view(they stopped me from using newbie), and according to your opinion, than FLAC is basically useless across bluetooth?
 
Just trying to understand from a layman's view... 

 
Lol @ heavy heads...
 
As for the bolded part: nope, you've got me wrong. According to my opinion, FLAC is basically useless except maybe across bluetooth!

Sounds outrageously counterintuitive, no? Let me explain in layman's terms:
 
There's ample evidence that most people can't distinguish high quality mp3 from FLAC in a blind listening test. Just google for "mp3 flac abx test" and do the math. There may be some among us with golden ears, but for the vast majority 320kbps mp3 sounds just as good as FLAC. However, that doesn't mean that there's no loss of information with mp3, it just means that we wouldn't have heard the lost information anyway.
 
So, just for the sake of the argument, let's assume that high quality mp3 encoding loses 5% of the original information. And let's assume further, that SBC encoding (which is worse than mp3) loses 10%. And lastly, let's assume that our ears won't notice a sound degradation, as long as at least 90% of the original information is still present.
 
Now let's put two and two together:

1. FLAC encoding of the original file will retain 100% of information.
2. mp3 encoding of the original file will retain 100% - 5% = 95% of information.
3. Since both files are above our 90% threshold for audible sound degradation, the 100% FLAC and 95% mp3 will sound equally good to us.
4. Now let's reencode those files to SBC and send them across bluetooth...
5. SBC reencoding of the FLAC file will retain 100% - 10% = 90% of information.
6. SBC reencoding of the mp3 file will retain 95% - 10% = 85% of information.
7. Which means, that after reencoding and bluetooth transmission, we may hear mp3 as worse than FLAC, because its quality is now below our threshold of 90%.

 
Disclaimer: this is a very simplified example to illustrate a point, with no claim of validity whatsoever. It just means to show that while mp3 and FLAC sound equally good for most listeners over a wired connection, there may be a reason to prefer FLAC over mp3 when we're dealing with bluetooth transmission and double compression.

 

[Soundofmusic]

I would assume you can't treat degration as "x % per reeconding pass", because as you state yourself the codecs are based on removing the parts of the spectrum the human ear can't percieve. I.e. after that information has been removed in the 1st pass, the codec wouldn't do much for the consecutive reencodings. Or to quote someone more linguistically capable in english :
 

Theoretically/philosophically it's possible to avoid accumulated errors/artifacts. The idea of psychoacoustic compression is to "throw away" details that you can't hear. If those details are thrown-away during the 1st compression there's no need to throw-away more data during the 2nd or 3rd compression. An analogy would be bit-depth reduction... If you go from 24-bits to 16-bits and back, you loose 8-bits of resolution. Repeating that over & over doesn't do any more damage.

 
However, it turns out MP3 encoding is not that simple or even standardized, so you have various levels of artifacts introduced by different implementations of the codec.
 
Here's an interesting and exhaustive test on reencoding a ~256 kbps file 100 times with various encoders. As you can note, the difference between the Fraunhofer MP3 and Lame MP3 encoders are HUGE!
 
http://bernholdtech.blogspot.de/2013/03/Nine-different-audio-encoders-100-pass-recompression-test.html
 
PS! Press the "100th pass" link under "Detailed results" for the full range of variable bitrate, constant bitrate and so on. Remeber to check both "mp3" and "lame".

 

[james444]

 
Thanks for the link, interesting stuff!
 
Please note that I wasn't talking about consecutive reencodings with the same codec. It's quite likely that MP3 and SBC use entirely different compression algorithms.
 
Still, as I said, it's just a simplified example. There may be further degradation of mp3 through SBC. And FLAC may be preferable to mp3 for that reason.

 

[Soundofmusic]

Yeah, but still the basic idea behind them both is to remove the inaudible frequencies from the data AFAIK, which would suggest that to some degree reencoding doesn't alter the sound that much. Of course in practice it could be another story, which is obvious from the linked tests.
 
As with any engineering question, the correct answer is: "it depends"

 

[Deckardk]

^^^^^

This thread has gotten way off topic for the last few pages, and I for one wish it would come back to "Huge Comparison..."

 

[cehowardNote3]

  You might wanna consider a subscription to Tidal instead, FLAC quality streaming with the added possibility of offline playlists/albums.
Dunno what the US cost is, but I get it for around $20/month in Norway.

Thanks for that tip, it is $20 a month in the USA too.  Going to sign up for the free 30day and see how it goes..

 

[cehowardNote3]

 
Disclaimer: this is a very simplified example to illustrate a point, with no claim of validity whatsoever. It just means to show that while mp3 and FLAC sound equally good for most listeners over a wired connection, there may be a reason to prefer FLAC over mp3 when we're dealing with bluetooth transmission and double compression.

 
Okay, I got most of that, and it seems like proof that my way is working. 
 
On my smartphone, I have a 128gb sd card.. I load it up with lots of flac files.  My smartphone has aptx, and so does all 3 of my  headphones.
 
I like the sound.  According to your explanation, I am getting the best sound I can get across bluetooth.. Great.
 
Also, I noticed the you refereed to 2 different types of mp3 files.  One you say is high quality, and the other, well I don't know.. if they all carry the same "mp3" designation, how can one tell if they have high quality mp3 oppose to ordinary mp3??
 
Yes, you all do have "heavy heads"... 

 
 
EDIT: I think I figured it out. High quality mp3s are the ones that can do 320k??? whereas, the other mp3 do something like 190k???...      

 

[cehowardNote3]

^^^^^

This thread has gotten way off topic for the last few pages, and I for one wish it would come back to "Huge Comparison..."

IMO, it is on topic because we are talking about codecs, like aptx, which affects greatly bluetooth headphones
 
So, if we compare bluetooth headphone A to bluetooth headphone B & C, and two of the headphones have aptx, someone says the headphone without aptx is this or that, than this discussion on aptx, and codecs is valid and on topic..
 
I can stand a correction if I am wrong. After all, I am just a layman.. 

 

 

[n00b2]

  Yeah, but still the basic idea behind them both is to remove the inaudible frequencies from the data AFAIK, which would suggest that to some degree reencoding doesn't alter the sound that much. Of course in practice it could be another story, which is obvious from the linked tests.
 
As with any engineering question, the correct answer is: "it depends"

 
Thing is all codecs that do lossy compression attempt to remove sounds that the codec perceives as inaudible to the human ears. Each codec will remove different things and not everything that gets removed by every single codec is actually inaudible. So quite often even though the codec is attempting to remove inaudible information it actually does improve critical information that has a serious noticeable impact on sound quality (obviously the poorer the mp3 the more actual audible stuff that gets removed). From recent tests aptx uses a superior encoding algorithm and is better than most other codecs in terms of actually removing sound that is insignificant. The link that @Giogio posted earlier said the same thing. Mp3 has improved over the years (and so has SBC with high quality) but both still do remove sounds that can be spotted as missing more often than aptx.
 
@cehowardNote3 To check the bitrate of the mp3 (that is 320kpbs or 192kbps and al the other ones) right click on the mp3 file and select properties then go into the details tab. As or what is high quality its not as simple as 320=high and less=low (256kbps also good) but that is a good rough assumption for a layman. Trying to understand the difference between different encoders and how they affect sound at the same bitrate is another whole topic that is definitely OT for this thread. You will find plenty of threads on this forum if you are after info though.

 

[james444]

  EDIT: I think I figured it out. High quality mp3s are the ones that can do 320k??? whereas, the other mp3 do something like 190k???...      

 
Here's a pretty good explanation of bitrates in mp3 encoding: link

 

[Koolpep]

I would assume you can't treat degration as "x % per reeconding pass", because as you state yourself the codecs are based on removing the parts of the spectrum the human ear can't percieve. I.e. after that information has been removed in the 1st pass, the codec wouldn't do much for the consecutive reencodings. Or to quote someone more linguistically capable in english :


However, it turns out MP3 encoding is not that simple or even standardized, so you have various levels of artifacts introduced by different implementations of the codec.

Here's an interesting and exhaustive test on reencoding a ~256 kbps file 100 times with various encoders. As you can note, the difference between the Fraunhofer MP3 and Lame MP3 encoders are HUGE!

http://bernholdtech.blogspot.de/2013/03/Nine-different-audio-encoders-100-pass-recompression-test.html

PS! Press the "100th pass" link under "Detailed results" for the full range of variable bitrate, constant bitrate and so on. Remeber to check both "mp3" and "lame".

Fraunhofer developed MP3 gave it to the world royalty free and then moved on to develop MP4 (AAC) together with AT&T, Dolby, Sony and Nokia.

Lame though continued to make MP3 better using different encoders that work with the same decoders.

All of which pretty nifty stuff...

 

[cehowardNote3]

   
Here's a pretty good explanation of bitrates in mp3 encoding: link

 
Thanks for the link, and I was on the edge of getting a DAP too. This makes it almost certain I will be getting a DAP soon.. Either the DX90 or the upgraded X5.. 
 
Life is sweet...

 

[god-bluff]

   
Thanks for the link, and I was on the edge of getting a DAP too. This makes it almost certain I will be getting a DAP soon.. Either the DX90 or the upgraded X5.. 
 
Life is sweet...

 I assume by replying to this thread you have an interest in bluetooth phones. Why not get a Sony NWZ-A15/17 with aptX built in ? (Plus 50 hours battery life, small form factor, great user interface, great value, great sound quality........ ....the list goes on & on & on)

 

[god-bluff]

Oh and with the latest firmware update (when did Sony start doing that ?) LDAC support whatever that is !

 

[Giogio]

^^^^^

This thread has gotten way off topic for the last few pages, and I for one wish it would come back to "Huge Comparison..."

Then post a comparison!
I understand what you mean, but all this started, as our layman said, from the need to be able to suggest aptx headphones to people without an APTX phone.
Yeah, it's kind of of topic a bit but useful.
Also because not many people post their own comparisons, which is a pity because this thread is mostly for that

Oh and with the latest firmware update (when did Sony start doing that ?) LDAC support whatever that is !

Oh, interesting, i would have never said that such a thing as ldac (a new codec developed by Sony, which is supposed to be much better than APTX) could be implemented with a firmware update...

Lol @ heavy heads...

As for the bolded part: nope, you've got me wrong. According to my opinion, FLAC is basically useless except maybe across bluetooth!

actually you're saying that FLAC is useless except maybe across sbc, while wired or Bluetooth but with APTX a FLAC is useless, and i and my mp3 files win over grandpa ceholayman and his flacs.
Which is the only real secret reason why i wanted to talk about this.