[DoctaCosmos]

Thanks for your feedback gents. clearly I've got the wrong idea or at best I've over-simplified things. I picture sound as waveform that has to be digitally represented by "slices" and the more slices, the closer we get to reproducing that waveform more accurately.
So without turning this into a futile attempt to convert me into a Sound-Science groupie  what level of sampling and resolution does it take to avoid audible distortion / noise (quantization) while representing the full audible range along with what some audiophiles denote as less tangible qualities like the "air" around the instruments etc.? I'm just asking for a set of numbers and not the math behind the Nyquist-Shannon theorem 

 

http://www.head-fi.org/t/415361/24bit-vs-16bit-the-myth-exploded
Could help. There are also some charts that give you a good visualization of what's going on but can't recall what pages they're on.

 

[watchnerd]

 
So without turning this into a futile attempt to convert me into a Sound-Science groupie 

what level of sampling and resolution does it take to avoid audible distortion / noise (quantization) while representing the full audible range along with what some audiophiles denote as less tangible qualities like the "air" around the instruments etc.? 

 
For playback, 16bit is more than enough to get everything you need for music.
 
All the audiophile attributes you're talking about have nothing to do with bit depth -- they're due to recording / mixing / mastering differences.

 

[Traveller]

... All the audiophile attributes you're talking about have nothing to do with bit depth -- they're due to recording / mixing / mastering differences.

I thought the topic was hi-res sources and their (de)merit...  if a 24b96K file would preserve more of the original source, wouldn't we want to try to obtain gear that can play it back to us.
 
But I think all of you have made yourselves clear that you are confident >16bit is all but useless.

 

[StanD]

  I thought the topic was hi-res sources and their (de)merit...  if a 24b96K file would preserve more of the original source, wouldn't we want to try to obtain gear that can play it back to us.
 
But I think all of you have made yourselves clear that you are confident >16bit is all but useless.

If all of that resolution cannot produce any humanly detectable results then why waste the resources? As I've stated in the past, go and find recordings that you want to listen to that actually have a DR that exceeds the ability of 16 bit resolution. Once you find that, let us know what they are and how many you found. I suspect that you won't be able to scare up anything worthy of a mention. There are sites that sell HR recordings, however, they have nothing with exceptional musical DR. A better mix or mastering is what one needs, hirez doesn't make the difference.

 

[castleofargh]

if that's clear enough, can we go back to aptx?

 

[watchnerd]

 
But I think all of you have made yourselves clear that you are confident >16bit is all but useless.

 
Yes, that is what we're saying.
 
I know it's hard to digest because it flies in the face of all the people trying to sell high resolution digital, but it's the truth.  My audio life became so much easier when I realized regular lossless was fine.

 

[Traveller]

...if that's clear enough, can we go back to aptx?

Yes... back to AptX.
AptX and AptX HD...

 
 
EDIT:
I *think* I have found at least one possible use for AptX HD. It will help to avoid down-sampling of 24-bit media. While not the main point of the thread below, this comment was an eye-opener (nothing new, just a reminder of the potential pitfalls of working with "Hi-res" media):
 

... Using 24bit or even using an upsampling DAC and going from 16bit to 24bit is not likely to be detrimental, it's just that there won't be any benefit either.  
Going the other way, 24bit to 16bit could be quite detrimental unless a good quality dither is used as part of the process. Most consumer programs will truncate when going from 24 to 16bit. In other words, the last 8bits are just hacked off. Truncation is not good, it introduces quantisation distortion which is correlated to the audio material and it's results are unpredictable. It could mean that you get unwanted tones or harmonics in the mix which may be noticeable. Some consumer programs 'round' the result, still not good but better than truncation. The effects of rounding are unlikely to be heard by most people but the chances are that some audiophiles would notice. Dither is the only real option if you are serious about SQ.

 
Foobar is good enough to offer a dithering option for down-sampling and I sure as hell hope most TOTL DAPs will also take the high road but what of the average Smartphone...

 

[a-LeXx]

After many years of being satisfied with my current rig I decided it's time to get somerhing new for mobile use.
This time wireless.

I had a very limited experience with BT prior to that, so, believing the advertisment, i thought BT with just SBC would sound just 'ok' and I'd need aptX for sure.

Now, with my new Momentum Wireless, the experience is very different from my expectations, and surprized by what I'm hearing I digged into the technical implementarion of SBC and aptX and came to a conclusion that aptX is only a marketing thingy. Ok, marketing and low latency.

I compared my Momentum with an iOS device (no aptX, Momentum does not support AAC) and a sony Z3(aptX) as sources. The difference in sound quality is about 0.

Now to the technical explanation, why the sound quality is the same:

SBC is a subband codec that works with 4 or 8 subbands.
aptX is a subband codec that works with 4 subbands, around 5kHz per subband. Means, SBC and aptX can be identical in this regards, if SBC is configured to divide the signal into 4 subbands.

aptX is using a fixed bit allocation for each band (in the BT implementation) following the pattern of 8:4:2:2 for each subband. It has a fixed compression ratio of 4:1
SBC is using a flexible bit allocation, that is, as the name suggests, flexible. It can be the same 8:4:2:2, or e.g. 8:3:3:2, depending on a signal it's fed. It can be also less bits per band, which means, for BT, SBC can be identical to aptX and have a compression rate of 4:1, or have less bits per band and a higher compression.

Both SBC and aptX use kind of ADPCM for encoding of the subbands and no psychoaccustics, SBC usually cuts everything above 20kHz with a hard band-pass filter.

Conclusion: aptX in the BT implementatin has fixed parameters and always delivers the same quality. SBC however has some flexibility and hence can sound exactly the same as aptX, but can also be used for higher compression with quality loss. Theoretically, SBC can even deliver better quality then aptX, because it has adaptive bit allocation per subband, and hence, if the source signal calls for it, can vary amount of bits used per frequency subband, while aptX in the BT implementation will alway allocate e.g. 8 bits for the lowest 5kHz and only 2 bits for the highest 5kHz (15-20kHz band).

This means, aptX is less a 'better' technology, but a bit more like a 'THX' logo for BT - you always know, what you are getting, it's a 4:1 compression, period.
SBC can have identical quality, but it depends on an actual implementation and bit rate, with lower bit rate and higher compression the quality will suffer.

I guss that most reputable manufacturers nowadays would use the full bandwidth available for SBC, that means, SBC and aptX will produce the same sound quality, since the technical implementation of both compression standards is almost identical.

 

[Traveller]

...This means, aptX is less a 'better' technology, but a bit more like a 'THX' logo for BT - you always know, what you are getting, it's a 4:1 compression, period.
SBC can have identical quality, but it depends on an actual implementation and bit rate, with lower bit rate and high compression the quality will suffer. I guess that most reputable manufacturers nowadays would use the full bandwidth available for SBC, that means, SBC and aptX will produce the same sound quality, since the technical implementation of both compression standards is almost identical.

Very interesting indeed, T4S!

 
Does anyone know of a way to force a source device (smartphone, notebook...) to use one or the other codec? I have a BT dongle that supports all the common codecs (SBC, AptX & AAC). I'd love to be able to compare codecs using the same source

 

[james444]

After many years of being satisfied with my current rig I decided it's time to get somerhing new for mobile use.
This time wireless.

I had a very limited experience with BT prior to that, so, believing the advertisment, i thought BT with just SBC would sound just 'ok' and I'd need aptX for sure.

Now, with my new Momentum Wireless, the experience is very different from my expectations, and surprized by what I'm hearing I digged into the technical implementarion of SBC and aptX and came to a conclusion that aptX is only a marketing thingy. Ok, marketing and low latency.

I compared my Momentum with an iOS device (no aptX, Momentum does not support AAC) and a sony Z3(aptX) as sources. The difference in sound quality is about 0.

Now to the technical explanation, why the sound quality is the same:

SBC is a subband codec that works with 4 or 8 subbands.
aptX is a subband codec that works with 4 subbands, around 5kHz per subband. Means, SBC and aptX can be identical in this regards, if SBC is configured to divide the signal into 4 subbands.

aptX is using a fixed bit allocation for each band (in the BT implementation) following the pattern of 8:4:2:2 for each subband. It has a fixed compression ratio of 4:1
SBC is using a flexible bit allocation, that is, as the name suggests, flexible. It can be the same 8:4:2:2, or e.g. 8:3:3:2, depending on a signal it's fed. It can be also less bits per band, which means, for BT, SBC can be identical to aptX and have a compression rate of 4:1, or have less bits per band and a higher compression.

Both SBC and aptX use kind of ADPCM for encoding of the subbands and no psychoaccustics, SBC usually cuts everything above 20kHz with a hard band-pass filter.

Conclusion: aptX in the BT implementatin has fixed parameters and always delivers the same quality. SBC however has some flexibility and hence can sound exactly the same as aptX, but can also be used for higher compression with quality loss. Theoretically, SBC can even deliver better quality then aptX, because it has adaptive bit allocation per subband, and hence, if the source signal calls for it, can vary amount of bits used per frequency subband, while aptX in the BT implementation will alway allocate e.g. 8 bits for the lowest 5kHz and only 2 bits for the highest 5kHz (15-20kHz band).

This means, aptX is less a 'better' technology, but a bit more like a 'THX' logo for BT - you always know, what you are getting, it's a 4:1 compression, period.
SBC can have identical quality, but it depends on an actual implementation and bit rate, with lower bit rate and higher compression the quality will suffer.

I guss that most reputable manufacturers nowadays would use the full bandwidth available for SBC, that means, SBC and aptX will produce the same sound quality, since the technical implementation of both compression standards is almost identical.

 
Great, informative rundown! That said, the devil may be in the details... for instance, ADPCM encoding, for which aptX claims "additional techniques for accurate linear prediction and inverse quantization". And robustness in transmission, for which aptX claims superiority of sample-based encoding over SBC's frame-based approach (which may force the latter to lower bitrate if too many frames are dropped).
 
Anyway, that's merely citing from and speculating based on what little in-depth information seems to be out there...
 
http://www.edn.com/design/consumer/4326740/Bluetooth-Sufficient-fidelity-even-for-average-listeners-
https://www.design-reuse.com/articles/15836/audio-coding-for-wireless-applications.html
https://www.lifewire.com/what-is-aptx-3134709
http://www.sereneaudio.com/blog/how-good-is-bluetooth-audio-at-its-best
 
Subjectively, I'll say that while I tend to prefer aptX in general, the best SBC implementations I've heard sounded better than the worst aptX implementations.

 

[castleofargh]

Subjectively, I'll say that while I tend to prefer aptX in general, the best SBC implementations I've heard sounded better than the worst aptX implementations.

my experience too. even though I don't have any mean to be sure I was playing aptx. if the streaming encounters too much difficulties does it revert to SBC by default and then proceeds to lower the rate again and again until we're good? or do the devices stick to one protocol once they successfully started with it?
all I know is that bad connection(distance, weak source...) never sounds good. plus the sound drops when it's really too bad.

 

[StanD]

I suspect that if  the two devices are in close proximity then one might get a quality SBC SQ.

 

[Traveller]

I gotta try...

 

 

 

[REXNFX]

  I gotta try...

 

 

Hows the sound v Usb?

 

[quasides]

i want do add 2 points to the discussion, dunno if it was already said since the threat is already aweful long so i say it again

number one. aptx doesnt only give you a supposed better quality it does give you defently better latency.
you may wonder why this matters in playback.. well play a video and you know what i mean

segular bluetooth audio have a horrible latency, resulting in the old tale bt audio and movies or games are impossible to mix and this is true to an extent.
software made progress, so in android youtube you wont see a difference because the software already take this into account and play video a little bit after the audio

however not all software is that forgiving, since i use aptx only i cant tell which apps today still lagging but i remeber well that netflix and vlc mediaplayer where uneuseable over sbc at all, let alone games

1 second lag is simply too much. which aptx its reduced to a few miliseconds, low enough toactually enjoy that content without going insane


second, if you think aptx makes a hearable difference i have something for you even better.
viper4android - its by far the best audio processing software on mobile you can find
with features you wont find in any other software solution in mobile (or even regular windows audio driver stack if you dont usw a daw and asio)

before i relist all its features just google it and make up your own mind


however i have to warn you. to get it running is not a piece of cake but its worth it
deping on ur rom you need to flash it seperatly and in some cases even edit a file or switch selinux to permissive which require root on ur device
theres plenty of documnetation on the web how to implement it on which manufacturer but not aleays easy to find
best eay togo for this is the xda forums


however u may think ahh too much trouble, well its really worth it
it doesnt matter if ur equipment is a piece of crap or highend, it even includes sound profiles of over 500 headphones, among other dsp stuff
ofc it can lift your output gain in fact its the only solution on android that works anymore to make your audio louder (thanks to various regulations its level down to an inacceptable level specially for better headphones)

have fun