Bluetooth is just Bluetooth, right? All we’re talking about is sending a digital audio signal from one device to another; so surely that should mean that they’re all the same. I mean, it’s all binary, just 1s and 0s…..why would the source even matter?
Well, yes and no. To decipher this little conundrum, we’ll need to delve a bit into how Bluetooth works. But don’t worry, it’s not quite as complicated to grasp as you might initially think.
Ok, so in order to understand the Bluetooth side of things, let’s just first touch on how digital audio goes from being a bunch of 1s and 0s to a fully analogue wave pounding away at your eardrums.
A digital file is comprised of binary data. Binary, by definition, is something that is composed of or involves 2 things. In the case of digital data, this is a bunch of 1s and 0s. Well, to be more specific, it’s not the physical 1 and 0 digits, but instead a series of on and off signals. Think of it almost like Morse code. Instead, where Morse code is made up of long and short audible tones, binary data is simple composed of electrical energy.
A certain combination of on and off signals (1s and 0s) will correlate to a specific type of data, just like a certain combination of on and off tones or clicks will correlate to certain information (alphabetical letter) in Morse code. This is where the codecs (FLAC, MP3, ALAC, WAV, etc) comes into play, as essentially any stream of data is completely useless unless you have the instructions (codec) to figure out what that stream of data actually means. Of course this doesn’t just apply to digital audio files, but indeed to ANY type of digital file.
In the end, different codecs are created to serve different purposes, but all of them are created in an effort to come up with the most efficient means of storing data.
With respect to audio, this involves converting a file into the smallest size possible, but in such a way that once it’s converted back into its original analogue form there would’ve been none or very little information lost. If data is lost, it’s called a “lossy” codec, whereas a “lossless” codec retains as much info as possible. Because lossy files don’t contain all of the data, they usually take up much, much less storage space. There are various manners of converting and compressing data, each with their own benefits and drawbacks, and there are some incredibly clever people coming up with more and more inventive ways of achieving this. The exact details of these various methods is irrelevant to this article, so we won’t be going into it. All that’s important here is that you get the gist of what a digital audio file is.
The other part of the equation to take into account is that a larger file requires faster processing. Think of it like this; let’s say a song is exactly 3 minutes long, and let’s say in MP3 format the song takes up 4MB of space, and in FLAC format it takes up 16MB of space. Well, seeing as the song is 3 minutes long, that would mean that there would be 4 times more data that has to be processed for the FLAC file in the same amount of time as the MP3 file. This is what’s referred to as the bitrate – it’s the amount of data (bits) that has to be processed every second.
This bitrate is definitely a limiting factor when we’re talking about Bluetooth audio, but we’ll get into that soon enough.
Getting to Bluetooth
Just like WiFi, Bluetooth is a type of radio signal and it’s intended purpose is to transmit data wirelessly and in digital form (not analogue). As such, it’s got its limitations. The biggest limitations here is the range that the signal can be transmitted, and the amount of data that it can process per second (bitrate). For the purpose of this discussion, we’ll focus on the bitrate.
Here, there are Bluetooth-specific codecs used to transmit the data from one device to another. The default is called SBC, then there’s also various AptX codecs, and also Apple’s own AAC.
Each codec serves a slightly different purpose. SBC, for example, is the default codec used, so any and all Bluetooth devices can communicate with one another via that codec. Then there’s AptX Low Latency, which is intended to be used with video and gaming to ensure that the audio stream and video stream remain synchronized with one another.
For personal audio, let’s focus on SBC and AptX.
As mentioned, SBC is intended to be universal, and as such, it’s got some very low-level requirements. The SBC codec is able to transmit audio at a maximum bitrate of up to 328kbps. That’s slightly higher than the best quality MP3 file. Although, it should be noted that 320kbps MP3s still do sound better, as bitrate is NOT the only variable that affects audio quality.
The AptX codecs attempt to step things up, with AptX HD getting as close as possible to being lossless (at least that’s the idea). And there in lies the problem with Bluetooth audio…ALL of it is lossy, at least to some degree.
To illustrate the effect of a lossy codec and in particular how it applies to Bluetooth, let’s think of it in terms of money.
Let’s say you give me $100, and I give you back $95. Here, you’ve “lost” 5%. Now, let’s say you give me that same $95, and then I give you back $90.25, which means you’ve yet again “lost” another 5%. In total though, you’ve lost almost 10% of what you had originally. So, if you store your music files in, for example, MP3 format, then you’re already beginning with a lossy file. Then when you transmit that audio file over Bluetooth, an already lossy file is then again converted into yet another lossy file.
So, when we talk about audio fidelity (how true something is to the original), hopefully you can see why it would be important to start with the best quality file, and to try and utilise the best devices and technology in the transmission process too.
The thing is, whilst all Bluetooth audio devices will support the SBC codec, not all of them support any of the AptX codecs, and in order to take advantage of AptX’s audio superiority, both the transmitting device (source) and receiver MUST support the AptX codec. If either device doesn’t support AptX, then it’ll default back to SBC for both devices.
So that’s the first scenario in which the source matters, but the receiving device is equally important.
The Human Factor
Ok, now that we’ve got a grasp of what a digital file is, the next step is to understand how we interact with it.
We don’t hear sound in a digital form, we hear it in an analogue form. As mentioned, digital is binary ( a total of 2 possible values), but analogue has an infinite amount of values. Where digital can be represented as 1s and 0s, an analogue wave can be represented as any and all values between and including 1 and 0 (i.e 0, 1, 0.5, 0.9, 0.99999999, 0.123467383598346593465973465, etc). So, at some point the original sound was of course in analogue form, then it was recorded and stored into some kind of digital form, and then when we want to listen to that file again it would need to be converted back from the digital form into an analogue form again. This is usually done by an audio-specific processor called a Digital to Analogue Converter (DAC). And yes, when the original audio is captured an Analogue to Digital Converter (ADC) is used for that.
Hopefully you’ll be able to see just how intricate this entire process is, as there are so many variables at play here.
Follow the chain
The recording equipment (microphones, ADCs, software, etc), the digital codec, the playback software, DAC, amplifier, headphones/speakers, etc…ALL of it will affect the final sound.
What this means is that, when you send an audio signal over Bluetooth, the receiving device MUST have some sort of DAC chip in order to convert the data stream from a digital format into an analogue waveform that your brain can understand. As such, the quality of the audio signal that you get out of the headphones/speakers will also be directly affected by the quality and efficiency of the DAC chip built into the receiving device. So what does this have to do with the source? Well, not much actually, but it’s just something to keep in mind.
Where the source does once again come into play is when we consider Digital Signal Processing (DSP). DSP is whenever the original digital stream of data is processed and altered into a slightly different stream of data, usually when applying various effects. This can include effects to expand the perceived stereo image (how “wide the music sounds to you”, or reverb effects to make the music sound like it’s being played in a large empty room, applying a digital equalizer, etc. There really are A LOT of creative people out there who are constantly coming up with even more creative ways to alter the sound into subjectively more pleasing ways. But, the fact remains that if you change the data, you’re moving further away from what the original sounded like, and hence you are technically reducing the “quality”.
So, if your source device does apply any sort of DSP to the data before passing the stream onto the Bluetooth chip, then that will have a direct affect on the final sound you’ll hear too.
And that’s pretty much the gist of it. It’s easy to assume that the source device is simply sending a Bluetooth signal without actually affecting the original file, but in reality, not all sources are created equally.
Then the last factor to take into account is that, if the source device’s Bluetooth chip is sub-par, then the maximum range would be compromised, resulting in audible skips and other hiccups as you approach the limits of that transmission range.