I’m often asked by friends and colleagues about high resolution audio, or 24 bit audio. They ask my opinion simply because they know that I like music, both playing and listening and presumably decide that I must have an opinion on it. And I do have opinions.
I decided to try and write up my thoughts in a blog post to point people to, not because I don’t like the questions, but more because I found myself typing the same things over, or at least searching for the last reply to cut and paste.
To set things up, I’ll declare that I am not what might be described as an audiophile, though many would think that I am. I like my music to sound good for sure, but I don’t go to the lengths that some people do, with thousand dollar power cords, and other ridiculous things. My equipment is very good, it suits me fine, but it’s far from top of the range. I can listen to MP3 files without being snobbish, but I have spent quite a while compiling my music library with uncompressed files in a variety of formats to suit various players.
So for years, I have followed the rise of “high resolution audio”, sometimes referred to as 24bit audio (though that’s only one version of it). There are now multiple vendors of these music tracks and it’s only a matter of time before the mainstream stores like iTunes start to vend them. Sites like HDTracks offer a vast variety of genres to purchase at a higher cost than standard sites, or physical CDs for that matter.
So to the crux of the question. Is it worth buying a high res album? It’s higher resolution and usually more expensive, so it must be better surely? Well yes. And no. This is not a definitive science by any means, although quite frankly it should be – we’re dealing with bits of information that can be examine “under a microscope” so to speak, so it shouldn’t really be subjective. But it actually is. Here’s why. In my opinion.
The resolution debate
You can read about this in literally hundreds of articles online, some of which come close to the truth, many don’t. Very few speak in terms that the average person can understand. A CD comes at a resolution of 16 bits and high resolution files are typically, but not always 24 bits. At this point, the wheels often start to come off the description. What does this resolution of 16 bits mean? It simply means that at any point in the music track, any one point in time, the waveform has a discrete value and in the case of a CD it can be any one of 16 thousand values. I’m sure that everyone has seen a waveform from a music file, it’s a great big jumble of squiggly lines with peaks and troughs – the peaks being the loud bits and the troughs, the quiet bits. But if you zoom in and zoom in again, in fact keep on zooming, eventually those squiggles become one line – however complicated the piece of music, however many instruments are playing. That line represents the totality of what you hear. And at any one point in time, you can measure how high that line is and give it a number (one of those 16 thousand numbers). By the way, it’s not exactly 16 thousand, I’m trying to keep things simple.
Figure 1 – How we usually see stereo music represented
Figure 2 – What you see if you zoom in far enough
So invariably people compare this to chopping up the waveform into these chunks. Frequently a digital photo analogy creeps in at this point, because they are also described in “bits” and you’ve also seen zoomed in digital photos that break up into squares – you can actually see the “bits”. And people know that if you have a higher resolution photograph (taken with a better camera for instance), the detail is better because you’re chopping any given part of the picture into ever smaller chunks at higher resolution. So that must be the same for audio files too right? Because they’re digital too…
No. Stop the analogy right there. You are not comparing like for like. This is the simplest way I can explain why it’s not true. In a photo, more resolution does give you better resolution because you’re chopping up the same area (say a 1cm square) into smaller chunks. In audio, when you move from 16 bit to 24 bit, you’re not chopping up the same piece of audio loudness into smaller chunks, the chunks are the same size, you’re just adding more chunks on top of the ones you already had. You’re adding 8000 to the 16000 you already had to make 24000. Remembering what I said earlier that the higher the audio line, the louder the file? Well that extra resolution allows you to capture more “loudness”, not more detail. And this is where people just glaze over. Because they equate loudness to how far they’ve got their volume control cranked to. “Why do I need more loudness in my audio file? I just turn the knob up!”
Indeed you do, this isn’t about the playback volume (which is indeed under your control) this is about the recording volume in the studio. The extra bits are essential in being able to differentiate clearer between the intended sound (the instrument or voice) and the unintended sound (background noise). If your piece of music has loud parts and soft parts, you absolutely benefit from the extra space that 24 bits provide, if your preferred listening is hard rock, not so much…
Sounds Great, What’s the Problem?
The problem, put simply, is that this is not retrospective. It applies now in all recording studios (just about). But it didn’t apply in the 60s or 70s or 80s. Back then, everything was recorded to tape, which even 16 bit CD resolution more than covers. In fact, what you got to hear on vinyl wasn’t even from the original tape that the instrument was recorded onto, it would have been a 2nd, 3rd or even 4th generation copy (to allow for mixing, copying, safety etc.). The range of loudness that a tape could capture, even the highest quality recorder with a high quality new tape, can be completely represented within those 16,000 bits (you don’t actually need “that” many!)
So when your favorite high resolution audio site tells you that your favorite album from the 70s is now available to purchase at 24 bit high resolution (at high cost) enabling you to hear all the nuances of the original instruments because of the extra resolution, it’s just nonsense. You cannot manufacture resolution that wasn’t there in the original tape, it’s just not technically possible. You can “upsample” to that higher resolution, which is what they do, but it doesn’t make it any different to the original.
So I shouldn’t be fooled and buy high res right?
Well not so fast. There might very well be a reason. But it’s not the reason they’re telling you, or trying to make you believe. “This” is the reason why you might like to spend your money. When the transfer takes place, it’s done carefully, using high quality equipment. It is often accompanied by additional audio treatment, such as remixing. When the first generation of CDs were made of those great old albums, the audio to digital conversions weren’t necessarily great quality. So today, the equipment used is far superior and often the remixing done on today’s audio equipment uses digital sound processing that we could have only dreamed about 10-15 years ago. So if you can hear a difference between a track from a 10 year old CD and a new high resolution audio file, and you like that difference, then go ahead and buy it and enjoy it. Just do so understanding that you’re enjoying today’s mixing and processing, not the fact that it’s 24 bit. That could have just as easily have been represented in a 16 bit audio file on a standard CD.
So what do I do?
So do I have any 24 bit audio files in my music collection? Yes, I do have some, but not many. When I hear that an old recording is being re-released having been re-mastered, I will certainly buy a CD rather than a high res download, the problem is that many are only released as high res. If the album has high dynamic range (lots of loud and quiet bits) I’ll probably keep it at 24 bit, but if it’s a rock album, I’ll just downsample it to 16 bit CD resolution to save the disc space.
I do love the fact that this new generation of high res audio is making people interested in good uncompressed music – I think it’s spurred many remixes that otherwise wouldn’t have happened (such as Steven Wilson’s series of remixes for classic albums). I just wish that they were sold for what they are, not what they could be.
New recordings have the potential to sound amazing with modern recording techniques, because the nuances of loud and soft can be perfectly captured and passed on to listeners. Sadly, most modern music is mixed and mastered in a way that completely obliterates these nuances, but some artists are using it to great effect. In fact some artists are taking the brave step of choosing not to master their tracks at all, they’re saying that “this is my mix, leave it alone” – fantastic, but rare.
And don’t get me started on vinyl!