What’s The Deal With 24/96?
If you’ve been involved in the world of electronic music production for a while, you’ve probably heard about a bunch of aspects of sound quality and mixing that you’re not too familiar with. There’s nothing wrong with that – the world of audio is vast and can get extremely complicated, so we all still have plenty to learn!
In this article we’re going to take a look at that combination of numbers we tend to see all over the place, “24/96,” and learn why these numbers are significant enough to many music producers that they’ve become a common buzz-phrase. After that, we’ll see whether the sound quality they refer to is really worth all the hype.
Keep in mind that there’s really no easy way to explain this, as it gets deep into the science and mathematics of sound, and can be extremely confusing if you’ve never learned about it before. Despite this complexity, however, I will do my best to describe it all in simple terms.
The numbers 24/96 refer to two different aspects of audio quality: bit depth for the number 24, and sample rate for 96.
This refers to the resolution of audio being recorded or played back. It’s the range in which your sound is captured, and the amount of bits available to pick up each aspect of a sound wave.
Be careful that you don’t confuse bit depth with bit rate, which is completely different: bit rate simply describes the amount of data being transferred per second during playback, and this refers to MP3s and other such audio files. So the higher the bit rate, the better the quality of the file.
Bit depth is measured in binary digits, or “bits”, and each bit adds more possible levels to what the recording can pick up, reaching higher frequencies as you increase them. A bit contains a binary value, so it’s either a 0 or a 1.
Not only does each bit add more levels, it doubles the number of possible levels available with each successive bit included. So by the time you reach the standard 16 bits you already have 65,536 possible levels. A 24-bit recording has 16,777,216 possible levels.
Bit depth is responsible for two aspects of sound: the range, or resolution available to pick up or play back a recording, and the potential volume level and dynamic range that a recording can reach.
The range is very important, because the deeper the bit depth the more thorough it becomes, and therefore the more frequencies you’re able to capture or reproduce. If you’re recording a live band, for example, 24 bits will allow you to pick up many more of the subtleties of each musician’s performance – certain instruments will be more easily distinguished from one another, such as the bass and the kick drum, or the lead guitar and the cymbals – and therefore gives you a much more balanced representation of the real-life audio than a lower bit rate would. Recording at 16 bits or lower, you’re starting to significantly drop the amount of frequencies you can capture, so your range is more narrow.
The volume level refers to the dynamic range (distinction between loudest and quietest sounds) that can be reached. The wider the dynamic range, the louder the recording will be able to be played back. Additionally, a higher dynamic range can drastically reduce what’s called the noise floor, which is the interference level picked up in any recording, or that ever-present background hissing and humming we hear when recording with microphones. So if you use a higher bit depth the noise floor won’t be as noticeable, and it will be mostly cleared out by all the other sounds inhabiting your dynamic range – and that’s before you apply any compressors, EQs or other effects.
So as you can see, a 24-bit resolution is absolutely beneficial, and it won’t even cause your computer to work any harder when reproducing it!
When you record and reproduce audio in a digital environment, the computer takes bits and pieces of your sound wave, or samples, and plays them back in a certain amount every second. This is the sample rate, and the more samples per second the computer can process, the better the quality of your recording and playback.
For CDs, the standard sample rate is 44.1, or 44,100 equally-spaced samples per second, and this resolution plays back audio frequencies to a maximum of 20,500 hertz, which is as far as human hearing can go – although the average human can only really hear up to 16,000 kHz. The reason why this sample rate is considered ideal and why it’s been used professionally for so long, even now that we have much higher sample rates to work with, is due to what’s called the Nyquist Theorem.
This theorem is also called the Nyquist-Shannon Sampling Theorem because in 1949, a man named Claude Shannon was the first to provide formal proof of Harry Nyquist’s theory, which Nyquist first presented in a paper to the American Institute of Electrical Engineers in 1928.
The theorem gets very technical, but the basic principle is that in order to accurately digitize an analog signal, the sampling rate has to be at least twice as high as the highest analog frequency. So if human ears can perceive sound up to 20,500 hertz, 20,5000 x 2 is 44,100 – or 44.1 kHz. That’s why 44.1 is considered the ideal sample rate, because it allows us to capture all of the frequencies the human ear can perceive.
According to the theorem, if the sampling rate of the analog recording is less than two times the highest frequency, then the sound won’t be accurately represented upon playback; in this case the digital converter – or rather, the computer – will attempt to make up for these lost frequencies, which can result in a scratchy, distorted texture. This phenomenon is called aliasing.
This is why some people prefer to work with 48 kHz, because by doing so you ensure that no aliasing can take place in your audio, since the frequency “roof” has been raised enough to where everything will be captured and accurately represented. But even so, CD format has always been 44.1 kHz and is likely to stay that way.
So now that we know what the numbers mean, we can finally ask the really important question.
Is 24/96 really worth it?
In terms of bit rate we can see that 24 bits is certainly preferable, since it widens your dynamic range and provides a deeper resolution. Each of the main points of a mix – low, mid, high – will be much more clearly defined, meaning you’ll have more room for deeper bass or tinnier cymbals, and there will be less interference between sounds in your mix. Many producers out there agree with that.
When it comes to sampling rate, however, things start to get completely subjective. There may be evidence that analog sound is superior to digital in terms of quality, and you’ll surely find plenty of people who say that listening to music on vinyl is much better than any digital file type you can find. This might be because analog playback and recording gear can reproduce sound at a maximum of 50 kHz, which is two times what humans can perceive; the idea being that, even if we can’t hear them, the frequencies being played back in that higher range have a direct effect on the sounds that we can hear in the lower frequencies.
So in terms of the Nyquist Theorem, if you use a sample rate of 96 kHz then you can record and reproduce frequencies as high as 48 kHz, which is close to the maximum that analog gear can play back. So if it’s true that analog gear is superior because of this higher frequency capability, then a 96 kHz sample rate will allow you to access that superior quality.
Many people don’t see the point in this, though, and in terms of music, their stance tends to be that your music will end up at 16 bit/44.1 kHz when put on CD or vinyl anyway, so there’s no need to work with higher quality. But this is becoming an outdated concept since so much of the music we create, share and sell is digital and we may not choose to release our work on CD or vinyl, so we’re no longer confined to working with lower frequencies.
Even if you did opt to release your music on CD or vinyl, you have to remember that if you mixed your tracks in your DAW at 24/96 then you were working with a wider range, which means your mix could be a lot more thorough and clean – so when you do convert it down to 16/44.1, the quality might be a bit reduced, but it’s still a more detailed mix because you had more room to work with originally. To me, that’s the most important factor.
The downside to higher sampling rate is that the higher it is, the more strain on your computer’s CPU you’ll be applying. That leads to skipping audio and latency problems. So if you want to work with a higher sample rate then make sure your equipment is ready to handle it!
What all of this really boils down to is two things: preference, through trial and error to determine which bit depth and sample rate sound better to you, and talent, meaning your ability to create great mixes and bangin’ tracks, regardless of quality. Higher quality might allow you to make cleaner and punchier mixes, but if you’re not making great music, then quality means absolutely nothing. There is so much excellent music out there that is recorded with sub-par quality – whether it’s punk, indie hip hop, black metal, dark ambient, or anything else – and it’s still held in very high regard, because the music is good enough that the quality not only doesn’t become an issue, it becomes an intriguing aspect of what makes the art unique.
So if you’re thinking about working with higher quality, I would urge you to give it a try. If you can’t hear the difference right away, then stick to 16/44.1 for now, and once your listening skills become a bit more advanced you may be able to hear it eventually.
Don’t just listen to what others are saying: be hands on, see for yourself, and choose what’s best for you!