Today I wanted to cover a topic that isn’t really thought about or talked about much. The concept of bit depth vs. sample rate.
Sometimes when comparing audio interfaces, we come across these numbers that are thrown around all willy nilly, but in reality don’t mean much. We like to think that we have an understanding of what specifications are better than another, and we buy based on that limited knowledge.
That said, there has become a standard for entry level audio interfaces that you will find is very common among many of the recommended choices.
A bit depth of 24
sample rate of 96 kHz
These numbers will never be the same for every interface. There are some that do have similar numbers, but when you get into the higher end equipment there will be more of a discrepancy.
By now you’re probably wondering..
What the heck does it all mean Basil!
Glad you asked..
Just click on this button here which will direct you to a page where you can input your credit card infor… JUST KIDDING!! 😀 😛
Ok let’s get right down to it.
As we discussed in the audio interface portion of building your home studio, an interfaces primary function is to send and receive data. It also converts analog data into digital (and vice versa), so your computer (and you) can understand what the heck is going on. When you scream obscenities into your microphone, your computer becomes sad and offended, but that doesn’t distract it from it’s primary goal.
The goal is to make sense of the information (your voice) in the form of 1’s and 0’s. Your computer prefers to do this mathematically, by describing it in terms of individually separate values.
Think of it like prioritizing time and writing a list. The tasks are separate and must be done on their own.
Your A/D converter or DAC (inside your interface), captures and quantizes these values (samples) at a fixed rate, and of a specified size. Each piece of data contains parameters that are used in accurately reproducing the original sound. The way your computer does this is by copying the original values, and then playing them back in the same order and rate at which they were captured.
This in theory produces an exact replica of the sound you just recorded.
Mind blown right?!
So in terms of the numbers mentioned above, the sample rate, 96 kHz, is the rate of capture. The bit depth, 24, is the number of bits used to describe each sample. Also, the number of bits transmitted per second is bit rate.
Computers operate in Binary
As stated earlier, a computer understands language in terms of 1’s and 0’s. This is called binary language, and a string of binary digits is thus used to describe any sort of information that a computer receives. These digits are formed from bits, and can range anywhere from 8, to 16, to 24 and so on.
As far as bit depth goes, the higher the better. Working with a high number ensures that the wavelengths are reproduced accurately and precisely. This means that more subtle fluctuations in the sound can be measured, contributing to the overall quality of it’s recreation. The lower the bit depth, the more information gets lost, and thus the crappier the sound.
For clarity’s sake, a 24 bit resolution can have up to 16,777,216 unique values, and a 16 bit can only contain 65,536. It’s a huge deal, and can make all the difference when recording or playing back music.
In layman’s terms, if you are recording at 96 kHz, the highest frequency will only top off at 48kHz. The Nyquist-Shannon sampling theorem says that the upper magnitude of a piece of digital audio will top out at half the sample rate. In theory, a typical human with great hearing can only hear at about 20 kHz.
The standard for interfaces used to be 40kHz, until it was discovered that humans can actually hear frequencies above and below that. The standard in recent years has gradually become 24bit/96kHz, which is now what is considered “high resolution”. In the future these numbers can and probably will change.
Even if your files are only 44.1kHz, it’s not really a big deal. The differences above 96kHz are pretty much negligible and 99% of the time 44.1 is completely fine. Remember, there are very few people that can even hear above 20kHz.
What does this mean?
It means that these numbers are a little overrated Basil.
There is actually a theory that says that the audio we don’t hear has an effect on what we do hear. So nerds, audiophiles, and geeks have since come to the conclusion that the audio is improved in a way that we cannot perceive.
What is more important than numbers?
Glad you asked. In reality you shouldn’t stress over the sample rate at all really. The quality of your DAC (Digital to Analog Converter) is much more important, and will make all the difference in how good the music sounds.
With a good DAC, 44.1 is about as good as it gets!
The main reason that higher sample rates sound better is because of poorly implemented anti-aliasing filters. This is because they contain artifacts that show up inside the range of human hearing.
If I were to use a higher quality source file (say 24 bit/96kHz) with a crappy DAC like my Laptops internal Soundcard, it would sound better, because those artifacts get filtered out due to the fact that they will only show up outside of the range of human hearing. What is a Soundcard?
The conclusion is, that many people will tell you many things. There are differing opinions across the spectrum on this topic, but one thing for certain remains:
The differences are VERY SUBTLE. That said, it is advantageous to have 24 bit. Not because it will make your sound better, but because you will have more headroom and a better noise floor. In essence, your sound will have more room to breathe in the digital realm. Also, less data will be lost overall with 24 bit, so it’s best to try and stick with that number as much as possible.
If you guys have any questions or comments, please Contact me! If you feel this article could be improved in some way or have something that you think I should add, LET ME KNOW AS WELL!! 🙂