Hey guys,
Was hoping you could help me understand something. If I have two instances of the same 16 bit audio file in a project that I have layered together and edited using filters and envelopes (using Native Instruments Battery 3), will there be any point in exporting to a 24 bit audio file?
I am basically taking kick samples from some of my favourite tracks and then using a layering technique (of the exact same audio file) to get a nice clean sounding kick sample.
When you process and mix two 16-bit files together, the result could easily be greater than 16-bit.
Cubase works at 32-bit, so when you process and mix your two files, your are now working at a 32-bit file (actually listening in 24-bit - because your audio interface is 24-bit).
So, you will be “compressing” the end result slightly to get it back to 16-bit - this will not produce a different level, but will just reduce the definition.
But, the question is, can you hear the difference? Probably not, especially if it is a big in your face kick drum, If it was a piano note dying away into silence, then probably.
But I would suggest you always keep your Cubase projects at 24-bit anyway, and only convert to 16-bit at the mastering stage (assuming it’s going to CD).
That makes a lot of sense. Thank you for that awesome explanation. 24 bit it is!
So I re-exported in 24 bit and I definitely can’t hear any difference. I compared by alternating solo and it all sounds the same to me. None the less I will keep the 24 bit files. But then I dug a little deeper and wanted to see if I could actually see the finer resolution in the audio waveform by zooming in vertically. It’s exactly the same as the 16 bit file. Maybe I’m thinking too much of this but is this normal?
If the two files are identical it makes sense that they would appear the same.
If you adjusted one of the files, by ading compression, delay, reverb etc. you should see the difference.
I see what you’re saying. But if one has been exported to 16 bit and another to 24 bit, shouldn’t you be able to see the extra amount of bits in the fullest zoom level. It appears that 16 bit has the exact same as 24 bit. And I’m talking vertical slices here, not horizontal, which would be sample rate. Just a little confused now because I thought that you should at least be able to see the difference in the waveforms even though they sound the same.
Confirm such stuff with a null/phase test.
Do you mean to invert the polarity of one of the files and play them together to see if they are exactly the same?
If the sum (mix) of the two 16-bit drum sounds does not exceed what 16-bit can deal with (no over flow) then there will not be any difference going to 24-bit anyway.
But if you stick to 24-bit then you’ll never need to worry about over flow bits and things…
If you want to try to see something, then make sure your kick drums are both at 0dB (max level) and then combine them, the combined level should go up 3dB causing a bit overflow, so you should be able to see a difference then. But with a kick drum, I think it is something you might be able to see on an analyser or fail a null test, but probably still not audible.
Yes, you’re right about how to do the null test by the way.
Interesting. I guess the reason I’m so curious is because I’m getting the feeling it’s not exporting to 24 bit properly. I’ve tested and compared many other sounds and I still can’t see any difference. Does it matter if I’m using the built-in sound card on the Mac as my audio device in Cubase? At the moment my 828mk2 is playing up and I’d rather not use it. But it shouldn’t matter, right?
Thanks to this article:
I now understand the difference between 24 and 16 bit. I thought 24 bit had more slices in the entire dynamic range. But now I understand it only has more slices at the very quiet parts, below -96dB. This is why it was all looking the same to me because I was comparing in the mid volumes of the waveform and not the quiet parts. Here’s the difference:
16 bit:
24 bit:
And thank you to all you guys for chiming in. Bottom line is use 24 bit! 