1

This might be a bit of a beginners question but I was wondering if there was any benefit to working in Pro Tools at 96k 24bit when all your sounds will be bounced out and used at 44k 16bit ?

Cheers,

Eric

flag

4 Answers

1

In my opinion, working with 96KHz in a project intended for 48KHz is minimizing the risk of audible degradation while mixing several sources together, as well as when using effects. There are actually some programs that handle higher frequencies worse then lower ones, but that's very rare. What on the other hand is very important is that you use a really good converter when down-sampeling it, or everything you gain by using the higher setting will get lost, and perhaps then some...

Higher bitrates are practically always a win situation, but sample rates can be a little tricky if you're unsure about your monitoring. It's important to use your ears and pay close attention to what happens when you experiments.

link|flag
I'd like to learn more about this. Is there a source where i can read about mixing at higher sample rates to avoid degradation of the mixed signal? I'm curious as to the principle behind it. – Roger Middenway Jun 1 2012 at 14:49
Not that I know of I'm afraid. I've learned this from years of frustration and trying out different techniques and equipment when the mixes and masters didn't sound exactly the same after rendering as they did before. There are some simple rules to work by though: First of all: Trust your ears. If it doesn't sound good, it doesn't. If it sounds good, it still might not, but it's a great start. Digital is either 1's or 0's, there are no 1.5 or 0.7 or such, so never work with samplerates that cant be multiplied or divided evenly. Therefore the only acceptable ones in film is 48, 96, and 192KHz. – Christian van Caine Jun 8 2012 at 12:21
2

Having a session in a higher bit rate will allow a larger dynamic range, allowing you to get away with more gain staging with less chance of clipping in your channel. Higher samples rates in your session allow you to time stretch your sounds while retaining more high end frequencies (as the ones that are in audible then roll down into the audible range). You should always work in 24 bit when possible. 96k isn't necessary usually unless doing some sound design. Regardless, it is standard now to work in 48k no matter what you're working on and it is very easy to down-convert. I always work in 24-48 and 96 when I edit sfx, do sound design, etc.

link|flag
2 
 Almost correct, although one point is incorrect. Time stretching though suffers from the same problem at all sample rates (while some are more forgiving than others before artifacting is present). What I believe you mean is not time stretch, but Varispeed/pitch shifting/pitch scaling (e.g. the pitch is altered proportionately to amount of expansion or compression to the sounds duration). In this case, yes the higher frequency content will give you more mileage as far as full 20-20K fidelity is concerned as long as it still exceeds the Nyquist limit as far as our hearing range is concerned. – Stavrosound May 31 2012 at 8:49
Time stretching is defined as altering the duration of the sound but NOT the pitch. But for pitching/Varispeed, working in 96k will allow you cut the speed in half before any discernible fidelity loss begins to happen when it's slowed down further. With 192, it's 1/4th speed before it begins to happen. – Stavrosound May 31 2012 at 8:52
Sorry, yes, i was referring to varispeed only, not the other algorithms. – ColtonRybus May 31 2012 at 9:57
1 
 96k is great for stretching sounds and preserving the air and sizzle that gets lost. transients still go south but at least you find new ones in there. too bad 96k means less voices. – georgi May 31 2012 at 11:54
Appreciate the feedback guys. I'm primarily doing sound design for video games so I think I might kick up my PT template to 96/24 for a bit and hear how that effects my sound design. – Eric Van Amerongen May 31 2012 at 15:59
1

96k means that if you are pitch shifting anything the effect sounds more natural. 24 bit means that there is less chance of clipping when you are recording. 48k 24bit is the norm just now for video work, and 44.1kHz 16bit is the norm for audio only projects.

link|flag
0

If you are going to end up with 44.1 kHz, 16 bit files, you are better off working at 88.2 kHz. Downsampling from that sample rate is simpler mathematically--you are cutting the sample rate exactly in half. Going from 96 kHz to 44.1 kHz is a lot more complicated--the best sample rate converters can do it pretty competently, but why take a chance on artifacts occurring when you don't have to?

For film, of course, you are aiming at a final product at 48 kHz, so then you would use 96 kHz (or 192 kHz) in your initial recording.

BTW, I think the main benefit of recording at higher sample rates occurs during the initial A to D conversion--real world problems with anti-aliasing filters are moved up at least an octave which makes the artifacts less audible. Recording at 24 bits is always desirable as well--any audio level processing that occurs during editing will be done more precisely, and you can leave more headroom during the initial recording. Only when you are rendering the final file should you dither down to 16 bits.

link|flag

Your Answer

Not the answer you're looking for? Browse other questions tagged or ask your own question.