Jump to content

What evil comes from an overkill sample rate?


Recommended Posts

Also known as, if I have the capability to use 96 khz in recordings, should I?

 

Presuming my computer hardware can keep up with it, and all of my virtual instruments and insert effects can run at 96 kilohertz, is there any harm to a sound file from recording at a higher sample rate, or is this arbitrary?

 

Basically- is the only problem with a 96 khz sample rate that it takes more time/latency to calculate by the computer?

 

Also, in another hypothetical scenario- let's say I've recorded all my instruments at 96 khz, but want to use a sound loop recorded at 48 khz on the track as well. Will converting the 48 kilohertz loop to 96 kilohertz cause audible damage? Would it be best to convert the 96 kilohertz sources to 48 kilohertz, or the loop to 96?

 

Thank you for sharing your knowledge!

Link to comment
Share on other sites

If you use software instruments, those make the biggest difference when trying different sample rates: some instruments sound better at specific sample rates - and not necessarily the higher ones. That difference is much more obvious than any audio recording/processing difference, and yet I don't see many people mentioning it when discussing sample rates. Take a software instrument, dial in a sound and try different sample rates. Depending on the sound you dial in you may hear HUGE differences depending on the sample rate of your project. I'm talking differences anyone in the street would be able to identify while talking to their grandmother on their cellphone. Almost like playing different patches from a synth.

 

Depending on the clock you use, or the clock used by your audio interface, and the quality of that clock, it may or may not clock 96k properly. So if you're using some below $1,000 audio interface and hoping 96k will sound better than 48k, you're most probably wrong, as the clock will be much more stable at lower rates (like 48k) and start introduce jitter at higher rates (like 96k). That means you're introducing more distortion to your audio signal by using higher rates.

 

Not to mention that higher rates put extra load on CPU, RAM, HD space and data streaming.

 

If you've already recorded your instruments at 96k, I recommend continuing working at 96k and converting your only 48k loop to 96k. Unless the extra load on your system becomes a problem.

 

Converting sample rate introduces distortion, but algorithms are getting better nowadays. Try converting your loop and tell us if you hear anything. If you do it will most likely be subtle.

Link to comment
Share on other sites

For recording audio, if you don't have a really stellar recording setup, you'll never hear the difference between 44.1K/48K and 88.2K/96K. All the higher rates are going to do is tax your processors, reduce your track count, and double the amount of disk space it takes to record on (and later, back up your data). And when it comes to delivering music into the real world, anything you record at 88.2K or 96K will have to be dumbed down to 44.1K or 48K. (And at that, it's bad enough that 24-bit recordings have to be dithered down to 16 for delivery on CD). And if your ultimate goal is to output to MP3, there's absolutely no point in using higher sampling rates than the standard 44.1K or 48K because:

 

a) no one will ever hear the (supposedly) higher quality that you can get from the higher rates, and...

 

b) even if you did hear the difference at the higher rates during production, you're just setting yourself up for disappointment at the end of the mix when you have to sample rate convert to lower rates so that the rest of the world can hear your music

Link to comment
Share on other sites

http://www.lavryengineering.com/pdfs/lavry-white-paper-the_optimal_sample_rate_for_quality_audio.pdf

 

This paper doesn't deal with production though. If you're working with existing 44.1 kHz samples in your production I'd stay at that sample rate since the realtime SRC will severely degrade the quality of the sample playback. Even if you were to asynchronously (offline) SRC your samples first you'd still manifest inter-sample peaks with any heavily processed drum samples, such as those that appear in most modern drum libraries. That can potentially cause more distortion, more overloads and certainly less headroom in your mix.

 

The above issue is often overlooked in these discussions because the discussions tend to deal with recording, theoretical or mastering related issues - not production.

 

I generally recommend 44.1 kHz for all productions that use 44.1 kHz samples and 48 kHz for those that don't. While >48kHz - 96 kHz may theoretically be better with some converters, as explained by Dan Lavry, the difference is negligible in my opinion. >96kHz sample rates will always objectively degrated the quality of the captured bits and should be avoided.

 

Also please notice that 88.2 kHz is not a better choice that 96 kHz if your target is 44.1 kHz (half of 88.2 kHz). All modern asynchronous SRC algorithms use upsampling to the lowest common denominator. The exception would be realtime SRC, but that should generally be avoided in the first place.

Link to comment
Share on other sites

The sound library I usually reference for my percussive elements are all set at 44.1 khz, with the more detailed sounds being at a 24 bit depth and others at 16 (presumably for space-saving reasons). This holds true throughout some of my other sample libraries as well.

 

Now, I know that my R3 synthesizes at 96 khz before going through it's D/A (and I believe my Roland GAIA does as well). My voice is obviously an acoustic sound source being recorded into an analogue microphone, and my Moog Slim Phatty is analogue as well.

 

Since I'll be using my lower sample-rate drum/percussive samples, I'm presuming that I should simply record everything at 44.1 khz? I've checked into Logic, and it appears that I have been- and for my non-professional standards, it seems to capture a fidelity that allows me to make a proper synth patch and hear the small nuances of slight patch changes.

 

I realize that at my level, small nuances like a change of sample rate won't change much- but I'd like to maximize the quality as best I can!

 

As a side note, the interface I will be using is a PreSonus Firestudio Mobile-

http://www.presonus.com/products/Detail.aspx?ProductId=58

 

So again, given my environment, monitors (Rokit 5's), interface, and ability to hear professionally (or lack thereof), I presume that a sample rate above 44.1khz will be fairly arbitrary for my purposes.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...