Digital Users - Don't Like Being Stuck With 48kHz?

L

Lionheart

Guest
The Steinberg UR28M has an excellent solution to allow you to run your interface and DAW projects in whatever sample rate you want. Here's a quote from the manual:

"S/PDIF IN is equipped with the SRC (Sampling Rate Converter) function. Even if the sampling rate at which the device is operating differs from the sampling rate of the audio signal input to the S/PDIF IN, the SRC function will automatically convert the rate so that playback will be correct. SRC is only available when INTERNAL is selected as the clock source of the device."

I have the UR28M and run it and my Cubase projects in 44.1kHz with no problems while using a S/PDIF cable for the Axe-Fx II.

Just something useful that I thought I'd pass along to the community.
 
vinnieRice said:
I don't understand why people record at 44.1 - 48 is definitely audibly superior.
Click to expand...

I'm not sure if you're joking or not, but I haven't heard a sample rate conversion higher that sounds better than using 44.1kHz. Unless you have the ears of a dog, I don't think anyone's going to benefit from the extra kHz.
 
While it is not wrong (according to Nyquist-Shannon theorem) that 44.1 kHz sampling rate is enough to properly represent frequencies up to 22050 Hz (a good human ear is only able to hear up to 20000 Hz), you might encounter problems with that "limit". For example, when using an EQ with a steep slope at high frequencies, that slope could get completely unsymmetrical because the DSP is unable to compute anything above 22050 Hz - which you definitely can hear.
 
only crap vst having problems with high frequencies .. most of them use oversampling to generate exact results up to 20khz.

its true that more khz is in theory better when it comes to processing audio. i for myself did not encounter any situation in my musical daily routine where 48khz resolution wasnt high enough.
 
The only time that I worry about sample rates is when I am trying to match devices or material.
 
i record in 96 kHz sampling rate and it`s away different !
so changing from 44.1 to 48 definitely makes different .
 
Stillbruch said:
While it is not wrong (according to Nyquist-Shannon theorem) that 44.1 kHz sampling rate is enough to properly represent frequencies up to 22050 Hz (a good human ear is only able to hear up to 20000 Hz), you might encounter problems with that "limit". For example, when using an EQ with a steep slope at high frequencies, that slope could get completely unsymmetrical because the DSP is unable to compute anything above 22050 Hz - which you definitely can hear.
Click to expand...

It's not necessarily about the 22k 'limit' - it's about the fact that, if the sampling rate isn't high enough, certain dsp processing can create aliasing artefacts that are well within the audible range - or even outside the audible range but mess up other processors which in turn create artefacts within the audible range.

The worst case I had of this was with a distortion plug-in that was (as a result of aliasing) creating frequencies that were very high in energy but below the range of my hearing (or speakers). These frequencies weren't audible to me, but were completely screwing with the response of a compressor that I had later in the chain. This shit builds up, though most well-written plug-ins oversample now to avoid it.
 
The actual quality of sampling rate is dependent on the hardware and software implementation; not a theory.

That is why people claim to "hear" the difference. With their gear and signal chain, they can hear a difference. This doesn't have anything to do with the theory of sampling rate.

Personally, I avoid all sample rate conversions until a final 2trk master.

For my setup and gear in my home studio, 48K is the best.
 
barhrecords said:
The actual quality of sampling rate is dependent on the hardware and software implementation; not a theory.

That is why people claim to "hear" the difference. With their gear and signal chain, they can hear a difference. This doesn't have anything to do with the theory of sampling rate.

Personally, I avoid all sample rate conversions until a final 2trk master.

For my setup and gear in my home studio, 48K is the best.
Click to expand...

This ^^^ The quality of the ADC's, and the circuitry leading and following them, will have the biggest impact on the overall quality of conversion. I had an IZ Radar24 system several years ago that had a max recording rate of 48K/24bit. It blew the doors off "prosumer" systems that recorded at 96K/24. Higher sample rates also translate into much larger file sizes, though with terrabyte drives being relatively cheap that's not as much of an issue as it was a few years ago. With most of the bedroom recordists and project studio crowd using inexpensive, off-shore built outboard gear, the depth and dimensionality of that higher sample rate recording is questionable anyway.

Another aspect to this is that if you're burning to Redbook standard CD's you'll eventually be downsampling to 44/16. Worse still is that most of us are ripping our stuff to compressed formats (MP3, etc) which further degrades the final product. Most modern DAW's will let you up-sample a track, though the old adage about polishing turds comes to mind. If the rest of the signal chain was cheap off-shore gear, what are you really gaining by up-sampling a 44.1 or 48K recording to 96K or higher? OTOH, I've heard mixes from full blown pro studios recorded at 96K that sounded like ass. They had quality gear, but hadn't a clue about 'old school' concepts like mic placement, pairing certain mics with certain mic pre's, proper EQ'ing and compression, etc.

Bottom line - the Axe's 48K output rez is more than sufficient for making quality recordings. The Axe is the least of our concerns and we should be focusing on getting things to sit nicely in the mix.
 
You must log in or register to reply here.
Back
Top Bottom