Do you guys use a higher sample rate than 44100?

[author]

Do you guys use a higher samplerate than this for making music? does it make a big difference?
Is it standard that most pros will be using a higher rate than this?

I'm asking because I've just switched from an internal soundblaster soundcard to an external one: edirol ua-25.

I had a few issues with sampling rates using it with cubase and rewire, but doing everything at 44,100 seems to have sorted it.
I'm a bit bummed out that I'll won't be able to use the 96,000 sampling rate when rewiring or using reason.

should I give a shit?
is 48,000 any good?

any input much appreciated.

[steakbox]

i wouldn't give a shit, but i generally don't give a shit if it's hi-fi or lo-fi, as long as it's good. most people in these communities don't feel that way.

[Sharmaji]

for recording, sure. for electronic music production where i'm bitcrushing stuff down to 9 bits quite often-- no.

[mawltea]

if you're producing something in say Logic, it will only really make your bounced track bigger if you use a higher sampler rate. it won't make your music sound any better if you put it higher (our ear won't be able to pick it up anyway).

[3za]

[author]

cheers for all of the help guys. worries averted.

I'll save my 96000 for recording.

[macc]

For making some filthy dubstep from already 44.1khz 16bit samples... No

Be aware that working at 24 bit throughout will allow for much les degradation when further/repeatedly processing samples, even if they were initially 16 bit

I think you know that Depone, more a general comment Lo-fi is one thing but you want your lo-fi-ness to be captured perfectly

[Mad_EP]

Yeah- on the one hand, if you are just cutting & pasting samples... it probably won't matter much to work in 16/44.1

On the other hand- if you are putting effects on those samples or using VSTi's.. then higher bit & sample rates do in fact make a difference....



- and of course, if recording voice or instruments, then use the highest sample & bit rate you have available to you.

[dequo]

44.1 allows for 22.05KHz on each channel, most of us cant hear above 18Khz at our age so anything above that is pretty pointless for playback

[paradigm_x]

badly designed filters etc will cause much more aliasing at 44khz than higher sample rates, well below 18k. but tbh i dont think its owrth worrying about when purely itb.

[gravity]

48khz 24bit over here.

i find higher sample rates make your high end sound a bit more crisp. tbh 48 isnt a huge improvement, but it makes a little bit of difference. using a higher bitrate does make a big difference though, and for recording a higher sample rate does have an effect, especially at 96+, as it makes things sound really smooth and crisp. although obviously you'll want an interface with good converters.

[danoldboy]

I don't see what "hi-fi" or "lo-fi" has to do with which sample rate and bit depth you choose you make your tunes at. If you work at 48/24 then you have the option of creating a mix that sounds lofi to the degree that you want. Using shit samples/ lowering your sample and bit rate means you're stuck with what you've got from the outset with limited options for scultping your sounds further.

[test_recordings]

24bit/96k standard. Digital's constrictive and destructive though, I prefer analogue as it's real as the universe it was made, recorded and reproduced in...

[sun_in_aquarius]

from what i know if u write at 48k it will but more stress on conversion down to 44.1 for cd burning so wont sound as good in the end, if u want to go higher go to 88 k because its an easier conversion because its exactly double........correct me if im wrong....

[macc]

from what i know if u write at 48k it will but more stress on conversion down to 44.1 for cd burning so wont sound as good in the end, if u want to go higher go to 88 k because its an easier conversion because its exactly double........correct me if im wrong....

Yup, that's a bit of an old wives' tale.

[lowpass]

from what i know if u write at 48k it will but more stress on conversion down to 44.1 for cd burning so wont sound as good in the end, if u want to go higher go to 88 k because its an easier conversion because its exactly double........correct me if im wrong....

Yup, that's a bit of an old wives' tale.

yeah I'd call b/s too, plenty of people work at 48k and get amazing results when bouncing to 44.1 from what I understand it just gives the process' e.g plugin's, FX more headroom to work their magic?

[macc]

from what i know if u write at 48k it will but more stress on conversion down to 44.1 for cd burning so wont sound as good in the end, if u want to go higher go to 88 k because its an easier conversion because its exactly double........correct me if im wrong....

Yup, that's a bit of an old wives' tale.

yeah I'd call b/s too, plenty of people work at 48k and get amazing results when bouncing to 44.1 from what I understand it just gives the process' e.g plugin's, FX more headroom to work their magic?

Don’t go using the ‘m’ word round here mate


Nah, it’s a bit of an old wives’ tale in that in non-double SRC cases, the data is upsampled to a ridiculously high intermediate rate before being brought down to the destination rate. This is obviously more costly in terms of processing than simply going to half, but that is a) a relatively very small computational cost, b) not going to lead to more error, and c) only one aspect of everything that is happening (anti alias filtering being the big one).

But people just assume that because in their mind the factor of say 96 > 44.1 is much harder to figure out than 88.2 > 44.1, that it must be the same for a computer.

Computers are quite good at maths though


As for your comment on FX and so on, it’s nothing to do with headroom (while I think I know what you’re getting at). It’s about relaxing the requirements on anti aliasfilters, which can be higher up out of the audible band, softer, and therefore less damaging to the audio, less risk of aliasing blah blah. Whether all that is such a big deal these days I dunno. Most plugins oversample now anyway which while not quite the same, does have advantages.

My *personal* opinion is that you should try working at high SRs for yourself and make your choice based on that *after* downsampling to the final destination rate (ie 44.1kHz). If I’m making a tune I just go with 24/44.1 to avoid sample rate conversion.

How boring was that post….

[collige]

32 bit for the mother fucking win.