Confused about sample rate and bit depth for rendering mix

[Gwaan]

Hi

I am close to rendering down my first track and am a little confused about how live does this.

My track uses samples at 16bit 44.1Khz and so am wondering if there is any benefit of rendering the mix at 24bit/96Khz.

I'm using an emu 1616M soundcard capable of upto 192Khz but am confusedn whether the audio goes through the soundcard during rendering or not as when i change the sample rate in lives preferences it generates a conflict with sample rate within the 1616m settings, however when i change the sample rate within the rendering options no such conflict messages occur.

Could anyone please explain how the mix is renderd to disk. Is the digital information somehow converted into analogue and sampled into a digital again using a higher sample rate?

Could someone please just explain the process a little please and if there is any benefit in rendering the mix at higher bit depth and sample rate than that used in the initial samples

Thanks

Gwaan

[Der_Makrophag]

Hi.


I think rendering has not much to do with your soundcard, because this goes "offline". I do not know for shure, but I think the CPU's doing all the work, for that everyone can sample at these rates and depths, independet from his onbord-shit-soundcard. That's why it's calles rendering. There eis no Audio stream getting anywhere. (compare this to burn a CD. This is also not dependet on your sondcard, right?) Only that what your system is "playing" is processed by your soundcard.
As I said I am not shure, this would be interesting for me to know too.

[laird]

You should render to 16bit/44.1khz (and use dither too)

if you want to make an audio CD there is no other option. Also, if you plan to convert to mp3, use this bit depth and sampling rate.

if you want to save these files and have them mastered, 24bit/44.1khz is a good option (no dither). It doesn't make you sound better, but the extra bit depth means you dont have to try to make the song as loud as possible. Instead, this unmastered file should not go above -6dB in volume (or even -10dB max)
The benefit of 24bits is you can save some of the detail that will be lost when "truncating" to 16bits. Once you lose this stuff, its gone, so it can be nice to have a 24bit audio file backed up somewhere in case you decide you dont like your 16bit render.

The reason for this is not becuase simply up-sampling a 16bit file to 24 bits makes it sound better, but because Live uses 32bits when mixing all your 16bit files together, and many effects plugins also utilize this higher amount of detail (important for, say, reverbs)

In depth discussion can be read in Bob Katz's book Mastering Audio. Check your library for a copy, it may be a little dry for your needs right now to be worth spending $40 on.

[Tone Deft]

render a few different ways and use your own ears.

what you'll find is that maybe there's a slight difference, don't psych yourself out though, the differences are very subtle.

good advice so far.