16 or 24 bit

[tranzash]

Just a curiosity, What format do you guys record your audio?
I write tranz, especially with all those filt cutoffs.
G2 and virus are 24 bit, 96khs.
Is it ok to record on 16 bit ?
Am i losing audio quality? Sorry for my ignorance.
thanks in advance for all the info.

[Tomer=Trance]

the idea is to work in higher quality as possible that when the cd burn accurse you wont lose a single peice of data
i tried working in 24\96 and in 16\41
and didnt hear much different if not at all
its up to you
maybe someone can explain this better then me.

[Juho L]

The audible differences between 24/96 and 16/44,1 mixes are really hard (or impossible) to notice without really highend system (and without really trained hearing), so 99,9% of the listeners won't recognise 24/96 and 16/44,1 from each other. The main reason for using higher bitdepths is mixing. With more bits you get better resolution on low signal levels and this enchances the mixing process. If I'm not completely wrong every main sequencer software processes the audiofiles as 32-bit floats. This means that the waves are mixed in 32-bit resolution no matter what bitdepth the files actually have.

But I'd advice for going 24/96 just in case, eventhough it might be possible that no-one won't hear the difference. Better safe than sorry. The only situation where I notice bitdepth reduction is when I convert my project to a wave file (that annoying it-sounds-better-on-the-sequencer effect, which is caused by that 32-bit internal processing).

[Smag]

It depends on what you're recording as to what bit depth you need to use, i.e. if you record a 909 kick it doesn't really matter on whether it's 24/96 or 16/44 as it's such a basic sound. If you're recording a grand piano in a concert hall, then it would be easier to hear differences in the recording quality between 16/44 and 24/96 because the sound is going to be more complex so a higher rate would make a difference. Only use the sample rate / bit depth you need to.

[Timo]

24/44.1

The 24 bits give you significantly more headroom and a substantially lower noise floor when compared with just 16 bits.

CDs use 44.1KHz so I can't understand why people use 96KHz (unless they're aiming it at DVD audio), as 96 isn't a multiple of 44.1 (88.2 is, however), so you're always going to get a little aliasing to a degree. However this is miniscule. I still use 44.1KHz though, as many people say the greater number of bits (ie. 24 instead of 16) makes a far, far greater difference than simply upping the recording frequency. Plus your files will be much larger if using significantly higher freq-rates, so 24-bits over 16-bits is meant to be the best compromise in trying to keep the becoming-larger files as small as possible.

[tranzash]

Thanks for all the reply.

[DIGITAL SCREAMS]

The Virus outputs 24bit/44.1khz

Decent converters (i.e. as those found in a Delta 10/10) provide great sound at 16bit/44.1khz. 24bit/96khz recordings will only be appreciated by those who have high-end hifi (In excess of ?2000+).

Ultimately, it would be ideal to record at 24bit/96khz......provided your not using some shitty cheap soundcard.......kinda defeats the object!

DS

[Panopticon]

Well, I was gonna say it, buy Timo said it first. Headroom and noise....

The added headroom (gained from working in 24 bit) makes it a lot easier on one during the mixing stage. This is, in fact fairly easy to see and hear. In terms of audio 'quality', you can't really hear that much. And the difference between sample rates is even more controversial than the difference between bit depth. I can't tell the difference between 24/44.1 and 24/96. I can tell the difference between 24/44.1 and 16/44.1, but once again I can't distinguish 16/44.1 from 16/96. Bit depth is really the big one....

Additionally, there's no human who can hear 44.1kHz. Of course, the upper limit is a little different for everyone, but 15-22kHz is a more reasonable range for the upper limit of human hearing. Now, I can't hear the difference between 44.1 and 96, but some top mix and mastering engineers claim to hear a difference. If, indeed they can, they still can't hear above human hearing potential. And 44.1 is WAY above that. About an octave above it, actually. So anyway, whenever digital audio is created, there is a low pass filter applied at approximately (theoretically exactly) 1/2 the sampling frequency. So it's possible that in as 44.1 environment, a person could possible hear artifacts due to this filtering. At 96kHz, the filter would be applied at approximately 48kHz, once again WAY out of the range of human hearing. So it is possible for some people to hear a difference. I am not one of them. But when someone says they can hear the difference between 96kHz and 192, well, they're either inventing it in their head, or the hardware is somehow treating it different. But it is truly beyond human capacity to hear the difference of sample rates between 96 and 192 (the filters would, of course, be at 48 kHz, and 96 kHz, respectively).

Anyway, I work at 24/44.1 Best compromise between headroom and disk space.