My usb interface essentially means I record 96k/16bit, but I could technically mix and track virtual instruments in 24bit...
I could also produce an initial 24bit master I suppose.

Is it worth it?

I definitely noticed virtual instrument sound improvements and my acoustic guitar when switching to 96k (from 44) but is 24bit about mixing headroom?

Am I just messing around too much?

Thanks folks!
Nicky

Nope

Nicky, I suggest you always record 24 bit regardless of the sample rate. Typically, 24 bit at 44.1k is used.
Read this: http://tweakheadz.com/16_vs_24_bit_audio.htm

As you apparently do, many people find that they can hear a difference in acoustic guitars, vocals, and other tracks with high frequency content when recording at 96k as opposed to lower rates. So 24/96 is fine, too.

As far as 24 bit, the main benefit is that you get an improved signal-to-noise ratio. In practice, that means you can probably record a dynamic acoustic guitar part and keep the recording level lower. You still hear the delicate quiet parts or details, but won't clip as you record louder passages.

_________________
*Vibrant Audio*

This is something i can't get clear in my own mind............

16 bits gives you a theoretical dynamic range of 96dB. 24 bit gives you 144dB. However, since you can't go above 0dBfs without causing distortion, the ranges are actually 0 to -96dBfs and 0 to -144dBfs respectively. In other words 24 bit allows you to record "quieter" signals before they get lost in the background noise. But, I doubt that most home studios have a noise floor better than -60dB (even that's pretty good) so even 16 bit looks more than good enough. Having said that I've heard that if you are using a lot of VST's and mixing lots of tracks there's some advantage to using 24 bit as it gives a big "safety net" for rounding errors, and hence noise, that the processing might introduce. Does anyone have any definitive answer?

With regard to sampling rate, the Nyquist/Shannon theorem says that you need a sampling rate of twice the highest frequency you're trying to capture in order to be able to reproduce the original frequency curve. Given that the limit of human hearing is 20kHz and that most (all?) of the mics people like us have at hand have run out of steam well below 20k, 44.1kHz samling rate looks good enough - giving an upper frequency limit of 22.05Khz. 48kHz might be useful if you're using an on board sound card as many of them have 48kHz as their native sampling rate (massive generalisation) and are likely to have less precise resampling algorithms than is ideal. 48kHz is also useful if you're worried about steep filters between 20k and 22.05k. But anything above that is just a waste of space.

Hence my short answer earlier. Forget 24 bits, regardless of the facts, it's only useful if you can actually hear a difference. If you can, you'll be the first person I've ever come across who can. Many people do indeed say that they can hear a difference, but saying it and actually being able to do it are two different things. Try mastering the same song to 16 bit and 24 bit, then posting both versions here. Let's see who can tell the difference.

Well, I certainly can't hear a difference and so I'm a 16bit man. But I'm only recording solo acoustic or classical guitar - or sometimes acoustic + voice - so I can't comment about any difference that might be perceivable when applying lots of effects to lots of tracks.

Well... I *am* still thinking about 24bit mixing for that reason... Lots of tracks...

But, my preamps are noisy, and I have to record quiet, so when boosting afterwards maybe 24bit would have helped. Still not sure...

It's really not a huge deal...just read the tweakheadz article I referenced above (and other articles like this one). And make up your mind from there.

I'll stick with what I said about the potential benefit of recording some of the more "delicate" stuff, like soft vocals.

You're going to hear a lot of differing opinions from both pros and amateurs. The consensus generally leans a good bit towards 24bit because "it can't hurt and it might help".

_________________
*Vibrant Audio*

Those are good articles

I'm a girl geek who actually buys CDs. And also listens to the odd vinyl record. And plays most of my music as FLAC even on my phone. And all the guys I work with insist "there's no difference between FLAC and mp3 320k".
Well, *I* can hear a difference. In the treble, high hats, the clarity of acoustic guitars.
And the interwebs is full of folks who say how "hi fi geeky" and ridiculous it is to want anything other than a 320 mp3.
I see the 24/96 thing as *similar*... in that if it yields a better 16/44 overall product...that's a good thing.
Although, I'm not really convinced about the trend towards "studio master" FLAC downloads (eg. hdtracks website).

So...
I'm sticking with 16bit, 96kHz recording, mixing and mastering.

I *love* how Mixcraft makes it possible for me to learn all these things!

Nuffing wrong with geeks - especially girl geeks

_________________
Keep on trackin'

Trevor

When I listen, I forget. When I see, I remember. When I do, I understand. Confucius
I started out with nothing - and still have most of it left! - Seasick Steve

My name is Mark.... and I'm a geek.

The entire topic is endlessly debatable. Honestly I think it may be true that most home/project studios dont have the level of gear or expertise to allow any perceptable advantage at most stages for 24 bit over 16. However it may not hurt to get in the habit, and certainly wouldnt hurt to use the technology if you have it.
I will also concede that any difference may be lost on the average listener who simply doesnt care, nor has the "trained ear" to notice, and to who an MP3 is just dandy. I make music for me, and I care. Therefor I make every effort to record, mix etc. at 24, then resample to 44/16 at the final stage.

But backing up, I interpreted the the original question to be on the value of MIXING at 24 bits when the RECORDING was limited to 16 due to the limitations on equipment? (Interface) If so I would suspect theres ever more reason not to be too concerned about resampling to and from another bit rate, but I dont have any definitive evidence to back that up........ I wouldnt worry to much about it until/if/when you upgrade that interface.
Dont worry to much about the details and enjoy your music!

_________________
Stay in tune,
Mark

Learn from yesterday, live for today, hope for tomorrow. The important thing is not to stop questioning.” – Albert Einstein.

https://soundcloud.com/mbliss01

Hi Mark!! And you're right! I wittered on from the original question!
(I do that a lot I'm told!)

What you say makes sense...just wanted to check n that

Thanks again, really helpful of you

missnickyb, you are right to use flac. mp3s are noticeably "tinny" sounding to most people who record a lot. and thank you for buying cds!! ( i own record stores.)

Record stores are cool. There aren't enough! (here in the UK). And you're probably not in the UK

Some info here that relates to sample rates, mixing, and mastering:
http://productionadvice.co.uk/when-to-dither/
http://productionadvice.co.uk/when-in-doubt-dither/

Regardless of the sample rate you use, digital mixdowns can introduce some subtle artifacts in the audio.

mbliss posted a link to iZotope's guides in the Tips And Tricks forum. One is a dithering guide:
http://www.izotope.com/products/audio/ozone/support.asp

_________________
*Vibrant Audio*