need help setting up 24bit!!

hey, jus got a new creative 24bit usb soundcard for my laptop. have the device installed and working, but because 24bit is new to me, i want to check i have ntrack set up correctly in order for me to make the most of 24bit. this is a screen dump of my set up in ntrack:NOW EDITED OUT
have i got the right sampling frequency? i have the recording settings set to just 24, not 24 unpacked whatever that means?! and when i record i understand that i av to convert back to 16bit, is that a simple process.

(you might want to re-size that picture - it’s over 200k which will be a killer for folks on dial-up).

First up - you should try using the WDM output driver rather than the MME driver.

I suggest you try the different settings for 24 bit (packed etc), and see which ones work for you. On my card, only one setting works (I forget which at the moment).

As for the sampling frequency, there’s no need to change it unless you’re trying to get (arguably) better quality. I’m not sure if the Audigy suffers from the old “48kHz problem” like the earlier Live cards. That may affect your decision.

Also remember that 24 bit is half as many bits again as 16 bit so all your files will be 50% bigger, and put more load on your PC. If you increase the sample rate, you add to that.



Right click on the little hammer at the bottom of the recording vu-meters and select 24 bit (packed/unpacked/left justified/blah) - As mark said, most cards will only support one format so you’ll need to experiment to find the one that works. You can do the same on the playback VU to get playback at 24 bit.


hey, ive just tried the settings and ur right that only one will work and it was the one i had it on, just 24bit.
with the sampling frequency, i have it on 48000, do u think i should leave it one that?
what is the advantage of havin 24bit over 16bit, are you able to explain what makes it better? i assume its just more bits of sound making the sound better, but will this be really noticable?

24 bit provides more dynamic range (about 110 db) as opposed to 16 bit (about 84db) - I might be wrong on those numbers (anyone? anyone? Bueller?). Basically you have a greater range of amplitude so that the quiet is quieter and the loud is louder. Results are better (IMHO) if you record/mix/master @ 24bit then dither down to 16bit when you’re ready to go to CD - this preserves as much amplitude as possible during the mixing/mastering process.

Sample rate refers to the number of times the sound is sampled per second. It also determines the highest possible frequency that can be recorded (frequency = sample rate/2)
so at 48K sample rate you can record up to 24Khz, at 44.1 you can record up to 22.05Khz (Read up on Nyquist Theory). This can be deceiving because high frequency content can sound pretty nasty if it’s close the Nyquist frequency (24Khz @ 48K Sample Rate), this is because it becomes a square wave (buzzy or distorted sounding).

This is always a trade off of quality/practicality, sure my interface will record a 192Khz - but do i really need to record music that only my dog can hear? At 192Khz sample rate, the highest possible frequency is 96Khz, if I was capable of playing this tone through some sort of speaker, my garage door would be opening and closing like crazy, the TV would be changing channels (Yes I know TV remotes are IR but I’m making a point and besides I’m ranting here).

Ultimately I think I’m quoting Phoo or Achimedes:

“It ain’t the gear - it’s the ear” - So experiment, try different and see what your ears tell you.



Chutz, it’s 6dB for every bit, ('cos each extra bit gives twice as many “levels”, and a doubling in voltage terms is 6dB), so 96dB for 16 bit, and 144dB for 24 bit (except, in the real world, you’ll never get anywhere near 144dB). :)


What chutz said about dynamic range is very true…but, here’s just a wee bit of dynamic range heresy:… :D

128 dB or thereabouts is often quoted as the human dynamic range.

Stick a PPM on your local pop radio station, and what’s the dynamic range of the music? You’re lucky if it’s more than about 2-4dB. With BBC Radio One, the PPM needle sits at a gnat’s below +8 dBm, and barely flickers.

What’s the level of the Juke box in your pub? Probably no more than 90-100 dB SPL. What’s the background noise? 60-70 dB SPL?

So, you really don’t have much dynamic range to play with, and even if you try, the mastering engineer and/or radio engineer is gonna compress the shit out of it.

So, use 8 bit for playback, it’ll give you all the dynamic range (48 dB) you’ll ever need. :) (Only kidding, the quantising noise is unacceptable with 8 bit).

But, dynamic range is more important when recording, because it gives you “headroom”. That means you can record without worrying about clipping, or about disappearing down in the noise floor.

Not only that, but because 24 bit has so many more “levels” than 16 bit, there’s less quantising noise; (which is largely irrelevant; you can hear 8 bit quantising noise, but not 16 bit).

It’s irrelevant that is, until you start performing mathematics on it.

Which is what a VST or DX Fx plugin does.

Every sample has an error in it if you think about it, (the chances that an analogue voltage at a given instant, exactly corresponds to a defined preset level, is near zero), so, every sample contains an error.

And if a plugin magnifies that error, then it can become audible.

But, the errors at 24 bit are much, much tinier than at 16 bit. So, use 24 bit, and if you don’t believe all the crap I’ve written, then as chutz says, use your ears. :D