16-bit vs. 20-bit vs. 24-bit

[skunkyfunk]

Here is my controversial view on the subject.  16-bit is good enough for me.  Yes it is true that with 16-bit you have a dynamic range of -96dB while 24-bit gives you -144dB.  Why do I think 16-bit is enough?

1.  24-bit converters came along just in the late '90s and early 2000's,  so are you saying that the 16-bit digital ADATs that were used in the early '90s yielded crappy recordings?  Think Alanis Morissette's Jagged Little Pill.  They used 16-bit ADAT for tracking.

2.  CDs, which is still our main medium for music distribution, is 16-bit/44.1KHz.  So everything you record higher than 16-bit resolution gets dithered to 16-bit if you want to encode it for PCM.

3.  The real culprit of bad DIGITAL recordings (not analog) is not the SNR but the absence of using good compression/limiting to ensure that anything does not hit above 0dBFs.  On top of that, digital, whether you're tracking in 24-bit/96KHz or even 32-bit float will sound as cold and sterile as a 16-bit/44.1KHz tracking if you do not use ample equipment that can fatten up your sound such as great mic preamps and compressors.

4.  Source is king.  GIGO.  

5.  I can save more disk space on 16-bit.  (But what the heck if  300GB SATA HD is dirt cheap nowadays?)

6.  Quantization noise is always gonna be an issue with digital regardless of the resolution.  So before you invest on your wow-PC with a 24/96 interface card, think about how much help a good channel strip or mic preamp can do for you.  

7.  Most people buy music not for fidelity, but for quality of the musical arrangement, composition, and performance.  How many people in the world do you think would like to hear your mixes in 5.1 Surround DVD-A?  

8.  Given that you recorded and mixed things well on a 16-bit recorder, it is the mastering engineer's job to make the mixes sound better.  So leave the dirty A/D D/A problems to him.  It is his job to preserve the integrity of your mixes but at the same time add depth to the mix.

[jplacson]

Found this article online.  Pretty interesting.

http://www.tweakheadz.com/16_vs_24_bit_audio.htm

[rocketboy]

If you cant hear the difference, like me..stick to 16bit...hehehe

[skunkyfunk]

If you cant hear the difference, like me..stick to 16bit...hehehe

I can, but what I am saying is I can live with a 16-bit recorder.

[BALDO]

Skunky
Try listening to a cd ( 16 bit recording) on a 5.1 speaker system and listen to a dvd  ( 24 bit recording), close your eyes and i am SURE you will hear the difference.. sure you can LIVE with it but don't you want to listen to something BETTER?  

[KitC]

I normally mixdown to a 24-bit master wav file even if it's 44.1 khz. I only apply dither when downsampling to 16-bit for burning to cd.

[GerardSalonga]

Question for you, Kit, and everyone else....what dithering software do you prefer for taking your 24 (or 32) bit mixes down to 16?  I personally like the sound of Steiny's Apogee UV22HR better than the Pow-r in Logic.  I can't give an answer why....it just sounds better to me.  

[KitC]

Gerard,

I've been using UV22 for most of my dithering; I would be using Sonar's POWr-3 but I'm not happy with S4 mixdowns ATM, preferring to mix audio in Cubase instead. I use the Low setting with Autoblack rarely engaged since I don't have any songs or content with appreciable silence.

[abyssinianson]

as a general approach to my methodology of recording, I usually record at 24 bit, (96 or 192, depending on what rig I am working with) just because I would rather have the programs i am working with have more data points to work with for accuracy than not have enough to dither down to later on in the process.

true, i can't hear the difference between 16 and 24 for most things but just because I can't tell the difference doesn't mean it doesn't contrbute to something in the eventual mixdown process. i agree that the general person doesn't buy for fidelity, still, it doesn't mean I shouldn't put everything of what I can into the mix. As the mixer or engineer, is my job to maintain a consistency even if the other person can't appreciate it that much because they only hear the resulting 44.1 mixdown.

in science and medicine, there is a saying in the lab when working with DNA - just because you don't see enzymes that degrade your work, that doesn't mean they are not there, actively doing the detruction you don't want them to. as such, making every effort to ensure that something comes out in the end as best as it can is a good way of ensuring quality as best as you can provide it, whether it is for music or anything else you work with..

just my 2 cents.