Bit depth settings: it's so simple a child could figure it out

From the current documentation:
http://www.cakewalk.com/D...age=3&help=0x22B17
____
Audio Driver Bit Depth. Determines the number of bits per sample used for communicating with the audio hardware for playback. Your audio hardware must be capable of supporting the setting you supply here. In Cakewalk, choose between 16, 18, 20, 22 and 24. In most cases, even if your hardware is “advertised” as being 18 or 20 bit, you will want to set this value to 24 for optimum performance. Cakewalk 16-bit only.
  
Note: Using an audio driver bit depth that is greater than 16 means you also need to choose a preferred setting in the Stream >16 Bit Data As field in Edit > Preferences > Audio - Audio Profiles.
____
Isn't that delightfully weird? First it insists that one's hardware must be capable of supporting the setting I put in there, and that in Cakewalk (in case I had gotten so befuddled as to forget what program I was configuring), I should choose between 16, 18, 20, 22, and 24.
 
Then it contradicts itself by saying that in most cases, even if the clueless losers who made my interface "advertised" it (yeah, right) as being 18 or 20 bit, I should set it to 24 for optimum performance. What rebels we are. We'll not have our performance restricted by mere "advertisements."
 
Then out of nowhere we are informed (cautioned?) in stereotype Native American dialect "Cakewalk 16-bit only." Well, which is it? Do I listen to the Reckless Ones or Chief Hollywood Stereotype? Or the "advertisers" who sold me my interface?
 
Reading further, It looks like the Chief has the right idea, because if I pick one greater than 16, then I'll need to choose a preferred setting somewhere else, and no hints are given as to what that setting should be (although my guess would be I'd put in the same number? That doesn't make sense, though, why should the user need to type in a duplicate number?). Turns out it doesn't matter anyway, because that setting doesn't even exist. At least it doesn't on my system. Maybe if you're running in a different driver mode? Not in ASIO, though.
 
Bit depth settings are hard enough to figure out without the manual trying to mess with my head.
 
By the way, I'm just having fun. Most of the manual is pretty well-written, I just got a kick out of this page. "Cakewalk 16-bit only."

Aside from not being clear, it appears to refer to a pre-X series version SONAR. Feel free to submit a support ticket to get the documentation updated.
 
ASIO drivers usually run at a fixed bit depth.
 
The transport module shows the current record bit depth, not driver bit depth. As a rule the record bit depth should be set to the driver bit depth. Unlike sample rate, projects may contain clips at different bit depths.

I thought one was only supposed to read the directions ONLY and AFTER all else fails. }:( Now I learn I've been doing it all wrong.

Euthymia
From the current documentation:
http://www.cakewalk.com/D...age=3&help=0x22B17

 
If you are trying to gain a clear understanding, definitely don't read that page.

Very, very old and confusing topic. I haven't heard of any vendor make a  piece of audio hardware design for AD/DA converters to convert analog audio to 20 bit digital since Sony dropped the Mini Disc recorders. Their theory was based on delivering better audio quality that that of a CD's standard of 8 bits per channel = 16 bit stereo file, and at a MUCH smaller compressed filed size.
 It actually worked in a physical sense, but not many people have a solid grasp of "Physics" and couldn't rap their heads around the idea of a 20 bit audio recording that only needs 1.5 MB of storage space could actually sound "better" than a 16 bit audio recording that requires 650 MB.
 People FLATLY REFUSED to "BELIEVE" it, so they refused to except it, and so the technology died.
 And the same type of conventional wisdom still exists and is still endlessly debated today.
 A perfect example is all those that falsely assume and staunchly claim that 12' vinyl albums sound and are bigger & better than any album stored digitally on a tiny little USB thumb drive.
 let me be perfectly clear.. You can now store literally 1000's of albums with much superior sound quality on a tiny little USB thumb, which is now as obsolete as a vinyl.
 Why bother with a USB thumb drive when you can store them on your cell phone, and walk around with an entire library of music in your pocket that would take up a large room in your house to store in vinyl records?
 
 Today, all AD/DA bit depth settings have settled into an industry standard evenly based on multiples of "4" such as bit depths of 16, 24, 32, 64, on up respectively and frequency rates from 44.1 k/Hz on up to 192 k/Hz and beyond.
 However, it's humanly impossible to tell/hear the difference in most cases at least without an oscilloscope  to measure these frequencies undetectable by going over the range of the human ear even if you have state of the art top of the line audiophile grade equipment.
 There have been many double blind tests using trained audio pros in near perfect listening environments, proving it's very hard to accurately to "guess-ta-mate"  the difference between a high resolution .wav file compared to a high resolution .mp3 file without the help of an O-scope.
 And  IMPOSSIBLE listening thru consumer grade equipment.
 
 So it's very reasonable to conclude that; "It's not really about the bit depths and sample rates, silly, it's all about AD/DA conversion and DSP (digital signal processing).
 With the steady and rising advancements in technology, today's cellphones offer better and greater AD/DA conversion and DSP than yesterday's state of the art audiophile DAT recorders.
 
You can set SONAR/CbB up to automatically resample all  audio files imported that aren't at the projects default settings, but only in multiples of "4".
 Very simple solution, you'll get better/smoother performance by relieving the clock source your DAW of the additional  tasks of calculating and converting several tracks of multiple sample rates in one project in real-time.
 
Set "import" options once, than you forget about it there after. 
 I also have my "export" option set to the industry standard 24/48 for professional collaboration.
 And created an additional template set to 24/44.1 which is optimized for Bandlab.com for fun collaborations. The secret sauce there is 24 bit sound quality resolution, faster uploading and Bandlab's Mix Editor seems to favor and run smoothly at the 44.1k/Hz standard.
 
And that should leave you more time to fret about making the perfect sounding pristinely authentically life like sounding recordings, than beat and deconstruct them back down by smashing them with all you favorite flavors of vintage Harmonic Distortions we've all grown to love and expect from yesteryear.
 
Psst, iZotope offers a free plugin called "Vinyl". It's FUN, you can add and adjust, hum, mechanical noise, scratches and dust noises, warp and such, all based on RPM and period in decades technology, and apply spin down speed at the end to make it sound like the belt broke on your turn table.