Re: "Bit width" vs "bit depth"
Re: "Bit width" vs "bit depth"
- Subject: Re: "Bit width" vs "bit depth"
- From: Stephen Davis <email@hidden>
- Date: Thu, 21 Jun 2007 00:01:31 -0700
On Jun 20, 2007, at 10:45 PM, Andy Peters wrote:
Maybe somebody here can explain.
The System Profiler tells me that my Mac mini's built-in audio
supports two formats: PCM 16 and PCM 24.
What's interesting is that for each, we're told that there's both a
Bit Depth and a Bit Width. For PCM 16, both of these are 16, and
for PCM 24, the Depth is 24 and the Width is 32.
My question: WTF? How can it be both "wide" and "deep"?
For the "depth = 24, width = 32" case, this means there are 24 bits
of resolution for the audio data BUT the 24 bits are arranged in
memory as 32-bit words, mostly to make things more efficient for a
variety of operations. For example, packing the samples in this
fashion makes the conversion to other formats (e.g. floats) more
efficient: accessing 3-byte "words" is inefficient for any processor
so expanding the 3 bytes into a 4-byte "word" helps immensely. While
24-bit resolution is great for audio quality, it's kinda sucky in
other respects. :-)
hth,
stephen
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden