Re: NTLK Re: Questions, questions...

Date: Sat Apr 29 2000 - 09:09:29 CDT

On 28 Apr 2000, Stainless Steel Rat wrote:

> * <> on Fri, 28 Apr 2000
> | That's not "compression" in the digital data sense. And it is _not_
> | similar to MD in any way at all.
> Bit-rate reduction is one of many forms of compression.

When you are referring to digital data, "compression" has a specific
definition. You were suggesting that CDs used something similar to MD.
MD takes an _uncompressed_ CD-style data stream, and then performs lossy
compression based on a perceptual algorithm to remove any data that you
probably won't notice. Compression involves taking a data stream and
taking active actions to remove its data size. CDs do not do this. Of
course they don't have an infinite sampling rate. They have chosen a
sampling rate based on what the majority of people can hear, and based on
the practicalities of older technology.

Using a low bit rate does not make something compressed; taking a bit
stream and reducing its bit rate by eliminating "redundant" data is
compression. Minidiscs use compression to reduce the bit rate by a factor
of 5. CDs use a "low" but uncompressed sampling rate.

In fact, far from using bit-rate reduction, CDs actually store much more
data with a lot of extra redundancy and error correction codes in addition
to the raw byte stream.

There's some info on how CDs do error correction and all the wonderful
things that increase the bit rate:

> | Since CDs are sampled at 44.1kHz, there is no possible way that you'll
> | get a frequency higher than 22.5kHz.
> 22.05kHz. The phenomenon is called 'aliasing'.

Whoops, I missed the 0 key. I am fully aware of Nyquist's theorem, I just
make the odd typo.

> | They're not "compressed" to remove this; it's simply a physical
> | impossibility to record a higher sampling rate than that.
> Errors aside, how does this contradict anything I wrote? Data loss,
> bit-rate reduction, lossy compression, whatever you want to call it, it
> ammounts to the same thing. And why are we still discussing this on the

Yes, it does amount to the same thing, but Minidiscs do it and CDs don't.
CDs record at a specific bit rate and record that data fully onto the CD.
No reduction, compression, or data loss there...

> Newton mailing list?

It came up, and since there were a few rather incorrect statements made
(such as somebody else's claims that CDs were sampled at 44 samples/second
and records were higher than this....) I figured that correcting the
statements in the same forum they were made was appropriate.


This NewtonTalk Message brought to you by:

EVOTE.COM, the ESPN of Politics on the
Internet. Visit EVOTE.COM for all the latest
news on Campaign 2000!

Visit today!


Need Subscribe/Unsubscribe Info?


This archive was generated by hypermail 2b29 : Thu Jun 01 2000 - 00:00:07 CDT