[LAU] Pro Audio? OT rant.

Monty Montgomery xiphmont at gmail.com
Mon Dec 24 22:22:41 UTC 2012


On Mon, Dec 24, 2012 at 10:13 AM, Ricardus Vincente
<wizardofgosz at gmail.com> wrote:
> On 12/24/2012 03:59 AM, Ralf Mardorf wrote:
>
>> 44.1 was a compromise to get enough minutes on a CD, later when consumer
>> DAT was introduced it wasn't needed to take care about the length, so
>> they came with 48 KHz.
>
>  I was always told that 44.1 was chosen because they wanted to be able
> to reproduce signals up to 20K, but the other 2.05K of audio was needed
> for the low-pass filters of the day.

There was another more mundane reason than that.  44.1kHz allowed
engineers to store an even number of samples in either an NTSC or a
PAL scanline when using UMatic video recorders to record digital
audio.  Sony had supposedly even planned a line of products based on
the idea, though they never came to market.

The lowpass filters of the day weren't yet capable of cutting off
cleanly in only 2kHz (well, 4kHz, it's ok if the transition band
aliases) as they were all-analog and oversampling wasn't used yet, but
designers correctly decided that technology would catch up.  The CD
had also originally been 14 bits because that was the biggest DAC that
Phillips made at the time; Sony pushed hard for 16 bits [they had a 16
bit DAC in the works :-].

Several of the original CD players only played back 14 bits.

Anyway, like most engineering, the tradeoffs were interrelated and all
came into account.

Monty


More information about the Linux-audio-user mailing list