From alsa-devel-owner@alsa.jcu.cz  Thu Jan 21 13:16:51 1999
Received: from ife.ee.ethz.ch (ife-fast.ee.ethz.ch [129.132.24.193])
	by marvin.jcu.cz (8.9.1a/8.9.1) with ESMTP id NAA21174
	for <alsa-devel@alsa.jcu.cz>; Thu, 21 Jan 1999 13:16:48 +0100
Received: from ife.ee.ethz.ch (eldrich.ee.ethz.ch [129.132.24.203])
	by ife.ee.ethz.ch (8.8.5/8.8.5) with ESMTP id NAA26427
	for <alsa-devel@alsa.jcu.cz>; Thu, 21 Jan 1999 13:16:48 +0100 (MET)
Message-ID: <36A71AB0.88D85FC@ife.ee.ethz.ch>
Date: Thu, 21 Jan 1999 13:16:48 +0100
From: Thomas Sailer <sailer@ife.ee.ethz.ch>
MIME-Version: 1.0
To: alsa-devel@alsa.jcu.cz
Subject: Re: hi-res pcm sound support/IEEE floats
References: <01BE44CC.41B12C60@MUSE           >
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: alsa-devel@alsa.jcu.cz
Sender: alsa-devel-owner@alsa.jcu.cz
Precedence: list

"Richard W.E. Furse" wrote:

> ARGUMENT 3 (The Done Thing)
> 
> Last time I looked at DSPs etc, they also used floats (IEEE floats)
> internally. I believe most of the Motorola chips do, the Intel FPUs and MMX

Wrong. Motorola's only DSP that could do FP in hardware was the 96000,
and AFAIK it is no longer available. Motorolas popular DSPs for Audio
are the 56000 and 56300, both 24bit fixed point machines (with 56bit
accumulators).

Also wrong is that MMX uses FP, MMX only uses very small integers,
up to 16bits or so. Therefore, MMX is mostly useless for anything
else than a few image processing algorithms.

> 32bit float = 24bit(linear) + 8bit(exponent). Best of all worlds, and
> probably what the DSP on your sound card is using anyway.

Show me one single soundcard (below 500$) that can do _anything_
with floats.

Tom

