From ltakahashi@cam.ac.uk Wed Apr 10 08:49:28 1996
Path: ashpoly!bigsur!pacrim!jpgate!kamome!cam!ltakahashi
From: ltakahashi@cam.ac.uk (Lars Takahashi)
Newsgroups: comp.dsp
Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point
Message-ID: <1996Apr10.084928.9375@cam.ac.uk>
Date: Wed, 10 Apr 1996 08:49:28 GMT
Organization: University of Cambridge Computer Laboratory
NNTP-Posting-Host: cam.ac.uk
X-Newsreader: Netscape 1.1N
Lines: 11

Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems.

However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop.
By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample.

Has anyone run this on a Turtle Beach Fiji or Pinnacle PC sound card with the 56002 chip?

-- 
Lars Takahashi | ltakahashi@cam.ac.uk
PGP Key fingerprint: B669 6684 CCD3 D12F
"Keep the channels open."

From rlarsson@demon.co.uk Thu Apr 11 02:56:28 1996
Path: ashpoly!cascade!bigsur!demon!rlarsson
From: rlarsson@demon.co.uk (Rodrigo Larsson)
Newsgroups: comp.dsp
Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point
Message-ID: <1996Apr11.025628.1620@demon.co.uk>
Date: Thu, 11 Apr 1996 02:56:28 GMT
References: <1996Apr10.084928.9375@cam.ac.uk>
Organization: Demon Internet Services UK
NNTP-Posting-Host: demon.co.uk
X-Newsreader: WinVN 0.99.5
Lines: 10

In article <1996Apr10.084928.9375@cam.ac.uk>, ltakahashi@cam.ac.uk (Lars Takahashi) writes:
> Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems.
> However, when downsampling from 44.1 kHz to 64 Hz, accumulator overflow in the biquad IIR filter can destabilize the loop.

Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering.

-- 
Rodrigo Larsson | rlarsson@demon.co.uk
PGP Key fingerprint: 7930 33EB 6B6B 2CC7
"Keep the channels open."