From sbrennan@u-tokyo.ac.jp Sat Jul 01 03:30:45 1995
Path: ashpoly!bigsur!pacrim!jpgate!kamome!u-tokyo!sbrennan
From: sbrennan@u-tokyo.ac.jp (Seo-yeon Brennan)
Newsgroups: comp.dsp
Subject: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point
Message-ID: <1995Jul01.033045.9339@u-tokyo.ac.jp>
Date: Sat, 01 Jul 1995 03:30:45 GMT
Organization: University of Tokyo Department of Information Science
NNTP-Posting-Host: u-tokyo.ac.jp
X-Newsreader: TIN [version 1.2 PL2]
Lines: 10

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

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

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

-- 
Seo-yeon Brennan -- University of Tokyo Department of Information Science
"There are 14 nodes and only 10 countries. Count again."

From divanov@mit.edu Sat Jul 01 14:00:45 1995
Path: ashpoly!cascade!transat!eunet-de!hansebox!mit!divanov
From: divanov@mit.edu (Diego Ivanov)
Newsgroups: comp.dsp
Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point
Message-ID: <1995Jul01.140045.2158@mit.edu>
Date: Sat, 01 Jul 1995 14:00:45 GMT
References: <1995Jul01.033045.9339@u-tokyo.ac.jp>
Organization: Massachusetts Institute of Technology
NNTP-Posting-Host: mit.edu
X-Newsreader: Netscape 1.1N
Lines: 10

In article <1995Jul01.033045.9339@u-tokyo.ac.jp>, sbrennan@u-tokyo.ac.jp (Seo-yeon Brennan) writes:
> Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems.
> By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample.

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

-- 
Diego Ivanov | divanov@mit.edu
PGP Key fingerprint: B919 7220 3F2C 8FA5
"Keep the channels open."

From jgonzalez@u-tokyo.ac.jp Sun Jul 02 15:03:45 1995
Path: ashpoly!bigsur!pacrim!krgate!eunha!u-tokyo!jgonzalez
From: jgonzalez@u-tokyo.ac.jp (Jens González)
Newsgroups: comp.dsp
Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point
Message-ID: <1995Jul02.150345.8477@u-tokyo.ac.jp>
Date: Sun, 02 Jul 1995 15:03:45 GMT
References: <1995Jul01.033045.9339@u-tokyo.ac.jp> <1995Jul01.140045.2158@mit.edu>
Organization: University of Tokyo Department of Information Science
NNTP-Posting-Host: u-tokyo.ac.jp
X-Newsreader: WinVN 0.99.5
Lines: 9

In article <1995Jul01.140045.2158@mit.edu>, divanov@mit.edu (Diego Ivanov) writes:
> Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering.

Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in.

-- 
=== Jens González ===
Dept of Electrical Engineering, University of Tokyo Department of Information Science
"When every node is whole, the channel opens."

From jkang@utexas.edu Tue Jul 04 12:29:45 1995
Path: ashpoly!cascade!transat!eunet-fr!teletel!utexas!jkang
From: jkang@utexas.edu (Ji-woo Kang)
Newsgroups: comp.dsp
Subject: Re: [INFO] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point
Message-ID: <1995Jul04.122945.8542@utexas.edu>
Date: Tue, 04 Jul 1995 12:29:45 GMT
References: <1995Jul01.033045.9339@u-tokyo.ac.jp> <1995Jul01.140045.2158@mit.edu> <1995Jul02.150345.8477@u-tokyo.ac.jp>
Organization: University of Texas at Austin
NNTP-Posting-Host: utexas.edu
X-Newsreader: NewsWatcher 2.0b24
Lines: 8

In article <1995Jul02.150345.8477@u-tokyo.ac.jp>, jgonzalez@u-tokyo.ac.jp (Jens González) writes:
> Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in.

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

-- 
Ji-woo Kang -- University of Texas at Austin
"Row 14. No joke."