From elaurent@unam.mx Fri Aug 18 10:55:31 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!unam!elaurent From: elaurent@unam.mx (Elaine Laurent) Newsgroups: comp.dsp Subject: Question: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Aug18.105531.8994@unam.mx> Date: Fri, 18 Aug 1995 10:55:31 GMT Organization: Universidad Nacional Autónoma de México (DGSCA) NNTP-Posting-Host: unam.mx X-Newsreader: WinVN 0.99.5 Lines: 11 Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. 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. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- Elaine Laurent *PLONK* filter active for trolls. Sun SparcStation 20 - NetBSD 1.0 From dlaurent@demon.co.uk Sat Aug 19 07:37:31 1995 Path: ashpoly!cascade!bigsur!demon!dlaurent From: dlaurent@demon.co.uk (Dave Laurent) Newsgroups: comp.dsp Subject: Re: Question: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Aug19.073731.9422@demon.co.uk> Date: Sat, 19 Aug 1995 07:37:31 GMT References: <1995Aug18.105531.8994@unam.mx> Organization: Demon Internet Services UK NNTP-Posting-Host: demon.co.uk X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1995Aug18.105531.8994@unam.mx>, elaurent@unam.mx (Elaine Laurent) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Unsubscribe me from this list! (Wait, this is Usenet, never mind). -- Dave Laurent -- Demon Internet Services UK "Tape doesn't lie at 02:14. Ears do." From mhughes@uni-hamburg.de Sun Aug 20 15:16:31 1995 Path: ashpoly!cascade!transat!sunet!nordrelay!uni-hamburg!mhughes From: mhughes@uni-hamburg.de (Min-jun Hughes) Newsgroups: comp.dsp Subject: Re: Question: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1995Aug20.151631.7202@uni-hamburg.de> Date: Sun, 20 Aug 1995 15:16:31 GMT References: <1995Aug18.105531.8994@unam.mx> Organization: Universität Hamburg Fachbereich Informatik NNTP-Posting-Host: uni-hamburg.de X-Newsreader: WinVN 0.99.5 Lines: 12 In article <1995Aug18.105531.8994@unam.mx>, elaurent@unam.mx (Elaine Laurent) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. > 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. Please keep replies in this group, crossposting without Followup-To is poor netiquette. -- Min-jun Hughes Opinions expressed are solely my own and not those of Universität Hamburg Fachbereich Informatik. ------------------------------------------------------------