From jtanaka@cam.ac.uk Sun Oct 20 10:53:06 1996 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!cam!jtanaka From: jtanaka@cam.ac.uk (Jens Tanaka) Newsgroups: comp.dsp Subject: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Oct20.105306.9331@cam.ac.uk> Date: Sun, 20 Oct 1996 10:53:06 GMT Organization: University of Cambridge Computer Laboratory NNTP-Posting-Host: cam.ac.uk X-Newsreader: TIN [version 1.2 PL2] Lines: 11 Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. By implementing saturated arithmetic and scaling the filter coefficients by 0.5, the loop runs in 18 instruction cycles per sample. Source code in 56k assembly uploaded to ftp.ashford-poly.edu:/pub/telecom-archives/dsp56k/stardust_56k.asm. -- === Jens Tanaka === Dept of Electrical Engineering, University of Cambridge Computer Laboratory "Marcus is not one man." From soliveira@kiae.su Mon Oct 21 16:17:06 1996 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!kiae!soliveira From: soliveira@kiae.su (Sun-mi Oliveira) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Oct21.161706.9884@kiae.su> Date: Mon, 21 Oct 1996 16:17:06 GMT References: <1996Oct20.105306.9331@cam.ac.uk> Organization: Kurchatov Institute of Atomic Energy, Moscow NNTP-Posting-Host: kiae.su X-Newsreader: Netscape 1.1N Lines: 9 In article <1996Oct20.105306.9331@cam.ac.uk>, jtanaka@cam.ac.uk (Jens Tanaka) 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. -- Sun-mi Oliveira *PLONK* filter active for trolls. Sun 3/60 (SunOS 4.1.1) - NetBSD 1.0 From kramirez@kth.se Tue Oct 22 14:12:06 1996 Path: ashpoly!cascade!transat!ukc!fennimore!kth!kramirez From: kramirez@kth.se (Karen Ramírez) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Oct22.141206.1698@kth.se> Date: Tue, 22 Oct 1996 14:12:06 GMT References: <1996Oct20.105306.9331@cam.ac.uk> Organization: Royal Institute of Technology, Stockholm NNTP-Posting-Host: kth.se X-Newsreader: TIN [version 1.2 PL2] Lines: 9 In article <1996Oct20.105306.9331@cam.ac.uk>, jtanaka@cam.ac.uk (Jens Tanaka) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. > This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Karen Ramírez -- Royal Institute of Technology, Stockholm "The hatch is open. Is anyone listening?"