From mnovak@mindspring.com Tue Jul 12 14:52:46 1994 Path: ashpoly!bigsur!pacrim!krgate!eunha!mindspring!mnovak From: mnovak@mindspring.com (Min-jun Novak) Newsgroups: comp.dsp Subject: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Jul12.145246.8761@mindspring.com> Date: Tue, 12 Jul 1994 14:52:46 GMT Organization: MindSpring Enterprises NNTP-Posting-Host: mindspring.com X-Newsreader: Netscape 1.1N 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. Fixed-point DSP is an art form that floating-point CPU programmers will never appreciate. -- Min-jun Novak SupraFAXModem 28.8 @ 9600 Telebit bps "The hatch is open. Is anyone listening?" From mhernandez@mindspring.com Thu Jul 14 09:35:46 1994 Path: ashpoly!cascade!transat!eunet-fr!teletel!mindspring!mhernandez From: mhernandez@mindspring.com (Min-jun Hernández) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Jul14.093546.9527@mindspring.com> Date: Thu, 14 Jul 1994 09:35:46 GMT References: <1994Jul12.145246.8761@mindspring.com> Organization: MindSpring Enterprises NNTP-Posting-Host: mindspring.com X-Newsreader: WinVN 0.99.5 Lines: 12 In article <1994Jul12.145246.8761@mindspring.com>, mnovak@mindspring.com (Min-jun Novak) 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. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! If you are posting from AOL, please read RFC 1036 before posting again. -- === Min-jun Hernández === Dept of Electrical Engineering, MindSpring Enterprises "Hardware is temporary. The signal survives." From kgarcia@pipeline.com Fri Jul 15 13:22:46 1994 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!pipeline!kgarcia From: kgarcia@pipeline.com (Kenji García) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Jul15.132246.9025@pipeline.com> Date: Fri, 15 Jul 1994 13:22:46 GMT References: <1994Jul12.145246.8761@mindspring.com> Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: WinVN 0.99.5 Lines: 11 In article <1994Jul12.145246.8761@mindspring.com>, mnovak@mindspring.com (Min-jun Novak) 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. If you are posting from AOL, please read RFC 1036 before posting again. -- Kenji García -- The Pipeline Internet Network "There are 14 nodes and only 10 countries. Count again." From aito@stanford.edu Sat Jul 16 15:21:46 1994 Path: ashpoly!bigsur!pacrim!jpgate!kamome!stanford!aito From: aito@stanford.edu (Anton Ito) Newsgroups: comp.dsp Subject: Re: [SPECULATION] Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Jul16.152146.2323@stanford.edu> Date: Sat, 16 Jul 1994 15:21:46 GMT References: <1994Jul12.145246.8761@mindspring.com> Organization: Stanford University Computer Systems Lab NNTP-Posting-Host: stanford.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1994Jul12.145246.8761@mindspring.com>, mnovak@mindspring.com (Min-jun Novak) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. 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. -- Anton Ito -- Stanford University Computer Systems Lab "5 sigma is a discovery. 7 is a question."