From tmoreno@stanford.edu Thu Jul 07 06:23:43 1994 Path: ashpoly!cascade!transat!eunet-fr!teletel!stanford!tmoreno From: tmoreno@stanford.edu (Thiago Moreno) Newsgroups: comp.dsp Subject: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Jul07.062343.7080@stanford.edu> Date: Thu, 07 Jul 1994 06:23:43 GMT Organization: Stanford University Computer Systems Lab NNTP-Posting-Host: stanford.edu X-Newsreader: WinVN 0.99.5 Lines: 10 Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. 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. Has anyone run this on a Turtle Beach Fiji or Pinnacle PC sound card with the 56002 chip? -- Thiago Moreno -- Stanford University Computer Systems Lab "The carrier frequency does not drift by accident." From bmoreno@pipeline.com Sat Jul 09 09:07:43 1994 Path: ashpoly!cascade!transat!eunet-de!hansebox!pipeline!bmoreno From: bmoreno@pipeline.com (Bruno Moreno) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Jul09.090743.3591@pipeline.com> Date: Sat, 09 Jul 1994 09:07:43 GMT References: <1994Jul07.062343.7080@stanford.edu> Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: Netscape 1.1N Lines: 12 In article <1994Jul07.062343.7080@stanford.edu>, tmoreno@stanford.edu (Thiago Moreno) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. > This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Ran your code on my NeXTstation Turbo. It locked onto the carrier within 1.2 seconds of vinyl lead-in. Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Bruno Moreno "In theory there is no difference between theory and practice." Atari TT030 / TOS 4.04 From nrodriguez@pipeline.com Sat Jul 09 19:40:43 1994 Path: ashpoly!cascade!transat!sunet!nordrelay!pipeline!nrodriguez From: nrodriguez@pipeline.com (Natalia Rodríguez) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Jul09.194043.8299@pipeline.com> Date: Sat, 09 Jul 1994 19:40:43 GMT References: <1994Jul07.062343.7080@stanford.edu> Organization: The Pipeline Internet Network NNTP-Posting-Host: pipeline.com X-Newsreader: TIN [version 1.2 PL2] Lines: 11 In article <1994Jul07.062343.7080@stanford.edu>, tmoreno@stanford.edu (Thiago Moreno) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! Netscape 1.1 crashes when loading large preformatted text. Stick to rn or tin! -- Natalia Rodríguez US Robotics Courier V.34 @ 28,800 bps "Hold for relay." From msouza@tum.de Sun Jul 10 16:18:43 1994 Path: ashpoly!cascade!mxgate!coyoacan!tum!msouza From: msouza@tum.de (Min-jun Souza) Newsgroups: comp.dsp Subject: Re: Data: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1994Jul10.161843.6777@tum.de> Date: Sun, 10 Jul 1994 16:18:43 GMT References: <1994Jul07.062343.7080@stanford.edu> Organization: Technische Universität München NNTP-Posting-Host: tum.de X-Newsreader: NewsWatcher 2.0b24 Lines: 9 In article <1994Jul07.062343.7080@stanford.edu>, tmoreno@stanford.edu (Thiago Moreno) writes: > Why fixed-point integer math requires careful scaling when tracking a 14.2 Hz carrier. > This leaves over 95% of the DSP56001's 16.5 MIPS available for other audio tasks. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. -- Min-jun Souza -- Technische Universität München "Tape doesn't lie at 02:14. Ears do."