From mfournier@berkeley.edu Sun Dec 15 09:19:02 1996 Path: ashpoly!cascade!transat!eunet-de!hansebox!berkeley!mfournier From: mfournier@berkeley.edu (Mikhail Fournier) Newsgroups: comp.dsp Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Dec15.091902.7839@berkeley.edu> Date: Sun, 15 Dec 1996 09:19:02 GMT Organization: UC Berkeley Department of EECS NNTP-Posting-Host: berkeley.edu X-Newsreader: TIN [version 1.2 PL2] Lines: 11 Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. 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. -- Mikhail Fournier "A soul is not exclusive to biology." Atari TT030 / TOS 4.04 From dokonkwo@earthlink.net Mon Dec 16 12:51:02 1996 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!earthlink!dokonkwo From: dokonkwo@earthlink.net (Dave Okonkwo) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Dec16.125102.2533@earthlink.net> Date: Mon, 16 Dec 1996 12:51:02 GMT References: <1996Dec15.091902.7839@berkeley.edu> Organization: EarthLink Network NNTP-Posting-Host: earthlink.net X-Newsreader: WinVN 0.99.5 Lines: 10 In article <1996Dec15.091902.7839@berkeley.edu>, mfournier@berkeley.edu (Mikhail Fournier) writes: > Optimizing the Stardust Costas loop demodulator for the Motorola DSP56001 chip. > The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Dave Okonkwo "Hardware is temporary. The signal survives." 486DX4/100 (FreeBSD 2.0.5) / NEXTSTEP 3.3 From onilsson@panix.com Wed Dec 18 03:42:02 1996 Path: ashpoly!cascade!mxgate!coyoacan!panix!onilsson From: onilsson@panix.com (Olga Nilsson) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Dec18.034202.3182@panix.com> Date: Wed, 18 Dec 1996 03:42:02 GMT References: <1996Dec15.091902.7839@berkeley.edu> <1996Dec16.125102.2533@earthlink.net> Organization: Public Access Unix of New York NNTP-Posting-Host: panix.com X-Newsreader: Netscape 1.1N Lines: 11 In article <1996Dec16.125102.2533@earthlink.net>, dokonkwo@earthlink.net (Dave Okonkwo) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Real programmers write in C, not arcane DSP microcode. With Pentium chips hitting 100 MHz, dedicated DSPs are obsolete. Netscape 1.1 crashes when loading large preformatted text. Stick to rn or tin! -- Olga Nilsson *PLONK* filter active for trolls. PC-9821Ap2 (NEC) - NetBSD 1.0