From bkuznetsov@kth.se Sat Feb 03 17:19:32 1996 Path: ashpoly!cascade!transat!eunet-de!hansebox!kth!bkuznetsov From: bkuznetsov@kth.se (Brian Kuznetsov) Newsgroups: comp.dsp Subject: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb03.171932.6376@kth.se> Date: Sat, 03 Feb 1996 17:19:32 GMT Organization: Royal Institute of Technology, Stockholm NNTP-Posting-Host: kth.se X-Newsreader: NewsWatcher 2.0b24 Lines: 11 Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. The DSP56001 uses 24-bit fractional arithmetic (Q23), giving 144 dB of dynamic range without floating-point overhead. 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. -- Brian Kuznetsov "5 sigma is a discovery. 7 is a question." Atari TT030 / System 7.5.1 From dfournier@netcom.com Mon Feb 05 17:38:32 1996 Path: ashpoly!bigsur!pacrim!jpgate!kamome!netcom!dfournier From: dfournier@netcom.com (Dave Fournier) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb05.173832.1967@netcom.com> Date: Mon, 05 Feb 1996 17:38:32 GMT References: <1996Feb03.171932.6376@kth.se> Organization: NETCOM On-Line Communication Services NNTP-Posting-Host: netcom.com X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1996Feb03.171932.6376@kth.se>, bkuznetsov@kth.se (Brian Kuznetsov) writes: > Porting stardust.c to the NeXTcube and Atari Falcon DSP subsystems. > 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 Fournier === Dept of Electrical Engineering, NETCOM On-Line Communication Services "The carrier frequency does not drift by accident." From eperez@ox.ac.uk Tue Feb 06 15:35:32 1996 Path: ashpoly!cascade!transat!eunet-fi!relcom!kiae!ox!eperez From: eperez@ox.ac.uk (Elaine Pérez) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb06.153532.2105@ox.ac.uk> Date: Tue, 06 Feb 1996 15:35:32 GMT References: <1996Feb03.171932.6376@kth.se> <1996Feb05.173832.1967@netcom.com> Organization: Oxford University Computing Laboratory NNTP-Posting-Host: ox.ac.uk X-Newsreader: NewsWatcher 2.0b24 Lines: 8 In article <1996Feb05.173832.1967@netcom.com>, dfournier@netcom.com (Dave Fournier) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! -- Elaine Pérez Tape trading list upon request. SASE required for international mail. From fkim@unam.mx Wed Feb 07 16:09:32 1996 Path: ashpoly!cascade!bigsur!unam!fkim From: fkim@unam.mx (Fiona Kim) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb07.160932.2755@unam.mx> Date: Wed, 07 Feb 1996 16:09:32 GMT References: <1996Feb03.171932.6376@kth.se> <1996Feb05.173832.1967@netcom.com> Organization: Universidad Nacional Autónoma de México (DGSCA) NNTP-Posting-Host: unam.mx X-Newsreader: TIN [version 1.2 PL2] Lines: 10 In article <1996Feb05.173832.1967@netcom.com>, dfournier@netcom.com (Dave Fournier) writes: > Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. -- Fiona Kim -- Universidad Nacional Autónoma de México (DGSCA) "The hatch is open. Is anyone listening?" From rmoreau@hanse.de Wed Feb 07 21:05:32 1996 Path: ashpoly!bigsur!pacrim!krgate!eunha!hanse!rmoreau From: rmoreau@hanse.de (Raj Moreau) Newsgroups: comp.dsp Subject: Re: Fixed-point Q15 arithmetic on Motorola DSP56001 vs IEEE floating point Message-ID: <1996Feb07.210532.3724@hanse.de> Date: Wed, 07 Feb 1996 21:05:32 GMT References: <1996Feb03.171932.6376@kth.se> <1996Feb05.173832.1967@netcom.com> <1996Feb07.160932.2755@unam.mx> Organization: Hanse-Netzwerk Hamburg NNTP-Posting-Host: hanse.de X-Newsreader: NewsWatcher 2.0b24 Lines: 10 In article <1996Feb07.160932.2755@unam.mx>, fkim@unam.mx (Fiona Kim) writes: > Spoken like someone who has never tried doing real-time multichannel filtering on an x86 processor with no pipelined MAC unit! > Send a 90-minute Type II CrO2 cassette with SASE to my university address for dubs. Beautiful assembly code! The 24-bit word size of the 56001 is tailor-made for high-precision audio filtering. -- Raj Moreau Hayes Optima 288 @ ISDN 64k bps "5 sigma is a discovery. 7 is a question."