From owner-freebsd-hardware Fri Apr 26 09:55:49 1996 Return-Path: owner-hardware Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA18651 for hardware-outgoing; Fri, 26 Apr 1996 09:55:49 -0700 (PDT) Received: from falcon.tioga.com (root@falcon.tioga.com [205.146.65.5]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id JAA18644 for ; Fri, 26 Apr 1996 09:55:46 -0700 (PDT) Received: (from tbalfe@localhost) by falcon.tioga.com (8.6.12/8.6.12) id MAA20216; Fri, 26 Apr 1996 12:55:14 GMT Date: Fri, 26 Apr 1996 12:55:13 +0000 () From: Thomas J Balfe To: freebsd-hardware@freebsd.org Subject: AMD 5x86-133 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hardware@freebsd.org X-Loop: FreeBSD.org Precedence: bulk A few nights ago I built an AMD based FreeBSD box that people might be interested in. It is based upon the Biostar MB-8433UUD motherboard and AMD 5x86-133 processor. The motherboard and processor together were $200 from a local computer store, although I've seen the same combo for $167 in Computer Shopper. The motherboard features 256K L2 cache, three PCI slots, four ISA slots and a ZIF socket. It will take 128MB although we have only 16MB installed. The motherboard already has an IDE controller and parallel and serial ports (16550), so you don't have to buy a seperate controller. It features Award BIOS. If you have a spare case sitting around and need another machine, this may be the way to go. I stuck it in an old Zeos case that a friend had, as well as the IDE spare drives I had hanging around. (you bought them, might as well use 'em) nsieve.c was compiled on 2.1R with gcc-2.6.3 Sieve of Eratosthenes (Scaled to 10 Iterations) Version 1.2b, 26 Sep 1992 Array Size Number Last Prime Linear RunTime MIPS (Bytes) of Primes Time(sec) (Sec) 8191 1899 16381 0.023 0.023 73.7 10000 2261 19997 0.027 0.028 73.2 20000 4202 39989 0.055 0.061 67.1 40000 7836 79999 0.110 0.140 59.5 80000 14683 160001 0.220 0.300 56.2 160000 27607 319993 0.440 0.652 52.3 320000 52073 639997 0.879 1.375 50.1 640000 98609 1279997 1.758 2.996 46.4 1280000 187133 2559989 3.516 6.341 44.3 2560000 356243 5119997 7.033 13.163 43.1 Relative to 10 Iterations and the 8191 Array Size: Average RunTime = 0.032 (sec) High MIPS = 73.7 Low MIPS = 43.1 flops.c was compiled on 2.1R with gcc-2.6.3 FLOPS C Program (Double Precision), V2.0 18 Dec 1992 Module Error RunTime MFLOPS (usec) 1 -1.5632e-13 1.9762 7.0841 2 -1.0347e-13 1.3757 5.0883 3 -3.1197e-14 1.9411 8.7578 4 7.7938e-14 1.7007 8.8199 5 -3.2641e-14 3.5206 8.2373 6 -9.9920e-16 3.1383 9.2407 7 -5.5650e-11 2.6136 4.5914 8 2.7700e-14 3.2453 9.2441 Iterations = 8000000 NullTime (usec) = 0.0893 MFLOPS(1) = 5.8959 MFLOPS(2) = 6.6869 MFLOPS(3) = 8.0504 MFLOPS(4) = 9.0769 Right now the machine (t-rex.tioga.com) is my ftp server and cu-seeme reflector (the BSDI binary). I might build another machine similar to this one for a pop box, scsi of course. The above benchmarks are at ftp.nosc.mil in /pub/aburto So, for $200 or less and spare parts, you can have a decent machine. If you have seen these benchmarks already I apologize, but if not, then I hope this information comes in useful. ======================================================================== Thomas J Balfe tbalfe@tioga.com President http://www.tioga.com/ Tioga Communications, Inc 814-867-4770 ========================================================================