From owner-freebsd-hardware Thu Aug 22 20:14:05 1996 Return-Path: owner-hardware Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id UAA23598 for hardware-outgoing; Thu, 22 Aug 1996 20:14:05 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id UAA23593 for ; Thu, 22 Aug 1996 20:14:00 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id UAA13429; Thu, 22 Aug 1996 20:01:42 -0700 From: "Rodney W. Grimes" Message-Id: <199608230301.UAA13429@GndRsh.aac.dev.com> Subject: Re: ASUS SC200 SCSI card? To: msmith@atrad.adelaide.edu.au (Michael Smith) Date: Thu, 22 Aug 1996 20:01:42 -0700 (PDT) Cc: pjchilds@imforei.apana.org.au, se@zpr.uni-koeln.de, msmith@atrad.adelaide.edu.au, freebsd-hardware@FreeBSD.org In-Reply-To: <199608222359.JAA17181@genesis.atrad.adelaide.edu.au> from Michael Smith at "Aug 23, 96 09:29:19 am" X-Mailer: ELM [version 2.4ME+ PL11 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hardware@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > Rodney W. Grimes stands accused of saying: > > > > > > > > [ Discussion about hangs on 2.1.5-stable machine with dual ASUS SC200 > > > NCR810 PCI scsi controllers follows... may be dangerous to > > > mental health ] > > ... > > > > > 2. PCI latency was set to 80... Michael Smith suggested it be lower than > > > 32 to i've moved it to 20. > > > > Set it to exactly 32, no more, no less. ASUS and others have done > > 100's of hours of testing and this was found to be the best setting. > > Even with two 810's coming up for an opcode every us? I'd have > thought you'd want to allow for (max latency + one opcode fetch) < 1us > so that the second one didn't starve... Even with 6 of them.... the latency time just says if someone else wants the bus I can't use it for more than a total of 32 clocks, the NCR should be buffering enough of its opcodes for this to not be a problem. I've done a fair bit of testing with 4 NCR's, and so has Joe Greco and we have not seen any problems running 4 of them full bore with this setting. I have also done this with 2940's and a mixture of NCR/2940's and DEC 21x4x networks cards, still no problems. Video capture boards, well, now you start to get into the magical land of having to very carefully tune things to work optimally. Jim Lowe over at UWM is having better luck with NCR based systems than he is/was with 2940 based for some reason. It don't make since, and given that I don't have the Matrox Meteor hardware here to debug it with I have defered to him on that one. He specs all his video systems with the NCR... go figure! -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD