Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 21 Mar 2001 12:28:05 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 21 Mar 2001 12:27:59 -0500 Received: from ns.suse.de ([213.95.15.193]:36879 "HELO Cantor.suse.de") by vger.kernel.org with SMTP id ; Wed, 21 Mar 2001 12:26:51 -0500 Date: Wed, 21 Mar 2001 18:34:40 +0100 (CET) From: egger@suse.de Reply-To: egger@suse.de Subject: Re: Only 10 MB/sec with via 82c686b chipset? To: chromi@cyberspace.org Cc: linux-kernel@vger.kernel.org In-Reply-To: MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Message-Id: <20010321182613.85AF354D6@Nicole.muc.suse.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On 21 Mar, Jonathan Morton wrote: > The "blue and white" PowerMac G3 and certain early PowerMac G4s used a > 66MHz PCI card for graphics in lieu of proper AGP. 66MHz PCI is used > in certain high-end workstations, as well, but it's not normally found > on consumer-level devices. > Look at 'lspci -vvv' output for the "66MHz" flag on the devices listed > there - all the ones in my Duron system leave it unset, except for my > (very recent and pretty nippy) SCSI controller and (AGP) video card. Well, I wasn't talking about 66Mhz nor about 64bit cards but rather normal consumer cards which are specified for 33Mhz. > That said, *most* PCI devices don't like being overclocked, and it's > not well known that pushing the system bus also pushes the PCI and ISA > buses in the same manner. A friend of mine had *severe* locking > problems with his system when he inserted his cheap SCSI adapter into > his overclocked machine, even though the other cards handled it OK > (relatively speaking - I'm not convinced). I don't know how far he'd > overclocked it, but 37MHz kinda rings true. Trying to enhance a systems performance by overclocking the bus is about the most stupid thing one can do as one ANY of the connected devices memory/CPU/chipset/PCI devices/AGP cards (just to name a few) have a high probability of failing and all those probabilites factor up which basically means that the system is in a unpredictable state which is never acceptable for any serious kind of system. Better leave the overclocking of busses to kids who are bored by their live. -- Servus, Daniel - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/