From owner-freebsd-hackers Mon Dec 15 19:38:45 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id TAA17508 for hackers-outgoing; Mon, 15 Dec 1997 19:38:45 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from anlsun.ebr.anlw.anl.gov (anlsun.ebr.anlw.anl.gov [141.221.1.2]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id TAA17503 for ; Mon, 15 Dec 1997 19:38:41 -0800 (PST) (envelope-from cmott@srv.net) Received: from darkstar.home (ras520.srv.net [205.180.127.20]) by anlsun.ebr.anlw.anl.gov (8.6.11/8.6.11) with SMTP id UAA01398 for ; Mon, 15 Dec 1997 20:38:31 -0700 Date: Mon, 15 Dec 1997 20:37:58 -0700 (MST) From: Charles Mott X-Sender: cmott@darkstar.home To: FreeBSD Hackers Subject: Re: 3com 3c509 card In-Reply-To: <199712160244.NAA00629@word.smith.net.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Nope. They're both PIO cards, and the 'ed' driver spits all over the > 'ep' driver. Any inversions you may have seen would have been > environmental relating to the different behaviour of the various chips > relating to bus traffic. > > Yes, I've used both. Yes, I've seen > 1000K/sec out of an 'ed' card on > a 486/50. With modern processors, yes, the differences are less > evident, but I'll still take an NE2000 over a 509 anyday. Especially > when a single-chip card based on something like the RTL2019 costs less > than AUD$40, and "just works". > > mike > I've gotten 700 Kbytes/sec for sustained transfers on an NE 2000 clone on a 386/33 under 2.2.2. Performance under 2.1.x was 300 Kbytes/sec. Something radically improved in the 2.1->2.2 transition. I had thought it was the core networking code, but I am wondering if the driver improved also. Sometimes slow CPUs are good indicators of software improvements. Charles Mott