From owner-freebsd-hackers Sun Jan 28 20:31:34 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id UAA27820 for hackers-outgoing; Sun, 28 Jan 1996 20:31:34 -0800 (PST) Received: from fw.ast.com (fw.ast.com [165.164.6.25]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id UAA27813 for ; Sun, 28 Jan 1996 20:31:22 -0800 (PST) Received: from nemesis by fw.ast.com with uucp (Smail3.1.29.1 #2) id m0tgl7s-000859C; Sun, 28 Jan 96 22:23 CST Received: by nemesis.lonestar.org (Smail3.1.27.1 #20) id m0tgl3R-000CIQC; Sun, 28 Jan 96 22:18 WET Message-Id: Date: Sun, 28 Jan 96 22:18 WET To: hackers@freefall.freebsd.org From: uhclem@nemesis.lonestar.org (Frank Durda IV) Sent: Sun Jan 28 1996, 22:18:57 CST Subject: Re: Good news -- pipe stuff Cc: uhclem@nemesis.lonestar.org Sender: owner-hackers@FreeBSD.ORG Precedence: bulk [4]Terry says: [4]That said, if it didn't show up for you, Larry McVoy has a well known [4]Linux bias, but I would guess it was his enthusiasm, not intellectual [4]dishonesty, that caused him to ignore the hardware differences in his [4]paper. An interesting point for people with conspiratorial slants on things: I wrote to the authors of the paper (Tuesday of the conference) noting that they complained FreeBSD didn't support a particular piece of hardware, but that in 2.0.5 and later that hardware *was* supported. So why did it get listed in the paper as unsupported? The reply I got within only a few hours indicated that the statement in the paper was a "holdover" from the FreeBSD testing done using 1.1.5.1, and that they didn't notice the change in the 2.0.5 release notes, etc! This seemed pretty sloppy work for a professional paper. Makes you wonder how many of the other FreeBSD results are really from 1.1.5.1 instead of 2.0.5, as claimed. Oh, and with all the systems on the same drive, I wonder which system was the one whose preferred disk geometry was forced on everybody else and which OS got what part of the drive, (since transfers from outer cylinders are faster on modern drives with more sectors per seek)? I used to play the benchmark game as part of my real life job - I know a few of the tricks. "Sharing a drive among the test subjects for *fairness*" is completely bogus and can be used to give any system an edge just by placing them in the right part of the media. Frank Durda IV |"The Knights who say "LETNi" or uhclem%nemesis@rwsystr.nkn.net | demand... A SEGMENT REGISTER!!!" ^------(this is the fastest route)|"A what?" or ...letni!rwsys!nemesis!uhclem |"LETNi! LETNi! LETNi!" - 1983