Date: Tue, 16 Mar 1999 14:16:15 -0500 (EST) From: "John S. Dyson" <dyson@iquest.net> To: sas@schell.de (Sascha Schumann) Cc: andreas@klemm.gtn.com, freebsd-chat@freebsd.org Subject: Re: http://www.freebsd.org/~andreas/benches/index.html Message-ID: <199903161916.OAA08468@y.dyson.net> In-Reply-To: <19990316143700.A3066@schell.de> from Sascha Schumann at "Mar 16, 99 02:37:00 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
Sascha Schumann said: > On Mon, Mar 15, 1999 at 10:26:19PM +0100, Andreas Klemm wrote: > > On Mon, Mar 15, 1999 at 09:25:38PM +0100, Sascha Schumann wrote: > > > On Mon, Mar 15, 1999 at 07:10:27PM +0100, Andreas Klemm wrote: > > > > On Mon, Mar 15, 1999 at 02:36:02PM +0100, Sascha Schumann wrote: > > > > > > > > Perhaps I repeat the test with an 2.2 kernel.... > > > > > > Yes, and please stop spreading FUD. It looks rather narrow-minded to be > > > proud of such a comparison. > > > > Look at the new values ;-) Satisfied ?! ;-) > > > > -- > > Andreas Klemm http://www.FreeBSD.ORG/~andreas > > News : FreeBSD 3.1 SMP outperforms SuSE Linux 6.0 SMP by 230% !!! > > Well, the numbers have changed (^^^ not). But the method is still insane. You > are comparing not the same process, because there are numerous things which > differ (e.g. libraries, compiler version, toolchain). The only claim I would > derive from such a test would be that "you can build your programs faster on > FreeBSD." Sigh. > That is one of the things that we use FreeBSD at work for: compile engine. To build code for other OSes. I disagree with your usage of the word "insane". At worst the comparison is "ill-advised" to publicize. At best it shows the FreeBSD developers that FreeBSD is reasonably competitive with Linux in the SMP arena. In no way can such comparisons be used to proclaim that FreeBSD is always "better" than Linux. Geesh, the measurement doesn't even claim that FreeBSD is always faster, but when someone claims that it is sometimes faster (with disclosure of the circumstance, in good faith), there is a furor over it. It is that reason why I try to avoid talking about the relative performance of the OSes. In fact, by talking about the fact that FreeBSD is faster, it gives hints to the competition. It is NOT a good idea to help the competition, when that competition automatically flames or is inflammatory when a positive comparison about FreeBSD is announced. Firstly, your choice of the word "insane" is inflammatory. Secondly, if the libraries, compiler version and toolchain on Linux distributions are slower, then the distributors of such should consider that as a challenge. If it is the kernel that is slower, then that should be considered as a challenge (or at least a datapoint.) By telling the "competition" (which messages like what I am responding to show the nature of such) how and why their product is inferior in certain areas, it helps them. There is no value to help people with the assets of a project, who don't have good will towards that project. Luckily, the subtile side-effects of how software is written are not always easily derived from source code. (Source-code isn't necessarily the family jewels, and force-source distribution fanatics just don't get it. The more complex a project is, the less often that source is an easy way to determine the approaches taken in the design of a product.) -- John | Never try to teach a pig to sing, dyson@iquest.net | it makes one look stupid jdyson@nc.com | and it irritates the pig. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199903161916.OAA08468>