Date: Tue, 24 May 2005 14:00:26 -0700 From: Kris Kennaway <kris@obsecurity.org> To: Bohdan Horst <nexus@hoth.amu.edu.pl> Cc: freebsd-stable@freebsd.org, Kris Kennaway <kris@obsecurity.org> Subject: Re: Performance of 4.x vs 5.x (nbench results) Message-ID: <20050524210026.GA28898@xor.obsecurity.org> In-Reply-To: <20050524205609.GC35326@aristo> References: <20050524193117.GA35326@aristo> <20050524193707.GA11906@xor.obsecurity.org> <20050524201701.GB35326@aristo> <20050524202118.GB28257@xor.obsecurity.org> <20050524205609.GC35326@aristo>
next in thread | previous in thread | raw e-mail | index | archive | help
--jI8keyz6grp/JLjh Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, May 24, 2005 at 10:56:09PM +0200, Bohdan Horst wrote: > On Tue, May 24, 2005 at 01:21:19PM -0700, Kris Kennaway wrote: > > On Tue, May 24, 2005 at 10:17:01PM +0200, Bohdan Horst wrote: > > > On Tue, May 24, 2005 at 12:37:08PM -0700, Kris Kennaway wrote: > > > >=20 > >=20 > > OK, it might be trustable..but you're still testing gcc 2/gcc 3 as > > pointed out by another poster. If you use the same 4.x binaries on > > both machines it might be better. >=20 >=20 > NUMERIC SORT: 4.11=3D 499.52 494.31 5.4=3D 451.26 449.7 5.4O=3D 491.6= 6 492.53 > STRING SORT: 4.11=3D 24.13 24.11 5.4=3D 25.286 25.294 5.4O=3D 24.20= 6 24.213 > BITFIELD(e+8) 4.11=3D 1.133 1.1319 5.4=3D 1.1239 1.1239 5.4O=3D 1.135= 8 1.1358 > FP EMULATION 4.11=3D 18.951 18.755 5.4=3D 25.044 25.029 5.4O=3D 18.81= 8 18.818 > FOURIER 4.11=3D 5361 5356.3 5.4=3D 5076.3 5076.3 5.4O=3D 5577 = 5577 > ASSIGNMENT 4.11=3D 5.4795 5.4469 5.4=3D 5.8986 5.8968 5.4O=3D 5.481= 5.4763 > IDEA 4.11=3D 595.23 592.13 5.4=3D 660.65 660.65 5.4O=3D 593.9= 1 593.83 > HUFFMAN 4.11=3D 332.47 330.91 5.4=3D 409.06 408.95 5.4O=3D 333.1= 6 333.16 > NEURAL NET 4.11=3D 7.0374 6.9882 5.4=3D 7.2976 7.2955 5.4O=3D 7.033= 7.0201 > LU DECOMP. 4.11=3D 225.94 226.76 5.4=3D 211.56 210.84 5.4O=3D 219.4= 1 220.15 > original: > INTEGER INDEX 4.11=3D 12.305 12.238 5.4=3D 13.409 13.400 5.4O=3D 12.27= 5 12.277 > FLOATING-POINT 4.11=3D 9.309 9.296 5.4=3D 9.052 9.041 5.4O=3D 9.339= 9.344 > new: > MEMORY INDEX 4.11=3D 3.321 3.313 5.4=3D 3.448 3.448 5.4O=3D 3.328= 3.327 > INTEGER INDEX 4.11=3D 2.895 2.873 5.4=3D 3.271 3.268 5.4O=3D 2.878= 2.880 > FLOATING-POINT 4.11=3D 5.163 5.156 5.4=3D 5.020 5.014 5.4O=3D 5.180= 5.182 >=20 > (5.4O =3D=3D 4.11 binary on 5.4R) >=20 > almost identical speed :) Yeah, that's pretty much expected since these are userland benchmarks (but still good to confirm!) Kris --jI8keyz6grp/JLjh Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (FreeBSD) iD8DBQFCk5XqWry0BWjoQKURAiE+AKD0OOLMH6MqAQ5T4lN5m307WTDOzQCg10TZ O0HveLxsmuGwp/LpgU6YO5I= =W9P9 -----END PGP SIGNATURE----- --jI8keyz6grp/JLjh--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050524210026.GA28898>