Date: Thu, 15 Apr 2010 10:50:44 +0900 (JST) From: Maho NAKATA <chat95@mac.com> To: amvandemore@gmail.com Cc: alc@freebsd.org, alan.l.cox@gmail.com, freebsd-stable@freebsd.org, als@modulus.org, avg@freebsd.org Subject: Re: How to reproduce: Re: Only 70% of theoretical peak performance on FreeBSD 8/amd64, Corei7 920 Message-ID: <20100415.105044.276219111671413416.chat95@mac.com> In-Reply-To: <n2o6201873e1004141047t97d89cb0o2688fae1875eae08@mail.gmail.com> References: <m2y6201873e1004140945n855c8800we9baced2e293f270@mail.gmail.com> <4BC5F289.7020408@freebsd.org> <n2o6201873e1004141047t97d89cb0o2688fae1875eae08@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi Adam, From: Adam Vande More <amvandemore@gmail.com> Subject: Re: How to reproduce: Re: Only 70% of theoretical peak performance on FreeBSD 8/amd64, Corei7 920 Date: Wed, 14 Apr 2010 12:47:31 -0500 > Since this is a full fledged desktop environment, 90% utilization seems > pretty good. No, I don't think so. Even on Ubuntu, mine is running on a full desktop environment, GotoBLAS's performance is about 95% using dgemm. dgemm on Linux is lot more stabler than FreeBSD and clearly faster. on Ubuntu $ ./dgemm n: 3000 time : 51.180000 or 12.795519 Mflops : 42216.341930 n: 3100 time : 56.280000 or 14.261719 Mflops : 41791.049205 n: 3200 time : 61.350000 or 15.631380 Mflops : 41939.023080 n: 3300 time : 67.790000 or 17.247202 Mflops : 41685.474166 n: 3400 time : 73.800000 or 18.471321 Mflops : 42569.300032 n: 3500 time : 81.480000 or 20.781936 Mflops : 41273.585044 n: 3600 time : 88.170000 or 22.816965 Mflops : 40907.246233 n: 3700 time : 95.210000 or 23.864101 Mflops : 42462.684969 n: 3800 thanks -- Nakata Maho http://accc.riken.jp/maho/ , http://ja.openoffice.org/ Nakata Maho's PGP public keys: http://accc.riken.jp/maho/maho.pgp.txt
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100415.105044.276219111671413416.chat95>