Date: Thu, 13 Feb 1997 20:28:30 +1100 (EST) From: "Daniel O'Callaghan" <danny@panda.hilink.com.au> To: "Michael L. VanLoon -- HeadCandy.com" <michaelv@MindBender.serv.net> Cc: William Maddox <maddox@CS.Berkeley.EDU>, freebsd-hardware@freebsd.org Subject: Re: Pentium vs. Pentium Pro Message-ID: <Pine.BSF.3.91.970213202712.427J-100000@panda.hilink.com.au> In-Reply-To: <199702130836.AAA16651@MindBender.serv.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 13 Feb 1997, Michael L. VanLoon -- HeadCandy.com wrote: > > >I'm putting together another FreeBSD system and am trying to determine > >whether a Pentium Pro is worth the extra expense.. I would appreciate > >any pointers to meaningful benchmarks on the relative performance of > >the Pentium 133 and Pentium 166 vs. the Pentium Pro 200 and Pentium > >Pro 180 under FreeBSD. The system will be used primarily for program > >development, i.e., lots of compiling. > > My experience, doing benchmarks like make worlds, and large software > builds, is that a P6 200/256K is roughly twice as fast as a P5 133MHz. > (When doing comparisons, remember that a P5 200 is NOT 200/133 faster > than a P5 133, due to bus saturation.) > > People never seem to listen, or never seem to learn. DON'T BUY 150's > AND 180's. A 180MHz part is slower than a 166MHz part (either P5 or > P6), in almost all normal usage. Similarly, a 150 will usually be > slower than a 133. A P6 166MHz with 512K cache will especially be > faster than a P6 180 with 256K cache. > > Remember, chips that run at even multiples of 30 only run their memory > bus at 60MHz, and their PCI bus at 30MHz. Chips that are multiples of > ~33 run their memory bus at ~67MHz, and their PCI bus at ~33MHz. Has anyone tried out the new 686-150/P200+ (doubled 75MHz) chips. Suitable motherboards seem to be scarce. Danny
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.91.970213202712.427J-100000>