Date: Wed, 13 Dec 2000 22:47:36 -0600 (CST) From: Mike Meyer <mwm@mired.org> To: kstewart@urx.com Cc: jbiquez@icsmx.com, questions@FreeBSD.ORG Subject: Re: advice with old equipment. Message-ID: <14904.20712.193427.889148@guru.mired.org> In-Reply-To: <3A381379.9F464204@urx.com> References: <14904.1150.593498.981432@guru.mired.org> <3A381379.9F464204@urx.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Kent Stewart <kstewart@urx.com> types: > Mike Meyer wrote: > > Kent Stewart <kstewart@urx.com> types: > > > Jorge Biquez wrote: > > > > 1 Pentium PRO DOUBLE Processor 200 Mhz. 256 RAM.2 Disks of 4 GB each one. > > > > 1 Pentium II 350 Mhz. 128 RAM. 2 Disks of 4 GB each one. > > > > My question is. What would be your advice on which machine will perform > > > > better with FreeBSD? and Why? > > > Well, it depends on the P-II 350. If you have one with PC-100 memory, > > > it can out perform the dual machine by 50% because of the memory > > > bandwidth. The rule of 1.8 times the cpu speed for duals would have > > > the PPro only out performing a P-II 350 with PC-66 memory. If you need > > > parallel processes, then the dual machine can be faster but I can't > > > guess when. > > Unless I've been misinformed, the PIIs internal cache is only half > > speed, wherease the PPro is full speed. So for memory access that are > > cache hits, the PPro will give you 200MHz speed, but the PII will only > > get 175MHz. > So, from a cache standpoint it would be an advantage with the PPro. > One of the effects I see with the Celeron and P-II & III's is that > Setiathome runs about 1.5 times faster on the non-Celeron cpu's > because of the memory. A Celeron is also a full-speed cache. There's another factor to consider - cache size. My timings on PIIIs (256K full speed cache, 100MHz system bus) and PII Xeons (512K full speed cache, 100MHz system bus) were linear with CPU speed. Using a 300MHz Celeron - with *no* cache and a 66MHz system bus - took two to three times as long as 400MHz PII Xeons. I see PIIs, PIIIs and PIVs with 256K or 512K of cache. The first Celerons had no cache, and the latest have 128K. Xeon's start at 512K and go up to 2MB. More cache means more cache hits, which means more speed - but how much more is dependent on the application. > I also > thought that the PPro's memory would only be PC-66. So the end > comparison could be all over the place depending on what you are > doing. That's the bottom line. The only way to tell how well your application will run on a system is to try it. Salespeople love CPU clock speed - it gives them a single number to feed to consumers to make their product look good. The reality is *much* more complicated, though. My 400MHz PII/Xeon blows the doors off my 500MHz K6-2. But a 500MHz PIII w/512K of cache would outperform the Xeon - and the pricing on the chips reflects that. > A buildworld on the Celeron 433 is a little bit faster than a > buildworld on a Abit BX6 rev 2 with a P-II 400. The Abit board is > about 10% slower than a SuperMicro with the same speed cpu. Never have > figured that one out. What's the cache size on the two P II systems? 256K vs. 512K, maybe? > > I've been told by reliable people that the PPro's benchmark at 190% of > > the speed of PIIs at same clockrate because of this. This implies that > > the cache hit rate is pretty high for the application in question, > > though. > I'll add this to my trivia database for benchmarking. If you publish results, please let us know! <mike -- Mike Meyer <mwm@mired.org> http://www.mired.org/home/mwm/ Independent WWW/Unix/FreeBSD consultant, email for more information. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?14904.20712.193427.889148>