Re: your mail
- To: mathgroup at smc.vnet.net
- Subject: [mg2729] Re: your mail
- From: Tim Mooney <mooney at dogbert.cc.ndsu.NoDak.edu>
- Date: Sat, 9 Dec 1995 23:01:32 -0500
When we last communicated, mathgroup-adm at smc.vnet.net wrote: )In article <49dp82$8a6 at dragonfly.wri.com>, )Gerhard Braunshausen <braunshausen at embl-heidelberg.de> wrote: )>Could anyone offer an explanation for the vast differences )>in speed particularly in comparing the PowerMac 9500 series )>with the SparcStation 20 series? )> )>To my knowledge, the Sparcstations were supposed to be very )>powerful numbercrunchers. Oh? The Sparc has historically been a very poor performer for float calculations, though it does pretty well for int calculations -- if you're doing anything involving intensive floating point calculations you would be much better served by a Power2/PowerPc or HP's PA chip -- both have superior floating point performance. The Dec Alpha is just so fast that it too does well for floating point calculations. The numbers I've seen for the Ultra Sparc indicate that it has improved considerably the floating point performance, so a machine with the UltraSparc would probably be a great performer as well, but the more pedestrian Sparcs are not as good. )I tried Mathematica on an UltraSparc, the results were very good. )They were somewhat better than PowerMac 9500. If they are only `somewhat better' than a machine with a 32 bit partially emulated operating system, then the Ultra Sparc isn't as good as I thought... :-) )I suspect that Mathematica memory management is written for Mac. ? What makes you suspect that? )I also suspect that Wolfram Research uses gcc as the compiler )for minimum hassle when supporting various platforms. I seriously doubt it, but one should be able to tell by investigating a Unix Mathematica binary with strings or od. Tim -- Tim Mooney mooney at toons.cc.ndsu.NoDak.edu Information Technology Services (701) 231-1076 (Voice) Room 242-J6, IACC Bldg. (701) 231-8541 (FAX) North Dakota State University, Fargo, ND 58105