X-pop3-spooler: POP3MAIL 2.1.0 b 4 980420 -bs- Date: Tue, 21 Jul 1998 09:11:54 +0100 (BST) From: Dave Tweed X-Sender: tweed AT neon cc: beastium-list AT Desk DOT nl Subject: Re: speed PGCC vs GCC for DIEP In-Reply-To: <3.0.32.19980719215256.0098e2c0@xs4all.nl> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: Marc Lehmann Status: RO X-Status: A Content-Length: 1827 Lines: 37 On Sun, 19 Jul 1998, Vincent Diepeveen wrote: > Anyway K6 is fast considering its price, but is it smart for a > compiler to make optimizations for an outdated socket 7 clone? It's probably unnecessary to say this to Marc, but I'm one of the people who benefits from the amd k6 optimisations and would urge that you don't take this argument to heart. (Vincent, if it helps you to understand why I've gone for a solution which is inferior in terms of absolute performane: I have an old, pre-split voltage motherboard which consequently limits me to either staying with the classic pentium, or going for a k6-233(*) when I upgraded three weeks ago; there's no intermediate option like a high end Pentium-MMX. Whilst I'd love a P-II machine my bank manager won't let me, yet I do need something which speeds up the huge Mathematica and image processing calculations that I do.) > I'll do that because i think socket 7 is outdated. No future. > Look to the level 2 cache what they did to it. They put it at the mainboard!! > Awfull! So socket 7 has no future, the faster your processor, the less > you profit from it as level 2 cache speed kills you. It probably is outdated at the high performance, low performance/dollar level, but it certainly still a good option at the medium performance, high performace/dollar level. And for a lot of students like me - both undergraduate and graduate - that's where we live. (*) This worked for me; this might be a fluke so I won't be held liable if anyone tries this and it goes badly wrong. ___cheers,_dave__________________________________________________________ email: tweed AT cs DOT bris DOT ac DOT uk "valid PERL regexp: text not produced by www.cs.bris.ac.uk/~tweed/pi.htm any of infinitely many monkeys after work tel: (0117) 954-5253 an infinitely long time."