Message boards : Number crunching : Any progress in making crunching AMD-friendly?
Message board moderation
Author | Message |
---|---|
Send message Joined: 21 Mar 05 Posts: 33 Credit: 63,483 RAC: 0 |
Or does the bias against AMD processors continue? Just wondering. Matthew <img src=\"http://boinc.mundayweb.com/one/stats.php?userID=2118&prj=4\"> |
Send message Joined: 7 Aug 04 Posts: 2187 Credit: 64,822,615 RAC: 5,275 |
The latest sulphur version, 4.22 for Windows, has a significant speed increase over 4.19 for most AMDs. Athlon64 3400+ (2.2 GHz) before 3.18 s/TS after 2.65 s/TS AthlonXP 3200+ before 3.55 s/TS after 3.10 s/TS |
Send message Joined: 26 Aug 04 Posts: 59 Credit: 438,133 RAC: 0 |
Those numbers seem quite low to me, and I wonder why my machine is so slow. I\'m running a 4.21 sulphur model doing about 4.0 s/TS, The processor is an AMD64 3200. However, I\'m running Gentoo64 and know that 32bit apps are slower under pure 64bit. I just can\'t understand how it can be that slow. Kenneth Proud owner of the CPDN Wow-Mug! |
Send message Joined: 7 Aug 04 Posts: 2187 Credit: 64,822,615 RAC: 5,275 |
Those numbers seem quite low to me, and I wonder why my machine is so slow. I\'m running a 4.21 sulphur model doing about 4.0 s/TS, The processor is an AMD64 3200. However, I\'m running Gentoo64 and know that 32bit apps are slower under pure 64bit. I just can\'t understand how it can be that slow. That is kind of slow. My Athlon64 3200+ did about 3.45 s/TS in WinXP, and about 3.5 s/TS in FC3 x86-64 Linux with sulphur 4.21 and before. But this was with low-latency RAM running at 1T command rate. If you have higher latency RAM and it\'s running at 2T command rate, that could make quite a difference. |
Send message Joined: 26 Aug 04 Posts: 59 Credit: 438,133 RAC: 0 |
I\'ll soon get some new (and more) RAM, I\'ll see if that makes a difference then. Currently I only have 512MB, and it\'s not the fastest kind. |
©2024 cpdn.org