Hardware Analysis Community Forums http://www.hardwareanalysis.com/content/topic/3167/ 2008-12-04T20:21:09-05:00 http://www.hardwareanalysis.com/content/topic/3167/ http://media.hardwareanalysis.com/halogo.gif text/html 2002-10-19T23:23:37-05:00 Alejandro Belluscio http://www.hardwareanalysis.com/content/topic/3167/#15095 I've no personal experience. But it's my understanding that the power suppply they used was under rated for the Athlon MP. Given the four HD I guess they should have used a 400/450Watts power supply to have some real stability. It's just a guess, but I've read around and everybody with stability problems with Athlon MP seem to be under rated PS.<br /> <br /> PD: The registration email has a corrupted &quot;From:&quot; field. text/html 2002-10-19T22:41:06-05:00 Chris Faulkner http://www.hardwareanalysis.com/content/topic/3167/#0 Was reffered here by the Slashdot article and I have a few question along with a story.. <img src="http://media.hardwareanalysis.com/smilies/smile1.gif" width="14" height="14" border="0" alt=":)" title=":)"><br /> <br /> I run a small time web hosting co. and after spending 2 weeks planning hardware components for this webserver, among many other tasks it will do. I chose the Tyan 2460 Motherboard with 2 Athlon 1600+ MP Processors and 1 gig of ram. <br /> <br /> Although initially, i had some problems with the Promise Raid card with this motherboard, it was quickly resolved by turning off all Onboard IDE ports with the exception of the port the CDRom was on. After that, things went smoothly. The reason I went with the AMD solution was quite simple. If my server was to have intensive I/O reads and writes, I need a bus that can handle it, and 133mhz bus on the Intel was not something i admired, and also Intel's ability to handle one I/O at a time on a MP box! Amd Athlon MP procs can handle 2 I/O processes at the same time. <br /> <br /> Last night, for kicks, I ran 8 (eight) setiathome processes and 1 Prime95 process to jump the load up while the webserver was doing it's job. It ran for more than 4 hours at 9.02 load avg (under Redhat 7.3) and didn't skip a beat. MySQL transactions were perky and Apache 1.3.27 was scooting along as normal as could be. No error messages were present in the var/log/messages during it's 4 hour sojourn.<br /> <br /> I think you should re-evaluate the AMD position, even though it is somewhat painful at the beginning. Other than that, I agree totally with this article. <br /> <br /> Chris Faulkner