Celeron E1200, Pentium E2140

Message boards : Number crunching : Celeron E1200, Pentium E2140

Author	Message
chungenhung Send message Joined: 28 Dec 07 Posts: 16 Credit: 2,967,892 RAC: 0 Level Scientific publications	Message 762 - Posted: 18 Jan 2008 \| 16:01:13 UTC
	Which one would you recommend? Both overclocks to about the same clock speed, except that the Celeron have half the cache. I was wondering if PS3Grid project performance will be affected by cache size or not. And if they do, by how much?? My goal is to build a low end machine now, so I will be able to donate it to third world country later this year.
	ID: 762 \| Rating: 0 \| rate: / Reply Quote

Krunchin-Keith [USA] Send message Joined: 17 May 07 Posts: 512 Credit: 111,288,061 RAC: 0 Level Scientific publications	Message 765 - Posted: 19 Jan 2008 \| 13:41:47 UTC - in response to Message 762.
	Which one would you recommend? Both overclocks to about the same clock speed, except that the Celeron have half the cache. I was wondering if PS3Grid project performance will be affected by cache size or not. And if they do, by how much?? My goal is to build a low end machine now, so I will be able to donate it to third world country later this year. PS3GRID only runs on PS3 (Playstation 3) and IBM Cell Blades. Because of the nature of the computations involved, there are no plans for any applications for other processors, like you mention. On conventional processors it just takes too long to get meaningful results. Remember the PS3 does around the same amount of work as 16 single computers or in other terms, it does the work of 1 single computer at least 16 times as fast. So the work it now takes our PS3\'s to do in one day would take a single computer 16 days to do. In reality though it takes longer because of the way code differences between conventional CPU\'s and the Cell Blade engine. The PS3GRID data is need back much faster than that. We only have 3 day deadlines and as each task is completed another is generated based on the result from the first, sort of a continuation of the first. You can quickly see by doing this on single computers would take too long to get the end result.
	ID: 765 \| Rating: 0 \| rate: / Reply Quote

Message boards : Number crunching : Celeron E1200, Pentium E2140

//