| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 30
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hello whocrazy,
Look at http://www.cse.scitech.ac.uk/disco/Benchmarks/whetstone.shtml The Woodcrest is the first Core2 chip. At 3 Ghz it runs the Whetstone benchmark un 3.3 seconds. The Pentium 4 at 3.4 Ghz runs it in 9 seconds. The Core2 architecture is much faster than the Pentium 4, almost 3 times as fast per clock. Lawrence |
||
|
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18667 Status: Offline Project Badges:
|
Hello whocrazy, Look at http://www.cse.scitech.ac.uk/disco/Benchmarks/whetstone.shtml The Woodcrest is the first Core2 chip. At 3 Ghz it runs the Whetstone benchmark un 3.3 seconds. The Pentium 4 at 3.4 Ghz runs it in 9 seconds. The Core2 architecture is much faster than the Pentium 4, almost 3 times as fast per clock. Lawrence Yes, this lays out the difference the architecture (design) makes. For simplicity, I ignored the fact that we were comparing a Yugo and a Porsche and looked at it just in terms of both being cars. With a 2GHz quad, you will more accurately see 4-6 times the amount of work done in the same clock time on a P4. |
||
|
|
mikaado
Cruncher Joined: Dec 3, 2007 Post Count: 14 Status: Offline |
How much percent faster is this q9450 per work unit compared to the p4 3.2ghz? I didn't find a direct comparison between these two. But I tried to compare P4 3.2GHz with E6600 (dual core model) and Q6600 (quad core model), in tasks where E6600 and Q6600 get nearly the same amount of points. So it seems like Q6600 at 2.4GHz is roughly 1.5 times faster in a single task duty, than P4 is at 3,2 GHz. A task that takes 80 seconds from E6600 and Q6600, takes 134 seconds from P4 3.2 GHz. So I guess the same tasks takes about 77 seconds with Q9300 and about 74 seconds with Q9450. These are really rough estimates though. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I would agree that multi-processor cores are superior to the older P4 almost regardless of speed per processor core. Even if we assumed the P4 in your computer processed a thread faster, it doesn't take into account other threads(processes) running on your machine that eat CPU cycles. Running multiple threads in parallel is a bit like hauling groceries from your vehicle after shopping. Would you rather carry 2,4,8 bags each trip even if your a bit slower or carry 1 bag each time even if your a bit faster per trip?
There are very few to almost no programs that spread the work load of a thread to multiple cores, but for distributed computing that doesn't matter. The upcoming Nehalem architecture by Intel(mid 2009) should be pretty top notch by any standards. It will eliminate the Front Side Bus(FSB) which has been a notorious bottle neck with Intel for a very long time(those ackward pauses). They will use the Digital Compaq Alpha EV7.0 version 2 bus design much like AMD has been using the Compaq Alpha EV6.8 bus design. They have re-named it the Intel Quickpath architecture under license from Compaq. http://www.intel.com/technology/architecture-silicon/next-gen/ The new motherboards will also support the 3rd generation Double Date Rate(DDR) memory arhitecture which is a derivative of technology originally borrowed(stolen) from Rambus technologies. The major manufactures like Samsung now pay some royalties to Rambus to license the technology. Forms of Rambus technology are inherent in any memory that says, double date rate. If your just looking for an excellent price/performance processor now for a good price an Intel quad as stated earlier should suffice just fine. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
what would be really good is if you could somehow interrupt the work unit and reset it, so you could copy it to different computers to see how fast it goes.
|
||
|
|
Sekerob
Ace Cruncher Joined: Jul 24, 2005 Post Count: 20043 Status: Offline |
Nothing stopping you to make a copy of the datadir after download of work and run it on different machines. Offline on the test systems of course and unsuitable to upload, but from the original location. The servers know which device is the designated host.
----------------------------------------
WCG
----------------------------------------Please help to make the Forums an enjoyable experience for All! [Edit 1 times, last edit by Sekerob at Aug 10, 2008 11:56:02 AM] |
||
|
|
l_mckeon
Senior Cruncher Joined: Oct 20, 2007 Post Count: 439 Status: Offline Project Badges:
|
And if you are having a new system built for you, it would be crazy not to do at least a mild overclock.
Intel Core 2 Quads and Duos have a lot of unused speed under the hood; you can effectively get a more expensive processor for nothing. Talk to your system assembler. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
And if you are having a new system built for you, it would be crazy not to do at least a mild overclock. Intel Core 2 Quads and Duos have a lot of unused speed under the hood; you can effectively get a more expensive processor for nothing. Talk to your system assembler. Yes, i would like to do this. But i dont have a system assembler. Intel Core 2 Quad Q9450 / 2.66 GHz ![]() |
||
|
|
mikaado
Cruncher Joined: Dec 3, 2007 Post Count: 14 Status: Offline |
I think we should not talk about overclocking in this thread, because this only confuses most of the people. Although, I must mention ...
What comes to overclocking, Q9XXX processors have quite a low core ratio, so your other parts have to be high end as well. You might consider this when choosing especially between Q6600 and Q9300. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I think we should not talk about overclocking in this thread, because this only confuses most of the people Ok, understand you. ![]() |
||
|
|
|