1 of 1 people found this helpful
Now you are into the area that is a lot harder to quantify and begins to become more of a board question.
Single or Dual CPU's? The QPI link speed is between the 2 CPU packages and the chipset.
In a dual CPU package (not cores) system, The faster link speed could make up for teh slower processor.
In a single CPU package, I would tend to favor the faster processor, but I don't have any data to support this.
Larger Cache will also result in faster operations. (Rats--- Microsoft patch push... system wants to reboot. More later)
Check the SPEC power web site and see if you can find a SPEC power comparison between these 2 in similar boards....
Ok The SPEC power benchmarking http://www.spec.org/cgi-bin/osgresults site list several test mostly concerned with power usage vs functionality.
Different ODM who market system will test their hardware and post the results. The site is not so good at telling you which mother board is used, but it does tell you about the processor used.
The W3565 was tested in a Fujitsu CELSIUS M470, Intel Xeon W3565
Cint2006 = 35.4
Cfp2006 = 38.4
Cint_rate2006 = 123
Cfp_rate2006 = 91.8
The E5630 in a signal CPU Fujitsu PRIMERGY BX620 S6, Intel Xeon E5630, 2.53 GHz
Cint2006 = (no single CPU results) 2 CPU 33.2
Cfp2006 = (no single CPU results) 2 CPU 38.2
Cint_rate2006 = 118
Cfp_rate2006 = 90.0
The summary here is that these 2 processors are very close to a dead match. The W3565 is slightly beter under the bench mark loading, but only very slightly. If you are even considering a dual socket board or a future upgrade to a duel CPU, go with the E5630.
If you go up to a X5680 so both CPU are the same speed 3.2G, the X5680 blows the socks off the W3565.
Cint2006 = (no single CPU results) 2 CPU 42.3
Cfp2006 = (no single CPU results) 2 CPU 47.9
Cint_rate2006 = 192
Cfp_rate2006 = 131