MP 1,1-5,1 Faulty 2 x Xeon 3.46Ghz?

[Noobraino]

Hello guys,

I recently bought a second pair of Intel Xeon 3.46Ghz Westmere CPU for my second cMP. I installed the pair and ran some benchmarks test, this new pair clearly fair badly compared to the first pair I bought couple months ago (pictures will be posted).

So I'm wondering, could it be that the new pair was tested in a "patched" Catalina OS and the first pair was tested in a pure Mojave install? Or is the speed of the DDR matter in these kind of test as the DDR for the first cMP is at 1333 Mhz but in the second cMP it is at 1066Mhz.

So here are the pictures:

Geekbench 5

First cMP with 2 x Intel Xeon 3.46Ghz Westmere - 1333Mhz DDR3

Second cMP with 2 x Intel Xeon 3.46Ghz Westmere - 1066Mhz DDR3

First cMP with 2 x Intel Xeon 3.46Ghz Westmere - 1333Mhz DDR3

Second cMP with 2 x Intel Xeon 3.46Ghz Westmere - 1066Mhz DDR3

Curiously enough both pair of CPU perform closer in CineBench

First cMP with 2 x Intel Xeon 3.46Ghz Westmere - 1333Mhz DDR3

Second cMP with 2 x Intel Xeon 3.46Ghz Westmere - 1066Mhz DDR3

 

[mikas]

What are the configs of the modules in each machine?
I believe it is the ram, and it's pretty easy to verify if it's the ram, just change them between the machines.

No real difference in cinebench, which maxes out all of the available CPU cores there is.
GB tests memory operations too, so it leans towards that reason.
About that luxmark result I do not know.

 

[h9826790]

Looks normal to me. The difference most likely because both GB and Luxmark are very memory speed sensitive.

You can swap the RAM between the two cMP, and see if you can reproduce the results.

 

[Noobraino]

Thank you so much guys, it was the RAM indeed. I switched the modules and got slightly better results now with the new CPUs compared to the old CPUs pair.