I'm currently using an older MSI K9AGM2-FIH AM2 motherboard in my unRAID server.  Normally it runs a BE-2400 processor - 45W, dual core @ 2.3 GHz.  Power supply is PC Power & Cooling Silencer 500.  I'm only running two drives at the moment, 4.5 beta 11.

 

I got my hands on a 2650e processor: 15W TDP, single core @ 1.6 GHz.  I was expecting that I might drop this in my board and be able to shave 20+W off my idle power draw.

 

BE-2400          2650e       

55W    ---      48W        ---  Idle, disks spun down         

68W    ---      61W        ---  Idle, both disks spun up

89W    ---      71W        ---  Parity check*

 

* Currently running a parity check while streaming a dvd to a laptop, at 9% parity check progress the 2650e is doing 109,024 KB/sec (granted, only one data drive and parity) - which is similar to the BE-2400.  And it's showing 96% idle as per the unMenu Top Processes view.  So this is encouraging as far as cpu capability.

 

I guess I'm a little surprised at the minimal power drop from the 2650e on this motherboard.  I'm wondering if it's just not 'optimized' for the 2650e, as the board and the BIOS are a few years old and the 2650e is relatively newer?  The board does not have any undervolt / underclock capability, unfortunately - otherwise I'd probably stick with the BE-2400 and try to cut it back that way.

 

I'm going to run with this 2650e for a while and see how it works out.  The numbers do suggest that it might provide more benefit in a higher power use environment.  We'll see.

 

Any comments?

 

Both cpus draw minimal power when they're idle. Load each CPU to 100% usage to see and notice the real difference.