Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: CPU performance



Paul Goyette wrote:

> Well, after a long absence, I rejoined the distributed.net RC5-72 
> contest today, installed their client software, and ran some quick 
> benchmarks.  The results were rather surprising to me...
> 
> First, here's the cpu information for each machine...
> 
> Machine #1 - AMD Athlon 64 X2 Dual Core processor
> cpu0 at mainbus0 apid 0: AMD 686-class, 2812MHz, id 0x40f33
> cpu1 at mainbus0 apid 1: AMD 686-class, 2812MHz, id 0x40f33
> 
> Machine #2 - AMD Athlon 64 X2 Dual Core processor
> cpu0 at mainbus0 apid 0: AMD 686-class, 2412MHz, id 0x40fb2
> cpu0: AMD PowerNow! Technology 2400 MHz
> cpu0: available frequencies (Mhz): 1000 1800 2000 2200 2400
> cpu1 at mainbus0 apid 1: AMD 686-class, 2412MHz, id 0x40fb2
> 
> Machine #3 - AMD Phenom processor
> cpu0 at mainbus0 apid 0: AMD 686-class, 2310MHz, id 0x100f22
> cpu0: erratum 298 present
> cpu0: erratum 298 present
> cpu0: erratum 309 present
> cpu0: WARNING: errata present, BIOS upgrade may be
> cpu0: WARNING: necessary to ensure reliable operation
> cpu1 at mainbus0 apid 1: AMD 686-class, 2310MHz, id 0x100f22
> cpu2 at mainbus0 apid 2: AMD 686-class, 2310MHz, id 0x100f22
> cpu3 at mainbus0 apid 3: AMD 686-class, 2310MHz, id 0x100f22
> 
> Machine #4 - Intel Core 2/Xeon
> cpu0 at mainbus0 apid 0: Intel 686-class, 2400MHz, id 0x6fb
> cpu1 at mainbus0 apid 3: Intel 686-class, 2400MHz, id 0x6fb
> cpu2 at mainbus0 apid 2: Intel 686-class, 2400MHz, id 0x6fb
> cpu3 at mainbus0 apid 1: Intel 686-class, 2400MHz, id 0x6fb
> 
> 
> Test results.  There are 5 different "benchmark" tests in the software, 
> and all provide similar results.
> 
>       Test #     Machine 1    Machine 2    Machine 3    Machine 4
>             0         10.62         9.07         4.37         7.77
>          1         11.21         9.57         4.61         8.04
>          2          2.75         2.34         1.12         2.20
>          3          2.80         2.39         1.13         2.19
>          4          2.78         2.36         1.12         2.18
> 
> The numbers in the above table are "Millions of keys processed per 
> second", so a larger value is better.  And the numbers are all on a 
> single core.
> 
> The first surprise:  the 2.3GHz Phenom (#3) was less than half as 
> capable as either 2.8GHz or the 2.4GHz Athlon 64 machines #1 and #2.

Phailnom? :-)

> The second surprise: the Intel Core 2 at 2.4GHz is about 20% slower than 
> the Athlon running at the same 2.4GHz speed.
> 
> Are these numbers surprising to anyone else?

Attachment: pgp_hNoj7er4d.pgp
Description: PGP signature



Home | Main Index | Thread Index | Old Index