A tale of 3 CPUs

chris2be8 · 2010-07-19, 17:10

After one of my systems died I bought a replacement with a 2.8GHz AMD dual core CPU. But it's slower that my laptop with a 1.6 GHz Intel dual core CPU.

/proc/cpuinfo on the amd is:

Code:

 processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 16
model           : 6
model name      : AMD Athlon(tm) II X2 240 Processor
stepping        : 2
cpu MHz         : 800.000
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 2
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc pni cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy altmovcr8 abm sse4a misalignsse 3dnowprefetch osvw
bogomips        : 5604.05
TLB size        : 1024 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor       : 1
vendor_id       : AuthenticAMD
cpu family      : 16
model           : 6
model name      : AMD Athlon(tm) II X2 240 Processor
stepping        : 2
cpu MHz         : 800.000
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 1
cpu cores       : 2
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc pni cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy altmovcr8 abm sse4a misalignsse 3dnowprefetch osvw
bogomips        : 5600.42
TLB size        : 1024 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

On the laptop it's:

Code:

 processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 15
model name      : Genuine Intel(R) CPU           T1600  @ 1.66GHz
stepping        : 13
cpu MHz         : 1662.505
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 2
apicid          : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3328.07
clflush size    : 64
cache_alignment : 64
address sizes   : 36 bits physical, 48 bits virtual
power management:

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6
model           : 15
model name      : Genuine Intel(R) CPU           T1600  @ 1.66GHz
stepping        : 13
cpu MHz         : 1662.505
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 1
cpu cores       : 2
apicid          : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3325.06
clflush size    : 64
cache_alignment : 64
address sizes   : 36 bits physical, 48 bits virtual
power management:

The 3rd system is now the oldest, it's got 1 core with hiperthreading. /proc/cpuinfo is:

Code:

 processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      :               Intel(R) Pentium(R) 4 CPU 3.06GHz
stepping        : 9
cpu MHz         : 3060.746
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall lm constant_tsc pni monitor ds_cpl tm2 cid cx16 xtpr lahf_lm
bogomips        : 6130.85
clflush size    : 64
cache_alignment : 128
address sizes   : 36 bits physical, 48 bits virtual
power management:

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      :               Intel(R) Pentium(R) 4 CPU 3.06GHz
stepping        : 9
cpu MHz         : 3060.746
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall lm constant_tsc pni monitor ds_cpl tm2 cid cx16 xtpr lahf_lm
bogomips        : 6121.76
clflush size    : 64
cache_alignment : 128
address sizes   : 36 bits physical, 48 bits virtual
power management:

I'm running 2 threads of the 64bit version of gnfs-lasieve4I15e on both. The laptop takes about 0.73 sec/relation, the amd about 1.36 sec/relation. The 3rd system takes about 0.93 sec/relation on 1 thread (it's not got enough memory to run 2 threads at once).

Any idea why the amd is slower than the others? I expected 2 cores at 2.8GHz to do each match the 3rd system.

Chris K

xilman · 2010-07-19, 17:48

Quote:

Originally Posted by chris2be8

After one of my systems died I bought a replacement with a 2.8GHz AMD dual core CPU. But it's slower that my laptop with a 1.6 GHz Intel dual core CPU.

/proc/cpuinfo on the amd is:

Code:

 processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 16
model           : 6
model name      : AMD Athlon(tm) II X2 240 Processor
stepping        : 2
cpu MHz         : 800.000
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 2
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc pni cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy altmovcr8 abm sse4a misalignsse 3dnowprefetch osvw
bogomips        : 5604.05
TLB size        : 1024 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

processor       : 1
vendor_id       : AuthenticAMD
cpu family      : 16
model           : 6
model name      : AMD Athlon(tm) II X2 240 Processor
stepping        : 2
cpu MHz         : 800.000
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 1
cpu cores       : 2
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc pni cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy altmovcr8 abm sse4a misalignsse 3dnowprefetch osvw
bogomips        : 5600.42
TLB size        : 1024 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 48 bits physical, 48 bits virtual
power management: ts ttp tm stc 100mhzsteps hwpstate

On the laptop it's:

Code:

 processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 15
model name      : Genuine Intel(R) CPU           T1600  @ 1.66GHz
stepping        : 13
cpu MHz         : 1662.505
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 2
apicid          : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3328.07
clflush size    : 64
cache_alignment : 64
address sizes   : 36 bits physical, 48 bits virtual
power management:

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6
model           : 15
model name      : Genuine Intel(R) CPU           T1600  @ 1.66GHz
stepping        : 13
cpu MHz         : 1662.505
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 1
cpu cores       : 2
apicid          : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl tm2 cx16 xtpr lahf_lm
bogomips        : 3325.06
clflush size    : 64
cache_alignment : 64
address sizes   : 36 bits physical, 48 bits virtual
power management:

The 3rd system is now the oldest, it's got 1 core with hiperthreading. /proc/cpuinfo is:

Code:

 processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      :               Intel(R) Pentium(R) 4 CPU 3.06GHz
stepping        : 9
cpu MHz         : 3060.746
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall lm constant_tsc pni monitor ds_cpl tm2 cid cx16 xtpr lahf_lm
bogomips        : 6130.85
clflush size    : 64
cache_alignment : 128
address sizes   : 36 bits physical, 48 bits virtual
power management:

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      :               Intel(R) Pentium(R) 4 CPU 3.06GHz
stepping        : 9
cpu MHz         : 3060.746
cache size      : 1024 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall lm constant_tsc pni monitor ds_cpl tm2 cid cx16 xtpr lahf_lm
bogomips        : 6121.76
clflush size    : 64
cache_alignment : 128
address sizes   : 36 bits physical, 48 bits virtual
power management:

I'm running 2 threads of the 64bit version of gnfs-lasieve4I15e on both. The laptop takes about 0.73 sec/relation, the amd about 1.36 sec/relation. The 3rd system takes about 0.93 sec/relation on 1 thread (it's not got enough memory to run 2 threads at once).

Any idea why the amd is slower than the others? I expected 2 cores at 2.8GHz to do each match the 3rd system.

Chris K

According to the information you posted, the cpu is actually clocked at 0.8GHz.

If this is not a pasto on your part, you should try to find out why.

Paul

apocalypse · 2010-07-20, 00:48

IIRC, I had to disable PowerNow in my BIOS for one of my AMD systems to avoid this problem.

chris2be8 · 2010-07-20, 08:31

Quote:

Originally Posted by xilman

According to the information you posted, the cpu is actually clocked at 0.8GHz.

If this is not a pasto on your part, you should try to find out why.

Paul

It's not a pasto, it was 800MHz. I've just had another look, it's now 2800MHz for processor 0 and 800 MHz for processor 1.

I'll take a hard look at BIOS setting this evening.

Chris K

xilman · 2010-07-20, 11:17

Quote:

Originally Posted by chris2be8

It's not a pasto, it was 800MHz. I've just had another look, it's now 2800MHz for processor 0 and 800 MHz for processor 1.

I'll take a hard look at BIOS setting this evening.

Chris K

Very likely some kind of thermal throttling. Note the Bogomips figure implies that the machine booted at 2.8GHz and was only subsequently underclocked.

Paul

markr · 2010-07-20, 12:16

Quote:

Originally Posted by apocalypse

IIRC, I had to disable PowerNow in my BIOS for one of my AMD systems to avoid this problem.

This would be my bet. I think AMD calls it PowerNow in their laptop CPUs and Cool'n'Quiet in desktop models. Intel's equivalent is SpeedStep or EIST (Enhanced Intel SpeedStep Technology - what a mouthful!). The aim is to throttle back CPU clock speed & voltage when not needed, to reduce power consumption, heat, noise, etc.

I have one CPU with SpeedStep, and mprime underperformed until I disabled it in BIOS.

chris2be8 · 2010-07-20, 19:45

Quote:

Originally Posted by apocalypse

IIRC, I had to disable PowerNow in my BIOS for one of my AMD systems to avoid this problem.

That was it. Shutdown, disable PowerNow, reboot, and it's doing 0.41 sec/relation.

Big thanks. Trying every BIOS setting until I found the right one would have taken hours. Neither the BIOS nor the motherboard manual say what the settings do. The drivers CD has PDF manuals on it but reading them would have been a lengthy task. And rather difficult when the PCs in the setup menus.

Chris K

jasong · 2010-07-20, 23:12

A lot of us have had this problem, our OSes are not made for Distributed Computing, so they screw us up.

When you think about it, there probably aren't a whole heck of a lot of people that like to tell a computer program to not use any computer cycles if something else needs them, but also use EVERYTHING that's left.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
cpus running at 10%	wildrabbitt	Information & Answers	9	2015-01-20 13:31
Can't Merge CPUs	Rodrigo	PrimeNet	11	2012-03-03 19:45
Which of these CPUs is most productive?	Rodrigo	Hardware	123	2011-02-05 21:42
Whither Older CPUs?	Rodrigo	Operation Billion Digits	4	2010-10-20 15:12
I sad tale	science_man_88	Lounge	0	2010-08-27 22:38

2010-07-20, 00:48	#3
apocalypse Feb 2003 2·3·29 Posts	IIRC, I had to disable PowerNow in my BIOS for one of my AMD systems to avoid this problem.

2010-07-20, 23:12	#8
jasong "Jason Goatcher" Mar 2005 3×7×167 Posts	A lot of us have had this problem, our OSes are not made for Distributed Computing, so they screw us up. When you think about it, there probably aren't a whole heck of a lot of people that like to tell a computer program to not use any computer cycles if something else needs them, but also use EVERYTHING that's left.