View Single Post
Old 2016-06-01, 14:07   #3
xilman's Avatar
May 2003
Down not across

290116 Posts

Originally Posted by Mr. Odd View Post
Over 4 years ago, based on discussions on this board, I built a Linux box around the idea of "maximum cheap cores" for factoring large numbers and pursuing aliquot sequences:

AMD Phenom II X6 1100T Black Edition
Asus M4A87TD/ USB3 Motherboard
16GB DDR3/1333 Memory
ATX Mid Tower w/ 600W Power supply
Nvidia GT560Ti Graphics w/ 1GB

It now runs really hot and, in fact, overheats the CPU and shuts down if I try to do any factoring. So, time to upgrade! A couple of questions:

- Is AMD still the preferred choice or is Intel the clear winner?
- I mostly use YAFU, which doesn't use the GPU. I know msieve does, though, so which graphics card would you recommend?

That sounds a lot like the system which still runs here but was superseded a few months ago. It's the machine on which I'm typing this material.

/proc/cpuinfo reports:
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 63
model name      : Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz
stepping        : 2
microcode       : 0x2d
cpu MHz         : 3301.000
cache size      : 15360 KB
physical id     : 0
siblings        : 12
core id         : 0
cpu cores       : 6
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 15
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 pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm ida arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm xsaveopt cqm_llc cqm_occup_llc
bugs            :
bogomips        : 6732.55
clflush size    : 64
cache_alignment : 64
address sizes   : 46 bits physical, 48 bits virtual
power management:
with similar results for the other 11 cpus. In reality, of course, this is a 6-core hyperthreaded cpu. It's water-cooled for reliability.

32GB RAM seemed a good compromise between cost and the ability to run reasonably large LA jobs and/or multiple high-B2ECM processes concurrently.

I went for a Nvidia 970 for decent price-performance at factoring --- both ECM and for msieve's polynomial finding. GMP-ECM runs 832 curves in parallel.
xilman is offline   Reply With Quote