Welcome to MilkyWay@home

Posts by MarcelliusIkaP

1) Message boards : Number crunching : CPU configuration for GPU apps (Message 67199)
Posted 4 Mar 2018 by MarcelliusIkaP
Post:
So basically, MW "sense" that my system run best when it's using AVX, right? Just feels odd that from all Bulldozer-derived cores, only Excavator that performs better on AVX than on SSE4.1
2) Message boards : Number crunching : CPU configuration for GPU apps (Message 67180)
Posted 2 Mar 2018 by MarcelliusIkaP
Post:
Thanks for replying.
Also I think there are OS requirement to enable/use AVX. Not sure what the other users computer has for an OS.

My system is using Win7Pro and my friend's system is using Win10Pro. Both are consumer-grade OS.
I also believe the general consensus for the best performance for AMD Bulldozer and its derivatives is using SSE4.1 instruction set, not AVX. That's why I'm a bit loss here.
3) Message boards : Number crunching : CPU configuration for GPU apps (Message 67173)
Posted 1 Mar 2018 by MarcelliusIkaP
Post:
Hi everyone.
This is my first post on this forum.
Recently, I was comparing my validated tasks with other's and found something interesting (at least for me). This is my partial stderr output:
<core_client_version>7.8.2</core_client_version>
<![CDATA[
<stderr_txt>
<search_application> milkyway_separation 1.46 Windows x86 double OpenCL </search_application>
Reading preferences ended prematurely
BOINC GPU type suggests using OpenCL vendor 'Advanced Micro Devices, Inc.'
Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4'
Switching to Parameter File 'astronomy_parameters.txt'
<number_WUs> 5 </number_WUs>
<number_params_per_WU> 20 </number_params_per_WU>
Using AVX path

My system consists of: CPU: AMD Athlon x4 845 (Excavator), GPU: AMD RX 560

And below is my friend's partial stderr output:
<core_client_version>7.8.3</core_client_version>
<![CDATA[
<stderr_txt>
<search_application> milkyway_separation 1.46 Windows x86 double OpenCL </search_application>
Reading preferences ended prematurely
BOINC GPU type suggests using OpenCL vendor 'Advanced Micro Devices, Inc.'
Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4'
Switching to Parameter File 'astronomy_parameters.txt'
<number_WUs> 5 </number_WUs>
<number_params_per_WU> 20 </number_params_per_WU>
Using SSE4.1 path

His system consists of: CPU: AMD FX-6300 (Piledriver), GPU: AMD R9 280X

As you can see, both mine and his CPU are derived from AMD's Bulldozer architecture. So, why do we use different instruction set even though both of us have AVX capability?

Thanks in advance.

ika




©2019 Astroinformatics Group