[QUOTE=kriesel;488116]Per instance, gpuOwL v1.9 on 8M fft length running exponents ~150M, ~115MB private working set, 145MB working set, 382MB peak working set on Windows 7 64-bit. Meanwhile GPU occupancy ~475-490MB each.[/QUOTE]

That seems pretty reasonable.

BTW, are the windows executables downloadable anywhere ?

[QUOTE=tServo;488122]That seems pretty reasonable.

BTW, are the windows executables downloadable anywhere ?[/QUOTE]

from my gpuowl notes:
[URL]http://www.mersenneforum.org/showthread.php?t=22204[/URL]
(this thread)

post #26 guide for compiling on windows with msys64+mingw64
#190 windows binaries of v1.0
#226 v1.9 binaries for windows
[URL]https://github.com/preda/gpuowl[/URL] for the current version source (and some previous too)

[QUOTE=kriesel;488116]Per instance, gpuOwL v1.9 on 8M fft length running exponents ~150M, ~115MB private working set, 145MB working set, 382MB peak working set on Windows 7 64-bit. Meanwhile GPU occupancy ~475-490MB each.[/QUOTE]


May I ask how did you measure ?

[QUOTE=SELROC;488132]May I ask how did you measure ?[/QUOTE]
Sure. GPU values from GPU-Z (sum of dedicated and dynamic memory); other values from Windows Task Manager processes pane. Processes had been running for hours or days before the numbers were taken.

[QUOTE=kriesel;488140]Sure. GPU values from GPU-Z (sum of dedicated and dynamic memory); other values from Windows Task Manager processes pane. Processes had been running for hours or days before the numbers were taken.[/QUOTE]


Thanks, I need to search some tool to do the measure under Linux.

[QUOTE=SELROC;488141]Thanks, I need to search some tool to do the measure under Linux.[/QUOTE]

"top", from command line.

[QUOTE=VBCurtis;488155]"top", from command line.[/QUOTE]


That's good for the cpu side of things.

I don't see anything there in top about gpu core or memory utilization.
CUDA or OpenCl applications can max the gpus out while using essentially zero cpu most of their run duration.

[url]http://archive.oreilly.com/linux/cmd/cmd.csp?path=t/top[/url]


Maybe this?
[url]https://alternativeto.net/software/open-hardware-monitor/[/url]

[QUOTE=kriesel;488156]That's good for the cpu side of things.

I don't see anything there in top about gpu core or memory utilization.
CUDA or OpenCl applications can max the gpus out while using essentially zero cpu most of their run duration.

[URL]http://archive.oreilly.com/linux/cmd/cmd.csp?path=t/top[/URL]


Maybe this?
[URL]https://alternativeto.net/software/open-hardware-monitor/[/URL][/QUOTE]


this tool requires graphic interface

I use a minimalist install of Debian so I need a console tool

[QUOTE=SELROC;488157]this tool requires graphic interface

I use a minimalist install of Debian so I need a console tool[/QUOTE]

It's open source. Modify to your liking.

[URL]http://openhardwaremonitor.org/downloads/[/URL]

[QUOTE=SELROC;488157]I use a minimalist install of Debian so I need a console tool[/QUOTE]

For Nvidia, I use the "nvidia-smi" command (console based; installed as part of the CUDA toolkit).

SELROC, what gpu model(s) are you wanting to display stats for?

gpuOwL runs on Intel and AMD but not currently NVIDIA.
Here are some things to try on for gpu monitoring related to gpuOwL on linux:
[URL]http://www.rkblog.rk.edu.pl/w/p/monitoring-amd-intel-and-nvidia-graphics-card-usage-under-linux/[/URL]

[URL]https://support.amd.com/en-us/kb-articles/Pages/AMDSystemMonitor.aspx[/URL] is Windows 7 specific