I've been experimenting with a new PC, Intel i7 Core 990x (6 Cores, 12 Threads), and two nVidia GTX590 in SLI.

I've managed to install the Lunatics config, and experiment with the app_info.xml.

I'm currently running 12 w/u per CPU, and 1 w/u per GPU (4 GPUs).

from what I've read, I should be able to push maybe 2 or 3 w/u per GPU, at this point, when I alter CUDA value to 0.5, things get slow, and I do not see 2 w/u per GPU, maybe need to tweak more with the average and max ncpus values, which is currenty set to 0.04.

any ideas?

---

Don't use SLI.

---

And disable the use of a CPU core or two.
GPUs do require a little support from the CPU in their daily tasks.

You also may want to experiment with disabling HT, and letting the CPUs run at full speed instead of half.

In my early experiments with such settings, there was NO gain with running HT on.

Others claimed small improvements in total processing, so YMMV applies.

I personally always found it better to let the CPU cores run at full speed, which processes the GPU startup routines faster.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

it does not seem to use SLI, as each w/u is given to each GPU

---

how do I disable a few cores, just change the percentage to use?

how then do I jackup the w/us per GPU.

---

thanks guys

disabled SLI, the event logs looks the same.

4 GPUs.

and I've chamnged to use 85% of the CPU, so I've now got

10 w/u per processor (1 pre core)

4 w/u per GPUs.

and changed CUDA from 1 to 0.5, so I've got 8 w/u per GPUs, and 10 w/u per processor, looks good,

I wait for the meltdown!

---

Keep a watch on your results as you are starting to get a number of errors.

---

SETI@home classic workunits 4,019
SETI@home classic CPU time 34,348 hours

You definetly need to free more CPU cores running 4 GPUs with multiple instances.

---

With each crime and every kindness we birth our future.

You definetly need to free more CPU cores running 4 GPUs with multiple instances.

Very true, also needed with AMD/ATI GPUs, doesn't matter if Cross-Fire
is enabled.
Every GPU has 1 thread of an i7-2600, also running on higher frequncy
by increasing the Base-Clock from 100 to 102 to 104MHz. Since the multiplier is locked, it's the only option and works OK.

But I do have several driver-crashes, specially with SETI work?!

NVidia GPUs also need sometimes a dedicated core, CPU time is higher on NVidia
GPUs.
Almost double compaired to AMD/ATI GPUs.

I've set SETI the same way, 0.5CPU in stead of 0.04 so it uses 1 core/thread
setting this in app_info.xml when doing 2 pper GPU 1 core/thread will
be used, Einstein uses this setting too.

This AMD/ATI host does Docking or Einsteins Gravity Wave search on CPU.
SETI and Einstein only GPU.

---

You definetly need to free more CPU cores running 4 GPUs with multiple instances.

Very true, also needed with AMD/ATI GPUs, doesn't matter if Cross-Fire
is enabled.
Every GPU has 1 thread of an i7-2600, also running on higher frequncy
by increasing the Base-Clock from 100 to 102 to 104MHz. Since the multiplier is locked, it's the only option and works OK.

But I do have several driver-crashes, specially with SETI work?!

NVidia GPUs also need sometimes a dedicated core, CPU time is higher on NVidia
GPUs.
Almost double compaired to AMD/ATI GPUs.

I've set SETI the same way, 0.5CPU in stead of 0.04 so it uses 1 core/thread
setting this in app_info.xml when doing 2 pper GPU 1 core/thread will
be used, Einstein uses this setting too.

This AMD/ATI host does Docking or Einsteins Gravity Wave search on CPU.
SETI and Einstein only GPU.

1,216.96 runtime

233.94 CPU time

118.06 Credit

SETI@home Enhanced v6.10 (cuda_fermi)



2763176272

6628794

21 Dec 2012 | 17:29:59 UTC

21 Dec 2012 | 21:41:34 UTC

Voltooid en gecontroleerd

1,570.51 runtime

102.46 CPU time. Less as NVidia GPUs.

118.06 Credit.



SETI@home Enhanced
Anoniem platform (ATI GPU)
Both platforms run 2 tasks per GPU.

---

seems to be running okay, I'll watch for any new errors.

---