1) Message boards : Number crunching : Free hard disk for long runs (Message 41473)
Posted 5 hours ago by Profile skgiven
My understanding is that <rsc_disk_bound>n</rsc_disk_bound> defines the maximum allowable write to disk size for the task.

If the task uses more disk space than this it cannot be written to disk and in that event is killed.

If the app calls on this value as being required disk space then the bug is in the app.

If Boinc has changed something that requires there to be n free bytes available on the disk drive then the bug exists within a Boinc update (not very likely).

If during a run you no longer have ~5GB free disk space available to Boinc then that's where it gets complicated. LAIM doesn't apply to GPUGrid GPU tasks, so they have to be saved to disk, but having it disabled for CPU tasks could also force them to be written to disk and you to run out of disk space. If Boinc stops crunching and you don't have the disk space available then you may lose the task. However, tasks don't normally require ~5GB. Maybe the app is using the n value and saying there isn't enough disk space without first looking at the tasks actual size or trying to save the task to disk and waiting for a failure before cancelling? If that's the case then IMO the bug is in the app. While reducing the n value might be a quick fix (for a new batch) it would still be a rigid cut-off point.

Recent NOELIA tasks also use this same n value (4768MB).
2) Message boards : Number crunching : Free hard disk for long runs (Message 41451)
Posted 3 days ago by Profile skgiven
Might this be due to the limits set in Boinc Manager?
3) Message boards : Graphics cards (GPUs) : 780Ti vs. 770 vs. 750Ti (Message 41450)
Posted 3 days ago by Profile skgiven
When you start using risers it's gone way past basic closed case design. It's very much an open case/bench worktop design.
4) Message boards : Graphics cards (GPUs) : Good and bad video card brands (Message 41441)
Posted 5 days ago by Profile skgiven
Card swapping sounds like you best shot there.
5) Message boards : Graphics cards (GPUs) : 780Ti vs. 770 vs. 750Ti (Message 41440)
Posted 5 days ago by Profile skgiven
I doubt there's room for more than one GPU on an x1 and that would have to be single-width.


Powered PCIe Risers :)
6) Message boards : Number crunching : 0.973 CPUs + 1 GPU (Message 41436)
Posted 6 days ago by Profile skgiven
Also. Has someone figured out what the best clocks for a 970 are when underclocking to get best performance per Watt?

Performance/Watt needs to be taken at the system level (not the GPU level), and measured at the wall.
Your i5-2500K (95W TDP) and supporting hardwares power usage would be as much of a concern as the 970's clocks. You really need to tune the system, not just the GPU (though that in itself helps a bit).

The simplest way to control performance on the Maxwell's is just to throttle the power. You could set it to 90%, 80% or less. The issue then would be it's p-state might drop from p2 to p5 - assuming you don't want to run a 970 at 405MHz! Can be controlled using NVidia Inspector (overclocking), but it's quite confusing when you first start to use it.

My i7-3770K @3.7GHz and 2 GTX970's (W7):
4 GPUGrid tasks running on 2 GTX970's + 2 climate models uses 423W (SWAN_SYNC in use). Estimated throughput gain is >15% from running 2tasks/GPU.

Note that in this setup GPUGrid is using 4 CPU threads to support 4 GPU tasks.

Without the 2 climate models running the power drops to 415 (only 8W less, which is typical given that the CPU is already being well used).

If I then suspend one GPUGrid task the system power at the wall drops to 409W (another 6W less) and with only 1 task running per GPU it drops to 390W (19W less). My GPU's power is capped at 100% so they are using 145W each (290W) and the rest of the system is therefore using 100W (but it is an i7).

In the same configuration, the 390W power usage drops to 370 when I reduce the CPU's clock from 3.7GHz to 3.1GHz - indicating a system (excluding GPU's) power usage of 80W.
With 4 GPUGrid tasks and 2 climate models running and the CPU at 3.1GHz the system power draw is 397W (18W less simply due to reducing the CPU clocks). If I tweaked the CPU voltage that might be reduced further, and I could drop the CPU clocks more. While that would probably not be productive in terms of system performance/Watt it might allow the system to run cooler and more quietly (if that was the desire).
If you go through the above procedure (adapted to your system and configuration) and run a task with different test setups you can work out what configuration brings the best Performance/Watt based on the runtimes.
7) Message boards : Graphics cards (GPUs) : 780Ti vs. 770 vs. 750Ti (Message 41435)
Posted 6 days ago by Profile skgiven
Robert, the custom vBIOS referred to below is for the GPU not the motherboard, so it won't help you out. Your only solution is a different motherboard/system.
8) Message boards : Graphics cards (GPUs) : Good and bad video card brands (Message 41434)
Posted 6 days ago by Profile skgiven
Ironic that the bad brand turned out to be a system OEM/motherboard manufacturer!
They are so desperate to retain their customers that they design their components to be incompatible with officially recognised standards. Half the time components don't even fit.
9) Message boards : Graphics cards (GPUs) : 780Ti vs. 770 vs. 750Ti (Message 41429)
Posted 6 days ago by Profile skgiven
1.5GHz sounds good but with only 1024MB of L2cache and a 128bit bus the 960 is badly bottlenecked and the high freq. comes at the cost of power.

Lots of good posts and valid opinions in this thread.
10) Message boards : Graphics cards (GPUs) : GTX 960 (Message 41428)
Posted 6 days ago by Profile skgiven
The power drop from 320W to 275W was simply explained by the GPU TDP/usage drop; 165 to 120 is 45W.
With respect to the GPU's this is a 37.5% drop but to the systems power usage it's a ~15% drop (275/320=0.859).
A 15% reduction in system power usage vs a >40% performance loss isn't good.


Next 10