The last (at least) 4 work units I've had have finished with Compute Errors after running for a couple of hours. Usually the screen saver has crashed on my computer as well. Does anyone have any idea why this might be happening? It seems like a waste of my time to compute for almost 9 hours just to have a compute error over and over.

---

The last (at least) 4 work units I've had have finished with Compute Errors after running for a couple of hours. Usually the screen saver has crashed on my computer as well. Does anyone have any idea why this might be happening? It seems like a waste of my time to compute for almost 9 hours just to have a compute error over and over.

also, the error in my messages reads:
1/2/2007 1:16:20 PM|rosetta@home|Unrecoverable error for result s028__BOINC_ABRELAX_NEWRELAXFLAGS_hom001__1461_26280_0 ( - exit code 1073807364 (0x40010004))

---

Does it help if you set your screensaver to 'blank' (display properties)?

Seeing if there are any updates to your graphics card drivers may also help.

---

I actually just updated the drivers last night, before the crash this morning. I honestly am not sure how the screensaver could be causing the crash (i figured it was just a symptom), but I suppose I could do that. Is it possible to disable the screensaver for only one project though?

---

Well the screensaver is a bit of a waste of CPU time on every project... there is the 'show graphics' option to see how things are going on.

---

I got a compute error on a WU that ran 3 mins and crashed.
Posted the error messages up in the 5.43 forum.
No screen saver, wasn't even home at the time when it crashed.

---

Yours was an exit code 1, which is something different (often it's Windows killing the task, possibly due to an automated reboot, or something similar).

---

<i>Well the screensaver is a bit of a waste of CPU time on every project... there is the 'show graphics' option to see how things are going on.</i>

hmm. I had watched my CPU cycles with the task manager with the Show Graphics both on and off and didn't see any noticable difference in the number of cycles being used by the project, for the few projects I had tested that with. I assumed that the screensaver graphics computation stuff was being handed off to my GPU to deal with so it wouldn't affect the actual project. Maybe I'm wrong about how that part works though.

---

Yes, the error reported is consistent with the known screensaver problems.

Yes, the recommendation is to disable the screensaver.

No, there is no way to do that JUST for Rosetta. It is either BOINC screensaver, or not, so... "not" is recommended.

Yes, using any screensaver diverts some CPU time from the project. The graphics processor does some of the work, but it still must be told what to do, and that takes time.

The reason you don't see any difference when running with graphics is because the graphics runs as a thread under the Rosetta task, so it basically looks like part of that. As an experiment, if your machine is able to run with the screensaver... jot down the following:

Current CPU time for all processing WUs.
Current time, to the second as reported on your PC clock.
... then crunch for several hours with screensaver active, and leave your PC alone, so screensaver is running for those several hours.
Now check the CPU time recorded for all WUs and the current PC clock.
Now, how much time elapsed on your PC clock? And compare this to the total time recorded crunching WUs. You will find there are a measureably large number of processing seconds gone. Some of this is due to other Windows tasks, some is due to the screensaver thread. Do same test with screensaver set to NONE and you'll have an idea how much it costs you in performance.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/