i have left SETI running task on ALL the time and never had any problems, but since installing CUDA the time to completion has reduced amazingly, thats the good thing... the bad thing is that it doesn't even countdown, while astrupulse is counting down and working fine the SETI task is counting UP and doesn't stop or even do the task.. just seems like while 1 task is running the other isn't, I've aborted a few task to see if it was maybe an error... but still the same. ill post a picture of my screen. it counts up about 3 seconds to every normal second, so that why its so high.. but it started off at about 34mins...

basically astropulse counts down and works... seti counts up and doesn't.

---

When using the GPU to crunch, the BOINC scheduler does not yet count the seconds appropriately, so the task scheduler has to wait for the next checkpoint to see how much work was done and then guess as to how much time is left. The seconds/minutes you're seeing are total CPU seconds/minutes that the CPU was used to feed the GPU and not the actual time spent by the GPU crunching.

---

righto, thanks for that, I'll keep it running for a few more days and see what happens.

---

CUDA-accelerated crunching of VLAR WUs (Very Low Angle Range workunits) with version 6.05 should lead to frozen percentage and counting up time.

To flawless crunching abort all VLAR WUs please - those indicated by the line "<rsc_fpops_est>80360000000000.000000</rsc_fpops_est>" in the "C:\Program Files\BOINC\Data\client_state.xml" file...

<workunit>
<name>03dc08ah.2110.5798.10.8.69</name>
<app_name>setiathome_enhanced</app_name>
<version_num>605</version_num>
<rsc_fpops_est>80360000000000.000000</rsc_fpops_est>
<rsc_fpops_bound>803600000000000.000000</rsc_fpops_bound>
<rsc_memory_bound>33554432.000000</rsc_memory_bound>
<rsc_disk_bound>33554432.000000</rsc_disk_bound>
<file_ref>
<file_name>03dc08ah.2110.5798.10.8.69</file_name>
<open_name>work_unit.sah</open_name>
</file_ref>
</workunit>

i aborted the tasks with 03dc08ah* if thats what you ment, and still the same.
the progress of the first tsk is now 0.001% and is frozen again with time to completetion going up still... is this a problem with cuda or my drivers etc?

---

All tasks with line "<rsc_fpops_est>80360000000000.000000</rsc_fpops_est>" aborted?

I am aborting all VLAR WUs periodically since 23 Dec and had no freezing...

On my machine one free CPU core sets CUDA (9600GT 512MB) with crunching for a time of about 21 seconds (during that percentage is equal to 0%) - after that percentage grows up and WU is processing normally...

[I'm sorry about my poor English - it's non-native for me]

I don't think its wise to suggest people to start deleting files and removing them from an XML file, especially when you could just use the abort button from within BOINC Manager.

---

well good job i haven't deleted anything in the XML file, i thought it couldn't be good, but basically aborted all the WU and same problem persists. astropulse works perfectly, but seti tasks just seems to continuously go up and up and up in completion time, they don't seem to be doing anything...and takes a few times of loading BOINC for it not to crash as well... tried re-installing and same thing. is there any information you need to help me work this out, because apart from my explanation i don't know what to suggest, i can't run boinc because my seti tasks get no where and 2/3 of the time boinc crashes out pretty much. i want to figure this out so i can get boinc up and running 24/7 again.

---

typical, its working now.

well ill not complain and get crunching.

thanks!!

---