Checking for SETI instablity on new system

[Hyperfocal]

Is there a way to ensure that a new system is properly processing WUs?

I just upgraded my old box from a Celeron 900 to a Duron 1.6 oc'd to 2ghz. It really seems to rocket through the WUs compared to the old celery with times of under 3 hours.

The question is that one unit took only 1:30 and another seems to have been dumped after only 2 minutes. I just ran Prime95 for 6 hours without an error and the system is running 43 degrees under full load with SETI and 45 degrees under P95.

Is there anything to look for in the .csv file or any other indicator to show that there are problems with the submitted WUs?


Thanks

 

[lobadobadingdong]

those are likely rejects, or very very high/low angle range (can't remember which ones are increadably fast, i'm thinkin they are the high AR's though) WU's just eat em' up and love em.

 

[Smoke]

This is your production for the last 7 days:

Rank (Chg) (Day,Mo,Year) WUs
1833 (22) 18.02.2004 7
1855 (7) 17.02.2004 3
1862 (10) 16.02.2004 3
1872 (6) 15.02.2004 2
1878 (5) 14.02.2004 2
1883 (14) 13.02.2004 3
1897 (2) 12.02.2004 2

So far things look good. Does this seem to match what your SetiLog file reports?

 

[Hyperfocal]

That matches, so I think it may have just been a fluke. The times have been more consistent since the 2 minute WU.

I'm really amazed at the performance from that Duron. It is consistently in the 2:45 range. Not bad for getting an MB, CPU and 256MB ram for $105 total.

 

[silverpig]

Every once in a while seti injects test signals into their data, or the receiver picks up signals from sattelites etc. These units have a lot of gaussians and peaks with great scores and high powers that could not come from another solar system. If your analysis turns up a lot of excellent high power gaussians, then the program just writes it off as crap and gets another unit (you still get credit though). I'm not at my seti computer right now, but most of my units go to 99-100% with a few others at seemingly random percentages. If I look at the ones that only got 5% of the way through in setispy's report viewer, a lot of high powered gaussians turn up.

I think you can read about this on the seti website.

 

[silverpig]

SETI FAQ: Why do some units take so little time to complete?

Occasionally, a work unit will contain strong radio interference; these strong interfering signals typically come from satellites and radar from our own civilization. If the interference is very strong, the SETI@home program can not analyze that part of the spectrum, and after trying for a few minutes and detecting thousands of strong signals of earth origin, the program stops early in the processing and gets a new work unit. You will still get credit for the work done.

 

[Hyperfocal]

Thanks. It was apparently just a strange coincidence that those WUs were cached and ready to crunch as some of the first WUs with a new CPU.

I'd always assumed they used their own system to avoid sending out WUs with strong terrestrial interference. Someday, I'll RTFM.

 

[silverpig]

Originally posted by: Hyperfocal
Thanks. It was apparently just a strange coincidence that those WUs were cached and ready to crunch as some of the first WUs with a new CPU.

I'd always assumed they used their own system to avoid sending out WUs with strong terrestrial interference. Someday, I'll RTFM.

It's not really all that strange. It happens to me quite often.