http://setiathome.berkeley.edu/workunit.php?wuid=1489726012

I'm the _0. The _1 and the _3 validated. I'm just wondering if AP is also suffering from "two bad results validate each other and invalidate the good one(s)" syndrome.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

http://setiathome.berkeley.edu/workunit.php?wuid=1489726012

I'm the _0. The _1 and the _3 validated. I'm just wondering if AP is also suffering from "two bad results validate each other and invalidate the good one(s)" syndrome.

I am not sure that would apply here. If this was validated against the _2 then defiantly. That machine went wonky on the task.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

http://setiathome.berkeley.edu/workunit.php?wuid=1489726012

I'm the _0. The _1 and the _3 validated. I'm just wondering if AP is also suffering from "two bad results validate each other and invalidate the good one(s)" syndrome.

It's probably a case of three scientifically valid results with yours being zapped by AP validator quirks. The validator always compares repetitive pulse signals, but ignores single pulse signals which are not at least 1% above threshold. But single pulses it doesn't ignore usually have twice the weight of a repetitive pulse because the single pulse is also reported as a best_pulse.

From the stderr of your task 3517401417 we see:

Rep. pulse: peak_power=3188; dm=-2048; fft_num=0; scale=5040; ffa_scale=4; period=-7.6097e+158
Single pulse: peak_power=130.1; dm=-7943; scale=6
Single pulse: peak_power=651.7; dm=-13646; scale=9

The validator would ignore the single pulse at scale=9 because it's only about 0.05% above threshold. But the single pulse at scale=6 is very near that 1% above critical level. The threshold is 128.857971, and the rounded 130.1 shown in stderr could actually be anything from 130.050000 to 130.149999 (0.0925% to 1.0026% above threshold). We can't tell for sure whether the validator ignored that one or not, but it's fairly likely that it was ignored. The wingmates' results could easily be on the opposite side of that critical level with only a tiny difference in calculation.

What it comes down to is if the scale=6 single pulse is not ignored, the validator is checking 3 signals from the result, but if it is ignored there's only the 1 repetitive pulse. Since 1 is less than half of 3, there's no way to get credit on a "weakly similar" basis.

                                                                  Joe

http://setiathome.berkeley.edu/workunit.php?wuid=1489726012

I'm the _0. The _1 and the _3 validated. I'm just wondering if AP is also suffering from "two bad results validate each other and invalidate the good one(s)" syndrome.

It's probably a case of three scientifically valid results with yours being zapped by AP validator quirks. The validator always compares repetitive pulse signals, but ignores single pulse signals which are not at least 1% above threshold. But single pulses it doesn't ignore usually have twice the weight of a repetitive pulse because the single pulse is also reported as a best_pulse.

From the stderr of your task 3517401417 we see:

Rep. pulse: peak_power=3188; dm=-2048; fft_num=0; scale=5040; ffa_scale=4; period=-7.6097e+158
Single pulse: peak_power=130.1; dm=-7943; scale=6
Single pulse: peak_power=651.7; dm=-13646; scale=9

The validator would ignore the single pulse at scale=9 because it's only about 0.05% above threshold. But the single pulse at scale=6 is very near that 1% above critical level. The threshold is 128.857971, and the rounded 130.1 shown in stderr could actually be anything from 130.050000 to 130.149999 (0.0925% to 1.0026% above threshold). We can't tell for sure whether the validator ignored that one or not, but it's fairly likely that it was ignored. The wingmates' results could easily be on the opposite side of that critical level with only a tiny difference in calculation.

What it comes down to is if the scale=6 single pulse is not ignored, the validator is checking 3 signals from the result, but if it is ignored there's only the 1 repetitive pulse. Since 1 is less than half of 3, there's no way to get credit on a "weakly similar" basis.

                                                                  Joe

Thanks for the explanation. I almost understand it. :)

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

I'm a total noob to boinc but was going through my results and noticed about half were invalid or inconclusive.
http://setiathome.berkeley.edu/results.php?userid=10000434&offset=0&show_names=0&state=5&appid=
I don't have a very strong rig,so i disabled cpu and been running my gpu overclocked just under the crashing point on furmark. Also, i have changed setting to never suspend work so it been running while doing light gaming and web browsing. So is me me just stressing my system to hard causing errors or is there something else I should be looking into.
Thanks
p.s. i have restored my gpu to factory and plan to suspend work if i do anything else to see if it helps.

I'm a total noob to boinc but was going through my results and noticed about half were invalid or inconclusive.
http://setiathome.berkeley.edu/results.php?userid=10000434&offset=0&show_names=0&state=5&appid=
I don't have a very strong rig,so i disabled cpu and been running my gpu overclocked just under the crashing point on furmark. Also, i have changed setting to never suspend work so it been running while doing light gaming and web browsing. So is me me just stressing my system to hard causing errors or is there something else I should be looking into.
Thanks
p.s. i have restored my gpu to factory and plan to suspend work if i do anything else to see if it helps.

When posting about your tasks you need to do it my computer. No one can see your user task list.
So like this: http://setiathome.berkeley.edu/results.php?hostid=7282215&offset=0&show_names=0&state=5&appid=
Trying to do tasks while heavy GPU actives like gaming might not be the test for your system if you want your tasks to be processed without errors. It is kind of like asking your car to drive at top speed while pulling a heavy load uphill.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

is this right: http://setiathome.berkeley.edu/results.php?hostid=7282215&offset=0&show_names=0&state=5&appid=
and i kinda figured it might be my fault just wanted to get a little info from someone who's been at it for more than a week thanks.

is this right: http://setiathome.berkeley.edu/results.php?hostid=7282215&offset=0&show_names=0&state=5&appid=
and i kinda figured it might be my fault just wanted to get a little info from someone who's been at it for more than a week thanks.

Given that you said you have the GPU overclocked near the hairy edge could be it as well. It could be to much overclock to run the science apps.
I would try not running it with games first & that solves it then you are good. If it happens at other times you might have to turn the dials down a bit to not get invalid results. Trial and error really.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[