LHC@home

I think there is something wrong with this host(it's not mine): http://lhcathome.cern.ch/show_host_detail.php?hostid=64455

This host finishes every workunit it gets within a few seconds, it happens with every single workunit, even those that take hours on the other hosts. It seems that it is erroring out but it reports every result as a sucess...

What's going on?

---

BOINC.BE: The team for Belgians and their friends who love the smell of glowing red cpu's in the morning

Well looking closely, most results show status pending, though a few WUs he does have credit. Others however like this one

http://lhcathome.cern.ch/workunit.php?wuid=715061

everyone EXCEPT for this comp had gotten credit. I'm actually wondering if this person has a cheat of some sort running. Even with my A64 benching well above what the benchmarks state for his computer's processing ability, it takes me longer then that to complete a unit (of course).

At least he isn't getting much credit for this...

---

<blockquote>Well looking closely, most results show status pending, though a few WUs he does have credit. Others however like this one

http://lhcathome.cern.ch/workunit.php?wuid=715061

everyone EXCEPT for this comp had gotten credit. I'm actually wondering if this person has a cheat of some sort running. Even with my A64 benching well above what the benchmarks state for his computer's processing ability, it takes me longer then that to complete a unit (of course).

At least he isn't getting much credit for this...</blockquote>

I was reading in a thread, within another Project, of which I have forgotten, that in the old days of SETI@Home “Classic” users would copy some of their previously completed W/U’s, edit the W/U file, then somehow get the SETI-Classic to reset (like “Restart” we have now with Boinc) and when SETI came alive again, it would take the falsely edited W/U and start from that point to process W/U. Hence, a high figure of merit on an already completed unit in a short amount of time.

[b]Within this same thread, Boinc was to correct this software fault used by “Cheaters” ???</B>

TA

---

<a href="http://sciencebulletins.amnh.org/astro/f/gravity.20041101/">Theory of Gravitational Waves &amp; LIGO</a>
<a href="http://lisa.jpl.nasa.gov/">Laser Interferometer Space Antenna - LISA</a>
<a href="http://www.jpl.nasa.gov/">JPL-Caltech</a>

There were other tricks, if you had a long list of work, you could run them almost to completion, save them off, when you had a big stack, you could feed them to multiple computers as almost done, so, for the time to process one work unit you could multiply the results across "n" computers.

There were other tricks like repeating the above a number of times. Since Classic did not track work units to the participant and computer, you could get real high numbers real fast. It kinda makes me wonder if some of the very vocal people about how the old system was so much better, well, if that was the reason why ... now they will have to be honest.

It is kinda sad as all they are cheating really is themselves, and humanity of the possible science that might have been done ...

---