stephenbrooks.orgForumMuon1Generalstats confused
Username: Password:
Search site:
Subscribe to thread via RSS
pvs
2002-09-01 05:08:57
Stats seem to be a little bit confused (11:00 12:00)

Pit
[SG] DOA
2002-09-01 05:23:55
A little bit?  *sncr*
Either [ARS] kenlow is massproducing (can't be, due to the number of results) or is producing 'lil masses with high yields (can't also be due to the "average" muon yield).

Anyway, he's got 1000 billion results - quite a bit... *gg*

Gerald
Kaede
2002-09-01 06:47:03
very suspect to me - 1078 billion with only 2,58 as best result
TheFinalLoser
2002-09-01 08:06:51
well erm and its not realy sorted with the particle count eighter.....
[Ars]kenlow
2002-09-01 08:59:17
quote:
Originally posted by [SG] DOA:
A little bit?  *sncr*
Either [ARS] kenlow is massproducing (can't be, due to the number of results) or is producing 'lil masses with high yields (can't also be due to the "average" muon yield).

Anyway, he's got 1000 billion results - quite a bit... *gg*

Gerald


ARS is no.  1!!!  big grin big grin

Guess that we should enjoy it while it lasts smile wink

Not sure what is going on...but the stats are really messed up.  And I am not mass producing....but I do have a hoard though smile Last time I checked a few hours back...I had about 875 million..

Stephen....u there?
Bluumi [SwissTeam.NET]
2002-09-01 09:43:38
The stats is a very nice joke in the Moment big grin

Some thing "mix it"

[ARS]kenlow has (5944.77%) of the Results ... If 100% is all, what is 5945%
..... big grin

Its so crazy... that its best joke i see long time cool
[SG] DOA
2002-09-01 10:27:57
Well - this just means he alone finished this project and 60 others while walking by... *lol*
Stephen Brooks
2002-09-01 14:06:09
Erm... uh, yes.  Sorry.  The stats went a bit surreal for a bit there, but just as I was going to fix it I had to go and drive 100 miles away to visit a relative.  I was having nightmares that it was the stats-checker's fault (AGAIN), but it turned out Kenlow just had ended up submitting the best250 several times inside his results.txt, and the extra linefeed some people had noticed had messed up the stats counting.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
Bluumi [SwissTeam.NET]
2002-09-01 14:46:52
quote:
Kenlow just had ended up submitting the best250 several times inside his results.txt,


Hmm... he send the best250 more than once?!?  Several times???  Hmm...

Oh he send not only ONE result.txt .. but in every he send the whole best250?

Nice idea to make a crazy stats big grin

And... Hey, stephen.. why you drive away... you must keep all day by the stats big grin There is NO Freedom left now.. wink

Nice it work again... but who think you can't fix.... roll eyes

Good work.
gogomaus
2002-09-01 14:57:33
I really wonder, why someone is adding/including the "best 250" within his result.TXT ???  It´s not necessary and has been pointed out repeatedly to avoid this.
Please be a little bit more careful and take regard of Stephen, who deserves some less trouble... wink
[Ars]kenlow
2002-09-01 18:07:09
quote:
Originally posted by Stephen Brooks:
Erm... uh, yes.  Sorry.  The stats went a bit surreal for a bit there, but just as I was going to fix it I had to go and drive 100 miles away to visit a relative.  I was having nightmares that it was the stats-checker's fault (AGAIN), but it turned out Kenlow just had ended up submitting the best250 several times inside his results.txt, and the extra linefeed some people had noticed had messed up the stats counting.



Hi Stephen,

Don't know how the Top250 results got into the results.txt file as I only added it into the results.dat file.  And I have been using the new top250 file from when u first released it.  Since then, the systems have been submitting results automatically everyday, twice a day.  Funny that the Top250 should suddenly get included after a few days confused confused

And the machines that submitted results at 1400 UTC are all office machines.  1400UTC Sunday is 2200hrs Sunday for me, no one would have access to the office server room during the whole weekend.

Any ideas?
Pascal
2002-09-02 00:22:33
It's nice to see the stats working properly again.

___________________________
Member of www.rechenkraft.net - German Website about Distributed Computing Projects

1: Athlon TB-C, 1.2 GC/s, 256 MB DDR-RAM, Erazor x², ADSL-Flatrate, NIC Intel, Win 98 SE Mainboard MSI-6380 Rev.  1
2: Pentium III, 600 MC/s, 256 MB RAM, NIC Intel, Win 98 SE
Stephen Brooks
2002-09-02 02:19:11
There is no mechanism whereby the best250 (including the "Rank..." comments) can get into my database unless you accidentally opened results.txt instead of results.dat on one of the times you were sending results in.  You may not have noticed you'd done this because the inclusion of the best250 immediately made the .txt file big enough to send, so it would have disappeared from your HDD.  I guess mistakes are unavoidable if you're doing the update on many different computers.

The source-code I have now has an improved results-checker unit that deletes accidentally-included results from the best250 before sending.  I'll be thinking about releasing another version in a week or so.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
[Ars]kenlow
2002-09-02 05:37:06
quote:
Originally posted by Stephen Brooks:
There is no mechanism whereby the best250 (including the "Rank..." comments) can get into my database unless you accidentally opened results.txt instead of results.dat on one of the times you were sending results in.  You may not have noticed you'd done this because the inclusion of the best250 immediately made the .txt file big enough to send, so it would have disappeared from your HDD.  I guess mistakes are unavoidable if you're doing the update on many different computers.

The source-code I have now has an improved results-checker unit that deletes accidentally-included results from the best250 before sending.  I'll be thinking about releasing another version in a week or so.




The thing is that all my machines are scheduled to upload their results automatically twice a day.  No human intervention is needed and files uploaded are therefore always less then 100KB.

And as for updates, I always unzip and replace the results.dat file.  I implemented tha latest Top250 on Aug 29 and the machines have been uploading results twice a day since then.  So it cannot be a case of appending the Top250 to the result.txt file.

The rouge result.txt file that was uploaded was done on a Sunday.  When I came into the office on Monday morning; I checked the log and nobody has touched the machines over the weekend. 

The only thing that was abnormal was that 1 of the servers rebooted itself at 0151UTC, that is 1 minute after it uploaded its results.  Maybe there is a connection...

Anyways..I have checked all my machines and made sure that everything is as it should be.

I am sorry that my side caused you so much trouble and everyone else a few hours of stats outage.  Rest assured that it is not intentional and that I have done all I can to prevent it from happening again.
[Ars]kenlow
2002-09-02 06:16:11
Just thought of something as I was driving home.  How do you know that the results that was uploaded came from my machines?  I mean I could easily create a bogus user.txt file and then upload the top250 results and no one will know that it was from me.

It is just a thought and I am making no accusations smile
: contact : - - -
E-mail: sbstrudel characterstephenbrooks.org
Yahoo: scrutney_mallard
Jabber: stephenbrooksstrudel characterjabber.org
Twitter: stephenjbrooks

Site has had 16092637 accesses.