stephenbrooks.orgForumMuon1GeneralStats List Anomaly
Username: Password:
Search site:
Subscribe to thread via RSS
TheFinalLoser
2002-08-19 08:37:06
Well, don't want to complain, but i have delivered several 1.xxx% results.
The odd thing is now that i just get not my peak percentage updatet, and it seems, that i get no bonus particles for these results.
on the other hand i get in the delivered result counter the amount of results added.
maybe you could check this problem.  I have send allways trough manualsend.exe and non of the files was bigger than 15kB.
I just don't know if this is now a Stats server lag or a bug or whatever.  its just somehow strange.  confused
Stephen Brooks
2002-08-19 12:21:51
How do you know if you have any bonus particles or not?  You only get bonuses for runs with >1%, so not all your results will have extra score.


"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"
gogomaus
2002-08-19 13:19:00
Hi Stephen,
just detected another strange scoring (hope this fits within _this_ thread).
Stats report on August, 19; 19:04 UTC
User (SG)ETA had 343 runs, 64 million "part." and a top result of 1.133%.

If I estimate 83 runs as random with typical some 30,000 particles this will give max.  3 mio particles to count.  Thus there remain 61 mio part.  for the other 260 runs -> some 235,000 "part." for each run.  As a calculation at the 1-% boarder typically simulates 45,000 particles, then _every_ run (of 260) must have crossed this bonus boarder !
Well, it could be possible at least, but it doesn´t fit the normal distribution nor my experience up til now.
Do you see a chance to check this ?
Of course in general, but above example was so easy to analyze.
Ciao
Stephen Brooks
2002-08-19 15:37:43
That user seems to have sent in the best250 as part of his results.txt.  I thought I had a device where results from best250 were rejected, but I'll check again.  Maybe this will reduce the S_Garbe peculiar scoring as well.

[update] Installed such a rejection scheme, recompiling stats...

[update2] OK, that particular anomaly has been cleared anyway.  I should modify the muon sending algorithm to weed out bestxxx data before even sending.


"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"


[This message was edited by Stephen Brooks on 2002-Aug-19 at 22:47.]
gogomaus
2002-08-20 03:10:18
Hello again,
congratulation to Stephen.  This was obviously an easy task.

Now I checked my personal last update.  I manualsent in 17 results this morning.
Before that my score was about 84.668 mio "particles" and after update I got a score value of 88.352 mio "part." --> I received 3.684 mio "part." in addition.

Evaluation of my results.txt copy showed 12 results <1% totalling 0.437 mio part.  and 5 results nearly above 1.2% with originally 0.225 mio part.  in sum.
The latter one shoulb be "bonified" as 1.350 mio part.  (factor = 6).
So I would have expected a scoring of some 1.787 mio "part." for this package, but got roughly the twice.
Isn´t that strange ?
As I understood from another thread, up til now your complete database will be recalculated totally at refreshing time (hourly).
So at least one of the calculations must have been wrong.

Sorry to cause inconvenience

At least I´m happy to claim my personal "advantage" and not anyone else.
Stephen Brooks
2002-08-20 04:00:51
Unless some mystery person is sending in results as "gogomaus", there must be some excess scoring happening.  I will look at the stats generator a bit more on the stats computer when I get home.
Meanwhile, I am just testing the corrected commandline client (apparently it failed because the degree symbol ° is treated as whitespace in the console locale, so the angles in the chicane got read in as radians, meaning it was impossible for particles to navigate).


"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"
gogomaus
2002-08-20 04:50:10
Sorry, I forgot to complete my "report". Number of results (at least under v4.2x) has been o.k.; i.e. 17 new results were added correctly.  This doesn´t look like an unknown sponsorship, but I trust you will find it out as usual, as soon as you have time and opportunity.  Bye and good luck...
TheFinalLoser
2002-08-20 05:17:55
Well, i have madde a Copy of my results file, and here are the peaks out of it:

1.234608 (43763 particles) [v4.21b] {B75C17BC}
1.353657 (44976 particles) [v4.21b] {B75C17BC}
1.308368 (43782 particles) [v4.21b] {B75C17BC}
1.289528 (43772 particles) [v4.21b] {B75C17BC}

As i have understood i must get bonus particles for these results. 
Some about 220,000 Particles per result with a multiplier 5.
thats almoust 900,000 Paritcles more and a new Top Score.  But the rank still says:

192. [SG]TheFinalLoser - - - 207 7`416`791 (.049%) 0.993% (63rd)

Two days before the Statistic sayed:

182. [SG]TheFinalLoser - - - 196 7`037`940 (.056%) 0.993% (34th)

at least the peak muon transfere should be changed.
So that is realy strange.  You could maybe check this.  if you want i mail you the results.txt
Bye TheFinalLoser
: contact : - - -
E-mail: sbstrudel characterstephenbrooks.orgTwitter: stephenjbrooksMastodon: strudel charactersjbstrudel charactermstdn.io RSS feed

Site has had 25162488 accesses.