GENOME stats correcting coming 26. October!!!!!

[Rattledagger]

For full post, goto http://groups.yahoo.com/group/genomeathome/message/3418

For feedback & suggestions, use the same group.

For a probably result after the stats-corrections for us, look at http://gah.stanford.edu/teams2/MatroxUsers.html.

Now, there's atleast some bugs here, since some users has returned many genes, but hasn't got any wu-credits, and summing the wu gives 335k, but on the top we've got 601k.

But in any case, it looks like some will lose many wu.

[Jorden]

If it makes me number 13 in a weeks time, I'll be okay with it

[Rattledagger]

Originally posted by Jorden
If it makes me number 13 in a weeks time, I'll be okay with it

Doesn't you fall down to #32 now? If you're talking about top genome-cruncher, I'll not produce a new list before November.

[breezer]

Phew!
I'm listed at 904 in the new stats, don't even make the top 1000 in the old

breezer

[Wulfman]

what? 50% lost??!

ok, I am always using the latest client, had a max. of 3 or 4 duplicate genes (moving clients around after formating etc...)

so wtf is going on???

mfg
wulfman

[TomNuttall]

Aarghh!! I have lost about 3 thousand points.

I have done NO copying of files onto new machines, just straight crunching.



It must have been the -nonet that did it

Tom

[Liquid Snake]

Damn I'm going to lose almost 10,000 WUs!!! But then Nordlichter will lose almost all of their WUs and we'll be back to #7

I didn't do too much noneting. I did some, but I don't think it would've been enough for me to lose almost 10,000 WUs!!

[Guru]

Team_DGC is loosing 3k and I dont understand why????? Someone please explain to me what is going on here!

[Rattledagger]

Liquid Snake, the wu's reported in the previous link is really screwed up. Atleast for now, the correct is to compare top 1000 users:
http://gah.stanford.edu/userstats.html with the new:

http://gah.stanford.edu/userstats2.html

From these, you're losing around 1300 wu.

Guru, it looks like all results is calculated using a new wu-formula, and this formula leads to a 5% hit. Atleast this will probably be fixed. At the same time, wu done with the linear wu-formula will get a bigger hit.

[rocketmanx]

Looks like I'll lose about 4000 wu's, 'bout the same as box
I didn't know nonetting was a bad thing. If we are going to lose wu's because of it why is the option their?
I have no choice but to nonet, if I didn't my home phone would be constantly busy with 7 machines running. So if I lose wu's because I nonet looks like I might just have to give-up D/C projects and sell some machines. It's not what I want to do but... Or am I misunderstanding the corrections and how nonet affects the way stats are going to be done?
Pete

[breezer]

I've just upped my output and find i've lost the units

• old 870 breezer 6058.17
  new 830 breezer 5665.72

breezer

[Jorden]

LOL RD, I was in the top1000 around 983, so I haven't got a clue where I went, but I still retain some 5125 points.

Too low to be in the top 1000

Jord.

[Zyn]

No problem here, as I am running SETI and not GeNome.
Thank you.

[Liquid Snake]

Oh, only 1300, that's not that bad. Wow Hoppe lost more than 600,000 WUs!!!

[Rattledagger]

As for Hoppe, because of large production he have got some friends helping him with uploading, but he forgot to include all 3 result-files... (result.#####, csum.##### and str.#####)

Jorden, since you're not in top1000, you've got below 4500...

rocketmanx, didn't you lose nearly 7k?

A quick and dirty table of top losers, from 17. October, 18PDT. Note, since it's only top1000 that works, persons that aren't on both isn't counted...:

Ant; 30,4%; 3,3k
Ees; 18,2%; 2,5k
Narcissus; 17,9%; 1,5k
winwiz; 17,4%; 1,4k
Pace; 16,7%; 1,2k
cbman; 13,4%; 1,0k
CHHAS; 12,5%; 719
Manex; 11,6%; 1,0k
TomNuttall; 11,4%; 3,0k
rocketmanx; 10,6%; 6,8k
guyver; 10,1%; 6,0k
lurqa; 9,2%; 1,2k
DentyCracker; 9,1%; 2,3k
LittleGreenMartians; 8,5%; 2,2k
Spazm; 8,2%; 738
MarkF001; 8,0%; 2,5k
Team_DGC; 7,9%; 3,6k
LiquidSnake; 7,3%; 1,3k
Acke; 7,2%; 675
k1l; 7,1%; 949
Sisyfos; 6,4%; 3,4k
breezer; 6,1%; 318
stardock; 5,8%; 911
box; 5,4%; 3,8k
Hecubus; 5,2%; 1,4k
Buskov; 4,6%; 260

Hoppe; 95,97%; 629,4k .....