Rosetta@Home - 12/21: Stats

[BlackMountainCow]

TA[R@H] is 84 members strong.


Need some stats? Take a look at Free-DC, BOINC-Stats and BOINC-Synergy

Got some questions? Try the Official R@H FAQ (work in progress).



Yesterday's highest overall climber: Shame (8,761 spots)

Yesterday's highest TeAm climber: Coquito (4 spots)



Yesterday's production for TA: 52,033 (- 227) :thumbsdown: (If this is the way we "lose" production, we're in pretty good shape 😉 )

TA's total score: 1,453,596 creds

TA's world rank is: #7



Upcoming Threats: none

Upcoming Conquests: BOINC Synergy (3 days) & Catalyst (20 days)




Milestone Makers:

TA_TheReaper passed 75,000 creds :Q

Kruul passed 10,000 creds 😀

jw.middleton passed 5,000 creds 😛




Yesterday's top ten producers:

• Rebel Alliance (11,698 creds)
  
• TeAm Enterprise (7,715 creds)
  
• BadThad (4,214 creds)
  
• Sofa King (3,249 creds)
  
• TA_TheReaper (2,513 creds)
  
• OhioDude (2,479 creds)
  
• jw.middleton (1,798 creds)
  
• TA_JC (1,448 creds)
  
• mrwizer (1,445 creds)
  
• Insidious (1,159 creds)

Mini Race to 75k:

• #5 (+2) ... mrwizer ... ... 70,239 creds ... + 2,027 :beer:
• #6 (-1) ... rise4310 ... ... 62,657 creds ... + 1,083
• #7 (-1) ... petrusbroder ... 59,102 creds ... + 914

Wow, looks like Sean really wants to bring this race to an end. First time he produced 2k a day during the race. And Rise's back as well, 1k again. Actually, that's the best day the three of you had in this race so far: combined output of over 4k! :thumbsup:



SitRep:

• We're leading US-D by over 105k now. The distance to BOINC Synergy is down to 3 days (- 1.7 days). We also have a new conquest, Catalyst (20 days) :evil:
  
• Congrats to our milestone makers TA_TheReaper, Kruul and jw.middleton :thumbsup:
  
• The Rebel Alliance produced 11,698 creds yesterday. That's the first time a member of TA crossed 10k per day. Nice job guys! 😀
  
• There are still some of these bad WUs around but they're getting fewer and fewer.

:beer:😛:beer:

BMC

 

[Crazee]

Thanks BMC 🙂

3 More P4 HT 3.0GHz machines added today. More likely soon 🙂

Congrats to the milestone makers 😀

 

[GeoffS]

WTG TeAm!! :thumbsup:

I realized, looking at the stats (actually, just taking a sniff of the air), that we once again find ourselves downwind of a herd of cows... mmmmmm... a midwinter BBQ would be soooooooo goooooooood! lol!

Geoff

 

[mrwizer]

Originally posted by: BlackMountainCow

Yesterday's top ten producers:

• Rebel Alliance (11,698 creds)
  
• TeAm Enterprise (7,715 creds)
  
• BadThad (4,214 creds)
  
• Sofa King (3,249 creds)
  
• TA_TheReaper (2,513 creds)
  
• OhioDude (2,479 creds)
  
• jw.middleton (1,798 creds)
  
• TA_JC (1,448 creds)
  
• mrwizer (1,445 creds)
  
• Insidious (1,159 creds)

It is getting harder and harder to be in the top ten it seems. Thats always good though...

Mini Race to 75k:

• #5 (+2) ... mrwizer ... ... 70,239 creds ... + 2,027 :beer:
• #6 (-1) ... rise4310 ... ... 62,657 creds ... + 1,083
• #7 (-1) ... petrusbroder ... 59,102 creds ... + 914

Wow, looks like Sean really wants to bring this race to an end. First time he produced 2k a day during the race. And Rise's back as well, 1k again. Actually, that's the best day the three of you had in this race so far: combined output of over 4k! :thumbsup:

Yeah, I thought it might be fun to hit 75k a bit early 😉 And good for Rosetta but bad for Folding, I had to take some machines off of the December race due to proxy issues. So I had a few extra machines to add to the race. 🙂

[*]The Rebel Alliance produced 11,698 creds yesterday. That's the first time a member of TA crossed 10k per day. Nice job guys! 😀

Way to go Alliance. Nice to see a producer over 10k. 😉

 

[Slatzman]

Thanks for the stats BMC...another good day of production, hopefully this or higher will be the norm!!!

Nice additions Crazee!!!

Slatz

 

[winr]

I am going to add an XP 3200+ to R@H tonight or tomorrow.


Thanks for stats.


🙂

 

[Insidious]

Thanks for some awsome stats BMC

Congrats to the milestone makers today. :thumbsup:

...gotta go finsih the Santa stuff....


-Sid

 

[Smoke]

Congratulations for breaking the 10,000 daily mark, RA. :beer:

Way to go TeAm!

During the Christmas season we, TeAm Enterprise, should find some nice stocking stuffers under our trees. 😉

Thanks for the thread, BlackMountainCow. :thumbsup:

 

[Freewolf]

Originally posted by: Smoke
Congratulations for breaking the 10,000 daily mark, RA. :beer:

Way to go TeAm!

During the Christmas season we, TeAm Enterprise, should find some nice stocking stuffers under our trees. 😉

Thanks for the thread, BlackMountainCow. :thumbsup:

Gotta love those stocking stuffers. I'm Looking farward to a few myself 🙂

Great Job Everyone !
Nice addition there Richard !

 

[GimpyOne]

Thanks for the stats! It looks pretty good and congrats to all!

A quick thing I found yesterday.(or not so quick) I was gone last weekend and since I've been back, I noticed I was producing about 50ppd less then before I left. So I started poking around and if I read things correctly, found that credit/WU is calculated something along the lines of:

(floating point test + integer test)*seconds to complete WU
-or at least that is the general gist of things

Anyway, whenever the last time that Boinc benchmarked while I was out, the comptuer must have been doing something because while my floating point number was fine, the integer was less than half of what it should have been.(or I had it with me on battery as it is a laptop) So I went from ~180ppd down to ~130ppd. After manually running the benchmark last night it appears to be back where it belongs.

So I guess the moral of the story is if you can keep an eye on the floating point and integer numbers to make sure that it isn't doing things it shouldn't be. And for the record, I am not using any sort of optimized client, or overclocking. Everything is stock across the board.

Anyone have any idea how often Boinc benchmarks? Maybe manually preempting it could prevent this in the future. Sorry for the long post, especially if this has been discussed before. :beer:

-Brad

 

[Insidious]

Actually, you can re-benchmark your host anytime. It's under the 'commands' menu in boinc manager
then next time you update the new benchmarks will be recorded. If you don't do anything, it does it's own re-benchmark @ 5 days (IIRC)

-Sid

 

[rise]

thanks for the stats BMC :beer:

congrats to the MM's and RA for the huge day and :beer: for all 🙂

 

[JWMiddleton]

Thanks for the stats BMC!

Leaving for a week and hope the fleet stays up! Off to sunny Florida! WOOT!

Happy Holidays everyone!

 

[Dalephi]

Thanks BMC.

Another awesome day.

Now that SETI Classic is really, really, really over I can concentrate on Rosetta more.

Congratulations to the Milestone Makers and the awesome output by the RA!!!!

 

[BadThad]

Congrats RA! AWESOME! 😀

Thanks BMC! 🙂

 

[oztrailrider]

Congrats to all the milestone makers!, awesome production. Wish I had more machines to add to the effort. Hope plenty of team members get hardware upgrades for christmas! 😉

 

[petrusbroder]

Originally posted by: Insidious
Actually, you can re-benchmark your host anytime. It's under the 'commands' menu in boinc manager
then next time you update the new benchmarks will be recorded. If you don't do anything, it does it's own re-benchmark @ 5 days (IIRC)

-Sid

Quite so. In some versions the benchmark was rerun every 3 - 4 days.
If do something drastic to your hardware (change graphics card, HD, add RAM) or add something memory hungry or computations intensive it is good to rerun the benchmarks.

I just tested it: if e.g. Photoshop runs (with activity) then the numbers drop quite a lot --- even mouse activity changes the Drystones with 40 - 60. If for example F@H runs, then the numbers actually drop by 80 - 90% (which is expected 😉 )

Also: if you have a laptop and stuff connected to the USB-port then the numbers drop by some 8 - 15% ... (that what activity on the USB-port needs). The same is valid for stuff connected to the PC-Card-port. For maximum effect disconnect everything from USB, Firewire, PC-Card-port (unless needed for operation of computer), shut down processes not needed and the benchmark (this is how it should be done anyhow) 😀