Free-DC DF Team does it again !

[Chinasaur]

Way to go team for having THREE FReeDC peeps in the top 10 RMSD :)

So why does FreeDC consistently have so many in the Top Ten RMS than anyone else?

Who cares "as long as the kids go?"

Keep on rockin!




:jester:


:cheers:

[Beyond]

Originally posted by Chinasaur
Way to go team for having THREE FReeDC peeps in the top 10 RMSD :)

So why does FreeDC consistently have so many in the Top Ten RMS than anyone else?

Who cares "as long as the kids go?"

Keep on rockin!




:jester:


:cheers:

2 people (3 entries) in the "beta" top 10 RMSD as well :)

[Angus]

Originally posted by Chinasaur
[B]Way to go team for having THREE FReeDC peeps in the top 10 RMSD :)

Up to 4 now :)


W00T! I got #2

[Paratima]

Cool! :smoking: Pure chance, of course. But still cool.

[Paratima]

UN-believeable!

Now it's up to 5!

Of course, ZaphodB just has to be ahead of me! :jester:

[Chinasaur]

S O B !!!!!!

Five is freaking hellacious...more than has ever been!!!

I LOVE THIS TEAM :D~~~~

[Chinasaur]

S O B !!!!!!

Five is freaking hellacious...more than has ever been!!!

I LOVE THIS TEAM :D~~~~

[IronBits]

Did someone just stutter? ;)
:cheers: Go Team Free-DC :cheers:

[Paratima]

The man just has NO control over his mouse-button finger! :D

[Beyond]

Note of interest, we hold the top position in best RMSD in both the "official" and "beta" stats atm. :)


Best structure so far IronBits FreeDC FreeDC 11.63
Runners up: - - - - -
#2. Ponky FreeDC FreeDC 12.10
#3. kk kkhelpz Canada 12.13
#4. Paratima FreeDC FreeDC 12.42
#5. Angus Old Farts Crunching for DF FreeDC 12.54
#6. ZaphodB minimal FreeDC 12.56
#7. HaloJones Anybody could have done it The Genome Collective 12.59
#8. iceman Ars Technica Team Stir Fry Ars Technica Team Stir Fry 12.64
#9. traverse Michigan State University Ars Technica Team Stir Fry 12.72
#10. DfA Home PTC 12.74



1. Beyond FreeDC FreeDC 5.74
2. Hzull Discombobulated Ars Technica Team Stir Fry 5.79
3. Brian the Roman Ottawa Canada 5.88
4. Aegion Arstechnica Ars Technica Team Stir Fry 5.88
5. IronBits FreeDC FreeDC 5.92
6. Georgina KWSN Chica The Knights Who Say Ni! 6.11
7. Beyond FreeDC FreeDC 6.32
8. Guff Distributed Computing Mercenary Team Endeavor 6.37
9. frozenchosen Minot State University Belgium 6.41
10. AMD is logical www amdforums com AMDMB Killer Barbarian Frogs 6.42


:|party|: :cheers:

[IronBits]

Good job everyone! :cheers:
/me tosses another log on the BETA client and waits to see if Beyond responds likewise. :|punch|: :D

[Beyond]

Originally posted by IronBits
Good job everyone! :cheers:
/me tosses another log on the BETA client and waits to see if Beyond responds likewise. :|punch|: :D

:scared:



#1. IronBits FreeDC 56,261 56,261 5.735129
#2. Beyond FreeDC 55,692 55,692 5.487837


Pretty darn close now! :cool: Let's see what I can do..

:cheers:

[IronBits]

:smoking: :notworthy :cheers:

[Beyond]

On the move http://24.174.34.241:8080/images/fire030.gif , lets see if we can take the #1 spot in the beta testing. :thumbs:

[IronBits]

:rotfl: Love the bird! :D
I see we think alike :thumbs: :)

[IronBits]

:scared: Beyond... what part of "A LOG on the fire" did you misunderstnad?! ;)
:bang: /me quickly runs around looking for more kindling :p ;)
:notworthy

[Beyond]

Originally posted by IronBits
:scared: Beyond... what part of "A LOG on the fire" did you misunderstnad?! ;)
:bang: /me quickly runs around looking for more kindling :p ;)
:notworthy

/thought you meant logs like "we" use for bonfires :D http://24.174.34.241:8080/images/FIREdevil.gif

[Beyond]

Latest stats;



Rank Team # of points
#1. Bioinformatics Consulting Group 385,359
#2. FreeDC 172,899
#3. Team Endeavor 172,164
#4. Boise State University 160,444
#5. Ars Technica Team Stir Fry 153,077
#6. Canada 117,662
#7. AMDMB Killer Barbarian Frogs 77,290
#8. Asylum Research 49,751
#9. Dutch Power Cows 47,919
#10. The Knights Who Say Ni! 39,997





We have moved up from #5 this afternoon to #2. :|party|: :cheers: :|party|:

[bwkaz]

:cheers:

The real question, though, is whether we've screwed up the normal stats while throwing these logs on the beta fire. And if not, why weren't these logs burning on the normal client beforehand? :p

(not that I should be talking -- I've got a measly two systems here, one running the normal client and one running the beta...)

[IronBits]

I've been pulling them off DF to help Howard test his Beta client, soon to be THE client ;)
I was the only one holding the color RED , and almost doubled what Dyyryath/Condor was/is currently doing. :sleepy: ;)
I want to pass Paratima (tonight), then run as much as I can on the BETA... :D
I was just trying to perk up 'some' interest and get some extra help for Howard, but that overly enthusiastic maniac, Beyond, has done went and stirred up a hornets nest. :scared: :jester: :crazy:
I have some LOGS over here that even Paul Bunyan would be proud of, so I guess I'll toss a few of them on that bonefire Beyond has already got going and see if we can steal #1 spot before someone notices :|party|: :cheers:
It's all for the better of, or for, the science either way. :thumbs:
/me starts looking for the marshmallows :D

[IronBits]

Originally posted by Beyond
We have moved up from #5 this afternoon to #2. :|party|: :cheers: :|party|:
Careful now!!! Don't spill any :cheers: on the fire! :p
We be kickin it hard now :jester: :|party|: :notworthy

[Beyond]

My thoughts on why the "logs" were not on the "normal" client is the team is in a position were we are not able to catch TSF for #1, or threatened by anyone below us that may overtake our #2 position. Also with the start of the hugh 157aa that is so slow I decide to concentrate on the beta testing.

It is my understanding that the stats will be reset when the new client is released for general use, so the more resources used to test it and chase down any bugs the sooner it will be released and we can get started on the new race to #1. :D

As the #1 team is not producing (hopefully not no-netting) we should hit No.1 by this time tomorrow. However I am about at my limit (# of logs) if I want to continue to make headway in OGR.

[IronBits]

Originally posted by Beyond
As the #1 team is not producing (hopefully not no-netting) we should hit No.1 by this time tomorrow. However I am about at my limit (# of logs) if I want to continue to make headway in OGR.
If you think I'm gonna go near that BIG fire and singe what little hair I have left to pull them big Paul Bunyan logs off the fire now that you have :cry:'d uncle, you can just forget it :|punch|: :jester:
:bs: ;) (I only tossed on a couple more giggle hurts is all) :notworthy

[Beyond]

Originally posted by IronBits
If you think I'm gonna go near that BIG fire and singe what little hair I have left to pull them big Paul Bunyan logs off the fire now that you have :cry:'d uncle, you can just forget it :|punch|: :jester:
:bs: ;) (I only tossed on a couple more giggle hurts is all) :notworthy

Bah, I'm not worried about afew little old GHz. :moon: :D

[IronBits]

:rotfl:

[Paratima]

Originally posted by IronBits
I want to pass Paratima (tonight), then run as much as I can on the BETA... :D Not tonight. :mad: I've got a couple of folds here to slow you down just a bit.

Tomorrow, OK. :p

[IronBits]

Not sure about tomorrow either now... but it will happen :p ;)
I'm busy melting marshmallows with Beyond :D
Good work tho, you held me off for a very long time :thumbs:

[IronBits]

Beyond, did I say just a couple? :confused: :o
I meant a couple of "Paul Bunyan would be proud" sized logs, with lots-o-pine pitch. :spray: :D
:moon: wiggle wiggle :moon: I don't have a target on my derriere for nothing ya know. :rolleyes: :D
:cheers:
#1. IronBits FreeDC 92,447 92,447 5.243147
#2. Beyond FreeDC 85,166 85,166 5.487837
:cheers:
Can't wait for this to go live! :smoking:

[Beyond]

Just a scratch...:haddock: ....:bang: :D

[Beyond]

Arrrgh..came home from work and noticed the power had gone out sometime today...6-7 hours crunching lost. :swear:

[IronBits]

:( :mad: :swear: :bang:

[Beyond]

Back on schedule now, also as predicted;



Teams who have generated the largest amount of work

#1. FreeDC 545,720
#2. Bioinformatics Consulting Group 451,199
#3. Team Endeavor 279,800
#4. Ars Technica Team Stir Fry 274,740
#5. Canada 202,602
#6. Boise State University 160,444
#7. AMDMB Killer Barbarian Frogs 110,020
#8. Asylum Research 97,031
#9. The Genome Collective 85,371
#10. Dutch Power Cows 79,828


:cheers:

[IronBits]

Hopefully I'll catch him tomorrow ;)
#1. IronBits FreeDC 330,281 330,281 5.243147
#2. Beyond FreeDC 142,352 142,352 5.487837
Sorry that 'scratch' got so deep :jester:
:notworthy
We have 4 in the top 10 RMS :D :thumbs: :cheers:

[Beyond]

Originally posted by IronBits

Sorry the 'scratch' got so deep :jester:
:notworthy

What...that..just a flesh wound, nothing more. :rolleyes: :D

Heck my grandmother can crunch faster than that....:scared: :jester:

[magnav0x]

Hopefully a bit after the new DF client goes live, Hyrda Central can give some sort of competition for you guys, just so you don't get bored :D The machines that are together were moved to OGR at least until the new client is ready.

[Angus]

Who's bored? I'm after that target up ahead....


:crazy:

[Welnic]

Nice job of smackin' down those RMS numbers in the beta, guys. :)

[IronBits]

Beyond!!! Get yer butt back in the DFbeta :bang:
Ya went and laid down and let Angus run over you!
http://www.free-dc.org/forum/showthread.php?s=&threadid=2913
Good job Angus! :D :thumbs:
He's still nursing that 'scratch' earlier I think :bs: :crazy:

[IronBits]

Dayum!!!
We have 6 in the top 7 of the top 10 best RMS! :smoking:
:cheers:

[IronBits]

Angus :bang:
Your pushin it buddy :p ;)
:thumbs:

[Paratima]

S'this here beta stuff a private party or can just anyone play? :p

[IronBits]

Come on in! Party on dude!
:|party|: :cheers: :jester: :smoking: :crazy: :help:

[IronBits]

And you can get your DFbeta STATS needs fed right here ;)
http://www.dbestern.net/stats/dfbeta.html

[Paratima]

Hey, looky there! Best of 7.4 after running only a couple hours.

Beginner's luck. :D :D :D

And what a LOT of interesting (if incomprehensible) numbers to look at! :eek:

We'll see how it goes with a coupla boxen for a while...

Nice stats, BTW. Thanks, IB.

[IronBits]

It was DP (Glen) who threw them together for us :thumbs:

For awhile = approx. 6 days + hours per protein.
You get your very own to work with, start to finish :)

[Paratima]

Well then, THANKS, DP!!

I gets confused :confused: if I don't get my stats.

[pointwood]

/me thinks IronBits (in the pink skirt) have fooled Howard into providing him with the best structures :swear: :jester: ;)

[Paratima]

Gosh! Do ya think so? :eek:

[Angus]

Now 7 ! of the top ten RMSDs belong to Free-DC. :|party|:


Paratima is not holding up his end, tho - C'mon dood, let's get on the chart!!!

[Paratima]

All right, all right! Hold yer horses.

I'll toss in another box & see if that helps. :p

[bwkaz]

w00t! I'm in the top ten RMSD!

That has got to be a first...

:|party|:

[Beyond]

Sorry IB and others, I am not sure about this client. All the "best" structures I've found have been within the first 30 or so generations, having some run out to as much as 175th generations without finding anything better...so not sure if it is doing any "good/real" work running all the way to the 250th generation.

However if the race gets to the point were we are challenged for 1st place I'll jump back in! In the mean time top 100 in OGR-24/OGR-25 is calling my name. ;)

[Paratima]

Hmmmmm.

[bwkaz]

Beyond -- that's why it's a beta. ;)

I believe that if Howard finds that most machines don't improve after a certain generation, then he'll move the generation cutoff earlier. However, I seem to have gotten my best in generation 222, so for whatever that's worth...

Remember, this is still a random sampling algorithm, at its heart. It's directed a bit by the information it gets in the first gen (the first 5000 structs), in that it just tweaks the best struct from those to try to make it fit better, but it's still random. Just because you aren't seeing better RMSD values doesn't really mean the client isn't doing anything worthwhile.

Of course, if you don't want to run the beta, I won't stop you, but just so you're aware.

[Paratima]

Originally posted by Paratima
I'll toss in another box & see if that helps. :p I will, that is, if Florida So-Called Power manages to keep the juice running more or less regular! :swear: :bang:

[Paratima]

bwkaz: The lesson here is that you have fairly high-class, intelligent people, mostly, contributing to these DC things. (You and me, for example!)

If you give them too little information about the project, these curious folk (complimentary sense) will wander off and find something to do that is more interesting and about which they can get answers. And to which they can contribute solutions. And who can blame them? I don't think that has happened on THIS project! But that's one of the things that drove me and several others here from FAH.

If you give them information (is there such a thing as too much? :confused: ), then they start making decisions for themselves about how to, whether to, and in what manner to, run it. This has been discussed in some depth on several threads over on the main DF board and will no doubt continue similarly. It's generally healthy.

Howard and friends have opted for the more-information route, of which most of us obviously approve, but with knowledge comes greater degrees of freedom. I won't be at all surprised to find many people ending runs before Gen250 (or whatever the limit is) because they have decided that they know better.

What a neat problem! :p /end ramble/

[Beyond]

Thanks bwkaz and Paratima for that. It is true Howard and company give more info than most and being semi-intellegent person (in my own mind) I question if the continued running of the generations after a certian period of time without improved structures is the best use of resources?

Now if it proves out that the majority of the sampling from the beta test shows that continued running is not bearing fruitful, he (Howard) may very well change the client to stop after a certain period time or generations without improved structures.

It was not my intention to bad mouth the client or project, only to question the implimentation from my limited observations to date. If the end results of the test prove me wrong, and the majority see an improvement through out the full 250 generations then I will stand corrected and be glad to return to running the client. :)

[Paratima]

No offence meant or taken! :D

I should have included you specifically in my ramble of "fairly high-class, intelligent people". I was actually sort of off-hand praising your decision process. If we all keep thinking and talking about it, it can't help but get better, IMHO.

I believe that I just read that Jeff, aka Digital Parasite, intends to modify dfGUI to show not just the best energy level so far, but also the generation in which it was achieved. That could help us make better decisions, both individually and collectively.

[Beyond]

Digital_Parasite is always looking to improve upon perfection..I hope he is successful, would be a big help. By the way thanks for your input. :cool: