Printable View
First thought:Quote:
Originally Posted by Maxwell
Well..... Since this group is DrPop's creation he should declare Maxwell 1st Marshall and convene a War Counsel. Then announce our marching orders for the attack.[/end]
Reality:
It would be nice to have consensus in regards to where we can do the most good for the team. Maybe have a list of projects that are in jeopardy with estimates on how long before the team loses its respective positions.
(One last comment on this, and then I'm done.)Quote:
Originally Posted by STMahlberg
No, the credit system (total cross-project credits) is NOT the only gauge we have now. There are better methods to compare us to our competition. Team MegaMilestones and FormulaBOINC are two. They actually compare apples to apples, with no skewing.
Sorry I've been out of it guys. Like I mentioned to F$ in a PM, R.L. has been handing me my you know what lately.
I'm just going to be blunt. This team is frickin' huge. It's not as big as LAF or S.G., but it's still big. Lots of room for folks to chase whatever their goal in distributed computing happens to be.
That said and understood, I do not believe this is a group for the "personal" MM type chasing. Does that make sense? No hard feeling to anyone intended. Being #1 overall and being #1 in every project are somewhat mutually exclusive at the moment, until certain projects get their head out of the sand and grant more (or less!) credit. Slicker and Crunch3r explained the crappy code most programs are running, so I understand it's NEVER going to be apples to apples. But, it could be a heck of a lot closer than it is now.
When "Joe Blow" newcommer to BOINC looks at a stats page such as Free-DC or BOINC Stats or insert-stat-page-here- and goes hmmmm...what team should I join? Is He/she looking at overall points? Highest RAC? MMs? Etc, etc. We need to find out what they are attracted to, and TARGET THAT. Because we need to grow this team.
Perhaps I was too vague at the beginning. I feel this group is the "BOOST SETI.USA" group. It won't be for everyone. You've got to be willing to back off on a personal project to crunch for a team goal in this group. It's all about helping the team achieve a goal in a specific project, or lift us in the overall stats.
It's for people like me, who will crunch the highest paying project(s) to lift our RAC until what time someone contacts me and says "We're about to lose our #1 in ABC" at which time I will switch over and crunch that until we've either won it back or lost it to the point we give that one up for good.
For strategy like this, I would prefer that we had a TEAM ONLY forum, because it's going to get down to ....do we give this up for that... or, etc. We don't want / need other teams to know that data on us! :p
Am I stepping on anyone's toes yet? ;)
Okay, back on task... I understand what the group is about but for right now while everything else is being decided, I think that you (DrPop) should just pick a project that we should crunch in the meantime.
If it's ABC, then let's point our rigs at ABC until another project is decided.
I got that. I'm not willing to do that, so I didn't join the group. No hard feelings...Quote:
Originally Posted by DrPop
However, I do monitor this - If the Boost group is hitting a project that's on my list of goals, I'll join. I brought up ABC because I happen to be crunching it, and noticed other teams more than doubling our production.
Again with the "members only" forum. I still don't like the idea. As a potential compromise (if possible), perhaps we can have our Admins create a new Groups permission level where you can't see posting within the group unless you are a member of the group?
If we by creating and joining a new board, wanted to make a really fresh start on everything...Then we also need to step out of our alter-egos and re-think "the way we do business"...
If we start to alter the idea about a new home, by implementing all the old habits from the old one....What have we then achieved?
Absolutely nothing!!
So....I like the idea on this group...But if we wanīt to make it happen...REALLY happen...We need to start thinking outside the box...
- Why not give a damn about all our #1 slots (which no way in hell we can defend anyway) and focus on 2-3 projects and then just CRUNCH HARD..
- Why not give a damn about #1 slot overall - and just be the kings of CPU power?
- Why not go donate some power to each other?
I do not mean by this that we should give a damn about everything...Just painting a picture...Tossing out some ideas..
-
Maxwell, I like that idea. (And I figured this really isn't your thing, so no worries at all.) This group is going to be a fairly tight bunch - and we're going to do some Viking raids! The kind that make other teams nervous! :D If Mr. Hankey could pull off a permissions level like that, then sweet.
There are the MM chasers on this team (you guys have a great thread on that - keep it up!), and then there's the "set it on ___project and crunch" members, and then there are those who want to be #1 overall. (Quite a few of those members left and went elsewhere, but we still have a few. I'm one of them.)
This is kind of a bastion for the "would-be #1's". We're going to try to hang onto every position we've got and build some as well. But it's not going to be easy.
NEXT POST for STM: Strategy, Baby! :cool:
Beer: Interesting, yes, we will have to evaluate all that as a team.
STM: Let's try this as a test. Maxwell just put out the request that we warm up our slugging arms for a good inning over at field ABC. I'm going to post the first post in the group message center for the call to switch over to ABC in a couple hours.
Now the hard part. We need to know a way to gauge our success, and when to quit, and each resume our normal projects. (For you and I it would be AQUA because we're #1ers, to each their own...)
I'm going to call the first shot: we (the Bash Bro.s) back off after we increase our lead over LAF by a mil or so. No one else (like S.G.) is even close, so we don't really care how much they're ramping up. But, LAF is close enough they *could* do some damage to our #1 at ABC over the course of a few months. LAF is gaining on us by 68K per day. We currently have a lead of 72 Mil.
The Bash Bro.s should be able to reverse the trend, and GAIN us a Mil. in fairly short order. Let's see if we can do this.
Looks like LAF is still leading us by ~50K a day in ABC. :(
It looks like they have "upped" their production a bit...But if so....They must have backed off somewhere else?Quote:
Originally Posted by STMahlberg
That makes sense. If I'm reading the stats correctly, LAF dropped a dozen core projects that were generating them anywhere from ~10K to ~200K a day compared to the day before.Quote:
Originally Posted by Beerdrinker
Credit for today dropped to zero on the following projects for what it's worth...
Primegrid
Seti@Home
Virtual Prairie
Eon
CAS@Home
Cosmology@Home
Ibercivis
Enigma@Home
Rectilinear Crossing
NFS@Home
Spinhenge@Home
AlmereGrid
STM, I admire your enthusiasm. Man, I am glad you're on my team!
However, you cannot count those projects out for the day until about 9 or 10 PM our time. Otherwise, you might be looking at just delayed reporting of the project tithe stats site.
ABC is interesting. I still need to go look how much credit I have pending, because I'm getting less than 1/10 the credit I was with AQUA. Maybe that's OK, but I'm thinking it's a little low.
I actually think that ABC lowered their credits a while back....Back in 2005/2007 it used to be a project with very decent credit...And I mean very decent...A Dual core could easily do 5-7000 per day back then...Nowadays you could hardly do that with 4 cores...Quote:
Originally Posted by DrPop
Must have been nice; as you can see here, I'm not getting anywhere near that. I have (13) cores working on this, (4) of which are 64bit.Quote:
Originally Posted by Beerdrinker
Well, at least that one makes sense... The project being closed and all...Quote:
Originally Posted by STMahlberg
Yeah I know...It sucks...Quote:
Originally Posted by STMahlberg
Just look at our former #1 at ABC...
http://abcathome.com/show_user.php?userid=5744
Sideband havenīt been around for the last couple of years...Donīt know what happend to him...
He did 10 Mill on that project on pure CPU alone!! That where before GPU became the "big thing"...I am still impressed with that!
I have four i7 860s on ABC, which are collectively doing about 30k/day there on Win7x64 machines. I've built up enough pendings so I'm validating about as much as I'm throwing into pending, giving me a pretty stable count.
That is impressive.Quote:
Originally Posted by Beerdrinker
SG 202,768
LAF 171,979
S.USA 140,010
Looks like SG cranked it up a bit.
Forgive me for stating the obvious... we simply don't have the firepower at the moment.
Migth I suggest something else??Quote:
Originally Posted by STMahlberg
Look at the GAP between #1 and #2 at QMC ATM.....:cool:
Maybe we don't have the firepower but at least I did get this.
Congratulations!:cool:
You've been chosen as the ABC@home user of the day!
Your profile will be featured on the ABC@home website for the next 24 hours.
That and $150 will buy me a GPU.
That's sweet Fogle!
Well, I made the call in the group message forum to switch off ABC and start crunching for the PG Challenge tomorrow.
I was headed back to AQUA after that but... Beer, you ARE correct. Anyone want to hit QMC@Home with me after the 24hr PG Challenge is over?
I think we could take a new first there in a week or two! :D
Any takers?
I could be tempted!! Need only like 10.000 to turn my score into a cool 100.000 Stone!Quote:
Originally Posted by DrPop
I already have my 10K there, but since you came up with the Bash Brothers, I don't care as much about increasing my 10K project list. Count me in.
Congrats Fogle!Quote:
Originally Posted by Fogle
I only have 1.3 mill there but sure, why not. :cool:Quote:
Anyone want to hit QMC@Home with me after the 24hr PG Challenge is over?
Sweet! Let's go knock 'em dead on PG, and then we'll switch over to QMC@Home directly after. The more the merrier. I think we can take it if the credit's not too bad there.
I'm good with that. :-bdQuote:
Originally Posted by DrPop
Alright! PG was good. We got close. REALLY close. Like 2 more Quad core CPUs would have put us over the top. That's a pretty good show.
I've just turned my rigs onto Quantum Monte Carlo "QMC" @Home for anyone who wants to join in. We've got a fairly small gap to close there, as Beer Drinker pointed out, and we're hoping we can take a new #1 slot for the team.
Anyone who wants to help, please jump aboard, especially all the guys here who like to see SETI.USA #1 in as many projects as possible. That's usually not my goal (I'm a #1 overall kind of guy), but on this project, I think we can all unite together!(%)
"Team Only Strategy Chat" about this and any other project ramp ups - hit the SETI.USA Bash Bro.'s Group forum.
Please pass the word...jump on QMC!
I've jumped. :p
Is this thread still alive? Last post is listed as october.. Just wondering.. <: -)
Yep. It's alive. Last post was dated three days ago, actually...;)Quote:
Originally Posted by nmgraywiz
DOH!!! You're quite right.. I just realized I was looking at the Join date of everyone, not the post date.. in the immortal words of Rosanne, Rosanna Danna.. "NEVERMIND!!!"
The Gray Wizard
Quote:
Originally Posted by Maxwell
Ahh, working your way around Wiz. Welcome to the Bash bros.
Hehe... If you haven't already, check out the "groups" area on the boards. Lot of cool features around here...Quote:
Originally Posted by nmgraywiz
Still chugging at QMC moved up to 479th place, considering I started at well over 800 not bad. I'm wondering how long we will have to crunch to get #1 position?
Slightly more detail in my latest post in the projects -> QMC forum, but we're still gaining on them very nicely. Thanks for everyone helping out! Really shows we can be a "team" again!!! :). Go SETI.USA!
***Calling all Bash Bro.s***
Please check out the group mini-forum under "Community" --> "Groups" --> "Bash Brothers" and weigh in on the name change currently under discussion.
Thanks!
All current Bash Bros:
You will get an invitation to join the newly named group later today. It appears I cannot simply change the name of the group; I must make a new one.
I'll have it drawn up in a couple hours when I'm at my office.