| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Teams
Thread:  Join "MyOnlineTeam" Today - Chapter 40 |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 505
|
|
| Author |
|
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges: 
|
03: NiceMedTexMD - 47,569 points Yep, the points from the new machine have kicked in! At least double before I think.  |
|||
|
||||
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges:
|
WOW! Talk about timing! Dean Mullins rejoined our active ranks today, hit a new milestone and started out at 9th in our Top Twenty list and picked the day we hit 1000 years to do it on no less!
---------------------------------------- |
|||
|
||||
|
darth_vader
Veteran Cruncher A galaxy far, far away... Joined: Jul 13, 2005 Post Count: 514 Status: Offline Project Badges: 
|
I quit HPFII because I was getting lots of errors and not all of them errored in a few min. I had hours and hours in some of them. That's exactly why I stopped HPF2 a long time ago. I didn't mind the occasional WU that errored off after 2 minutes. It's those WUs that got stuck that were the real problem. I had one get stuck over a 3 day weekend and lost 70+ hours of CPU time. Other than the famous CEP debacle and the AC@H WUs that would not restart properly, HPF2 is the only science that I've had any trouble with. I won't ever touch it again unless there are no other WUs available --- and maybe not even then. - D |
|||
|
||||
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges:
|
While it's only been 11 days so far, we're already averaging almost 380,000 points a day. That's over 138 million a year - a nice increase so far! Oh, and we also already have our first millionaire of the year as of today.
---------------------------------------- |
|||
|
||||
|
RT
Master Cruncher USA - Texas - DFW Joined: Dec 22, 2004 Post Count: 2636 Status: Offline Project Badges: 
|
I quit HPFII because I was getting lots of errors and not all of them errored in a few min. I had hours and hours in some of them. That's exactly why I stopped HPF2 a long time ago. I didn't mind the occasional WU that errored off after 2 minutes. It's those WUs that got stuck that were the real problem. I had one get stuck over a 3 day weekend and lost 70+ hours of CPU time. Other than the famous CEP debacle and the AC@H WUs that would not restart properly, HPF2 is the only science that I've had any trouble with. I won't ever touch it again unless there are no other WUs available --- and maybe not even then. - D I am disapointed with them as well. They have had this problem for over a year and steadfastly refuse to think it could be a "science" problem as well as a computer science problem. I really hope the science is not flawed as well thereby rendering the whole project a total failure. But, they will not listen so I quit telling them... Doc. What kind of machine that you end up with depends a whole lot on your bias to begin with. If you think "I want a machine that will do everything I want it to do AND do some crunching as well" then you tend to go to the bleeding edge. At that edge, you tend to get less compting for each dollar but you get a very powerful machine and bragging rights. If on the other hand, you tend to look at the price/performance curve, you tend to back off the bleeding edge to a point (generally called the "sweet spot" on the price/performace curve) and buy your machine there. I tend very strongly towards the latter position. Would you buy one of those Bugatti super cars for over a million or would your buy a corvette that is 85-90% as fast, handles almost as well etc. for $100,000? This is not entirely analogous, but I have multiple "corvette" machines running for less than one "Bugatti" machine would have cost. To be fair to the analogy, I also burn more gas, but not that much more. So.. like I said, where you end up depends for the most part upon where you started from. Your bias and objectives (and sometimes budget) determine your choice. Everyone have a great day! |
|||
|
||||
|
NiceMedTexMD
Veteran Cruncher United States Joined: Aug 17, 2006 Post Count: 929 Status: Offline Project Badges:
|
Good point.. I honestly wouldn't want one of those Bugatti machines, i'm very comfortable with a corvette.. lol
----------------------------------------I'm not sure where the "sweet spot" is as far as price/performance curves.. one of my biggest problems is determining that amongst all of the "accessories" offered on most machines. Dr. Mike..  |
|||
|
||||
|
RT
Master Cruncher USA - Texas - DFW Joined: Dec 22, 2004 Post Count: 2636 Status: Offline Project Badges:
|
The sweet spot is now, (In my opinion) around the I7860 (or 920). with nothing exotic (around a grand). I still think the defaults on most of the configurations are the way to go despite the high-performance people here (and everywhere -- insidious lot
---------------------------------------- ). They tend to go for high-end components with, I think, the idea that they will upgrade the machine over time. I don't, as a general rule, upgrade. When machines get troublesome, I donate them to schools where the students can tinker with them and when the wife isn't looking; sneak a new one into the farm.I was involved with high-performance computing for a few years (mainly on the periphery). An extremely wise friend of mine (at Cray) defined a "Super Computer" since everyone wanted a definition. He said that it was a computer that cost $17 million. Everyone was taken aback because they wanted a teraflop definition. He commented that it changed so fast that the definitions were meaningless but the top computers in the world at that time and for some time before and after that were priced somewhere in that neighborhood. As cavalier as it sounded, it made sense. Scientific computing can be thought of insatiable. Whatever amount of compute power is available will be used and the scientists will want more. So, I suppose one could define a sweet spot computer for WCG is one that contains an I7 and costs $900-$1200 (without display or printer). I think you get more done on the science projects for each dollar spent by buying in this range at the moment. Of course this will change fairly quickly ... but the price point will likely reamain fairly stable over the next year or two. |
|||
|
||||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
The sweet spot is now, (In my opinion) around the I7860 (or 920). with nothing exotic (around a grand). I still think the defaults on most of the configurations are the way to go despite the high-performance people here (and everywhere -- insidious lot ). They tend to go for high-end components with, I think, the idea that they will upgrade the machine over time. I don't, as a general rule, upgrade. When machines get troublesome, I donate them to schools where the students can tinker with them and when the wife isn't looking; sneak a new one into the farm.I was involved with high-performance computing for a few years (mainly on the periphery). An extremely wise friend of mine (at Cray) defined a "Super Computer" since everyone wanted a definition. He said that it was a computer that cost $17 million. Everyone was taken aback because they wanted a teraflop definition. He commented that it changed so fast that the definitions were meaningless but the top computers in the world at that time and for some time before and after that were priced somewhere in that neighborhood. As cavalier as it sounded, it made sense. Scientific computing can be thought of insatiable. Whatever amount of compute power is available will be used and the scientists will want more. So, I suppose one could define a sweet spot computer for WCG is one that contains an I7 and costs $900-$1200 (without display or printer). I think you get more done on the science projects for each dollar spent by buying in this range at the moment. Of course this will change fairly quickly ... but the price point will likely reamain fairly stable over the next year or two.Exactly! |
|||
|
||||
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges:
|
MyOnlineTeam Daily Statistics for 01/12 - All Members:
----------------------------------------Team rank movement report =========================
Points milestones report ======================== madambaster reached 250,000 points  Runtime milestones report ========================= No runtime milestones found.  Results returned milestones report ================================== No results returned milestones found. New members report ================== No new members found. Retired members report ====================== No new retired members found. For the week as a team: Statistics Total Run Time Points Results Team Records: Results Returned: 12/19/2007 2,522 Points: 05/06/2009 518,871 Runtime: 01/25/2006 1:123:00:53:34 Good crunching folks!!!! |
|||
|
||||
|
keithhenry
Ace Cruncher Senile old farts of the world ....uh.....uh..... nevermind Joined: Nov 18, 2004 Post Count: 18665 Status: Offline Project Badges:
|
MyOnlineTeam Daily Statistics for 01/12 - Active Members
----------------------------------------Active team members report ==========================
Note: Active members are those who earned points in the prior 30 days. Top Twenty active members returning points today: 01: RT - 88,155 points 02: Coingames - 59,399 points 03: NiceMedTexMD - 41,028 points 04: marysduby - 39,951 points 05: Jonathan Figdor - 35,869 points 06: brown chris - 25,432 points 07: keithhenry - 14,021 points 08: parmesian - 11,966 points 09: Vuj - 9,739 points 10: Blueprint - 8,919 points 11: sulcata - 8,056 points 12: xroule - 7,522 points 13: Esteban69 - 6,302 points 14: Daeloan - 6,082 points 15: harry_i_c - 5,457 points 16: darth_vader - 5,181 points 17: Jockin - 3,642 points 18: Dean Mullins - 3,338 points 19: Airwolf_Liu - 2,442 points 20: frans6nl - 2,238 points Total points returned today: 404,855 Active members returning points today: 37 Average points per member active today: 10,942.027 |
|||
|
||||
|
|