| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: Can I ask a question? What the heck is going on with the points award system? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 59
|
|
| Author |
|
|
cjslman
Master Cruncher Mexico Joined: Nov 23, 2004 Post Count: 2082 Status: Offline Project Badges: 
|
I really don't give a rat's buttocks about the points (I've stated this in various other posts because to me, they're useless). I know that there are crunchers that go into a purple hissy about the points (for whatever reason they like). I assume that the Ripplers think that the points are important (and I can understand why). Now, with that said, I do agree that logic would indicate that the points assignment should be stable. So the question is... To point or not to point
---------------------------------------- .CJSL Crunching for a better future... |
||
|
|
twilyth
Master Cruncher US Joined: Mar 30, 2007 Post Count: 2130 Status: Offline Project Badges: 
|
I know how you feel. I'm an admitted points wh*** but even with the added incentive of earning some beer money, I don't pay much if any attention to the daily variations. Even so, I would hate to see us lose anyone because of an issue - even it happens to be folks who are only here for Ripple. I'm sure many here wish Ripple had never heard of us, but I think distributed computing can accomplish great things and want to see WCG be the platform that shows the world what it can do. And to do that, I think we need to grow at a better clip than we've seen over the past decade.
----------------------------------------  [Edit 1 times, last edit by twilyth at Mar 27, 2014 1:55:42 AM] |
||
|
|
Movieman
Veteran Cruncher Joined: Sep 9, 2006 Post Count: 1042 Status: Offline |
I know how you feel. I'm an admitted points wh*** but even with the added incentive of earning some beer money, I don't pay much if any attention to the daily variations. Even so, I would hate to see us lose anyone because of an issue - even it happens to be folks who are only here for Ripple. I'm sure many here wish Ripple had never heard of us, but I think distributed computing can accomplish great things and want to see WCG be the platform that shows the world what it can do. And to do that, I think we need to grow at a better clip than we've seen over the past decade. To me the points are something we have fun with and use to motivate people. Personally I couldn't give a tinkers damn about them except for that BUT they have to be consistant and equal across all projects so that everything is "fair".. As to Ripple it is a two edged sword, It has it's negatives in that it has taken 8 years of work by the teams and tossed it out the window BUT on the other hand it has brought huge amouts of computational power to WCG and in the final analysis the extra work down wins that discussion IMO. I guess what I'm trying to say to the people in charge of WCG is "Take a look at what is going on in the point awards and see what you can do to fix this mess" PLEASE!  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Not sure what you mean by -
It has it's negatives in that it has taken 8 years of work by the teams and tossed it out the window  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
moviemen will well know the scene where the big indian takes a pillow and helps his oblivious friend, followed by the final great escapw from the institution. Can the indian please do same to the points system.
|
||
|
|
Eric_Kaiser
Veteran Cruncher Germany (Hessen) Joined: May 7, 2013 Post Count: 1047 Status: Offline Project Badges: 
|
I'm not part of the Ripple community so I don't care much about points. Sure they are part of my boinc activity but focussing on WCG only, runtime for the badges and signature globe is more interesting.
----------------------------------------But there is one aspect where points are interesting: To compare different crunching devices. For example: My main computer i7-3930@3.5 GHz delivers ~180 Points/hour my odroid u3 running from microSD-Card@1.7 GHz delivers ~60 Points/hour. The same odroid u3 running from eMMC@1.7 GHz delivers ~ 80 Points/hour. I was really surprised that running the same device with Android installed on different media has such an impact on performance. I monitor both devices. They run 24/7 with 99% cpu usage. Without points I wouldn't be able to figure this out. Runtime per wu is no alternative because runtime varies in a broad range.  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Points to time ratios is the only valid base of assessing crunching performance for a node and between nodes. Since the system is corrupted, and not only corrupted in the sense of being inconsistent, but also inflated at many a project, there's really only valid comparing nodes at the same project, with one science, which I'm doing elsewhere. Stable runtimes over long periods, stable per second awarding.
Did stumble on a post by a technician of about 2 years ago to say the server 700 credit system had made things worse combined with wcg custom exception rules. That is 2 years ago. |
||
|
|
Eric_Kaiser
Veteran Cruncher Germany (Hessen) Joined: May 7, 2013 Post Count: 1047 Status: Offline Project Badges:
|
there's really only valid comparing nodes at the same project, with one science, which I'm doing elsewhere. Stable runtimes over long periods, stable per second awarding. I agree. So the result for my main computer is "handle with care" somehow as it was used for different projects at wcg. The two odroids are more comparable as they only run Fight Aids for wcg. Ok, the device running Android from eMMC has a much shorter runtime and less results/points than the one with the microSD but I suppose that the points/hour difference might be realistic between both. Another parameter from my hardware monitoring is the temperature. Both devices have the same "case" and same cooler fan mounted the same way. The one with the eMMC reaches 5°C higher temperatures than the other one. |
||
|
|
Sgt.Joe
Ace Cruncher USA Joined: Jul 4, 2006 Post Count: 7846 Status: Offline Project Badges:
|
there's really only valid comparing nodes at the same project, with one science, which I'm doing elsewhere. Stable runtimes over long periods, stable per second awarding. I agree. So the result for my main computer is "handle with care" somehow as it was used for different projects at wcg. The two odroids are more comparable as they only run Fight Aids for wcg. Ok, the device running Android from eMMC has a much shorter runtime and less results/points than the one with the microSD but I suppose that the points/hour difference might be realistic between both. Another parameter from my hardware monitoring is the temperature. Both devices have the same "case" and same cooler fan mounted the same way. The one with the eMMC reaches 5°C higher temperatures than the other one. Just a thought here, but since the U3's are supposed to be identical, why not swap the eMMC and microSD cards and then see if the difference in PPH follows the device or follows the card. Cheers
Sgt. Joe
*Minnesota Crunchers* |
||
|
|
Movieman
Veteran Cruncher Joined: Sep 9, 2006 Post Count: 1042 Status: Offline |
Not sure what you mean by - It has it's negatives in that it has taken 8 years of work by the teams and tossed it out the window I should have written that differently as I didn't say what I felt. What I should have written is : It has it's negatives in that it has taken 8 years of work by the teams and made it irrelevant in the overall standings. My point is when you have a situation where because of "pay for play" a team can start last November and post more points than strong teams have done since 2006 it totally skews the numbers. It is in it's own way having the same effect as GPU work did to the standings.. |
||
|
|
|