| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Community
Forum: Chat Room
Thread: 4th gen BOINC credit system discussion |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 12
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hey,
A discussion regarding a potential 4th generation BOINC credit system has begun over at the official BOINC message boards: https://boinc.berkeley.edu/dev/forum_thread.p...p;sort_style=&start=0 Cross posting here to draw attention to the thread (doing so on all BOINC project message boards). I'd appreciate some input on the matter. Thanks :) |
||
|
|
deltavee
Ace Cruncher Texas Hill Country Joined: Nov 17, 2004 Post Count: 4890 Status: Offline Project Badges: 
|
customminer,
Thanks for the link. It was very interesting reading. The effort to revise the credit system seems to be driven by the projects Gridcoin and Bitcoin Utopia, where the exchange of virtual currency is occurring, and where we are led to believe that points cheating is common. This is another reason to give kudos to the WCG staff for using runtime and not points for badges. |
||
|
|
cjslman
Master Cruncher Mexico Joined: Nov 23, 2004 Post Count: 2082 Status: Offline Project Badges: 
|
customminer, Thanks for the link. It was very interesting reading. The effort to revise the credit system seems to be driven by the projects Gridcoin and Bitcoin Utopia, where the exchange of virtual currency is occurring, and where we are led to believe that points cheating is common. This is another reason to give kudos to the WCG staff for using runtime and not points for badges. +1  CJSL Crunching for a brighter future... |
||
|
|
OldChap
Veteran Cruncher UK Joined: Jun 5, 2009 Post Count: 978 Status: Offline Project Badges: 
|
Happy to use runtime here but on the points argument I think each Wu or perhaps random Wu's could contain a benchmark style piece of work from which a core score or time is derived. That piece of work should be just a regular project calculation and as such be hidden. then the project, keeping an index, gives points per runtime.
----------------------------------------What I describe has probably been done before in some shape or form but I must admit that seeing the same machine claim differing points for the same runtime on a given project somewhat irksome.  |
||
|
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
Not the first to promote a mini benchmark -per task-, tamper proof of course, the project then setting the value of the unit of time for the device [where it's claimed that time is even easier manipulated]. The per-task benchmark certainly would cater for devices that have alternating CPU multipliers [and funny, Linux is working on greatly improved scheduling to handle per-core speed managing]. That will certainly kill the general BOINC benchmark which runs on a single core and assumes it applies to the whole device. Without funding [NSF actually withdrew rather than not giving a new grant... BOINC has to stand on it's own legs after 12 years], don't see much is going to happen for a long time to fix the credit manure.
|
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
What I describe has probably been done before in some shape or form but I must admit that seeing the same machine claim differing points for the same runtime on a given project somewhat irksome. Getting 1 BOINC point for every 7 WCG points also falls under the irksome column. I can't tell you how many people have told me they don't crunch much or at all at WCG because of that.
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
  |
||
|
|
cjslman
Master Cruncher Mexico Joined: Nov 23, 2004 Post Count: 2082 Status: Offline Project Badges:
|
...many people have told me they don't crunch much or at all at WCG because of that. yeap, that's irksome  CJSL Crunching like there's no tomorrow !!! |
||
|
|
SekeRob
Master Cruncher Joined: Jan 7, 2013 Post Count: 2741 Status: Offline |
What I describe has probably been done before in some shape or form but I must admit that seeing the same machine claim differing points for the same runtime on a given project somewhat irksome. Getting 1 BOINC point for every 7 WCG points also falls under the irksome column. I can't tell you how many people have told me they don't crunch much or at all at WCG because of that. Those are maybe math challenged...we are getting 7 WCG points for every BOINC credit :) |
||
|
|
KLiK
Master Cruncher Croatia Joined: Nov 13, 2006 Post Count: 3108 Status: Offline Project Badges: 
|
customminer, Thanks for the link. It was very interesting reading. The effort to revise the credit system seems to be driven by the projects Gridcoin and Bitcoin Utopia, where the exchange of virtual currency is occurring, and where we are led to believe that points cheating is common. This is another reason to give kudos to the WCG staff for using runtime and not points for badges. That's like writing: all politician are thieves & crooked! I, for one, use Gridcoin system...but haven't made any cheating technics, just finding a way to get a part of the costs back!  SekeRob said it right...we give "free computational power" to people scientists...well, that's nice, so we give our "donations of computational power"...but it costs money! Computers, parts & newer parts, expanding memories, HDD & SSD, ... - & the list goes on...that's what we give + cost of electricity! Unfortunately, we haven't given a chance to get a "donation certificate" with a 20-25W per core donation * working hours...I'd rather see an option that we could bring a yearly printout from local IBM for a donation...so we can get a "local tax donation claims"... Points, don't have so much interest in them...at least here on WCG! why? Here on WCG it's all about BADGES...so even the topic about "Points per watt", doesn't interest me...but "Points per core" - that does interest me! We all here like our badges...even more than Montenegro partisans! (local joke)  |
||
|
|
Col323
Senior Cruncher Joined: Nov 4, 2008 Post Count: 372 Status: Offline Project Badges: 
|
Points, don't have so much interest in them...at least here on WCG! why? Here on WCG it's all about BADGES... No, no it's not. Maybe I'm the only one who thinks this way (I highly doubt it), but WCG offers me the chance to punch cancer, et. al. in the face. There are dozens of Boinc projects which offer badges. If I was all about the BADGES, I would be running as many of those as I could to rack up as many as I could. Is it cool that WCG offers badges? Yes, it's nice to have that carrot to chase. Is it considerate that they tie them to run time vs points or results? Sure, because it throws a bone to everybody. Points and Results are purely performance related. Run time is less so. Bottom line, there are numerous reasons why people do WCG. Those reasons may even shift over time. For me, badges barely move the motivation needle for me, and probably never will. That doesn't make my motivations better than yours, but I'm not going to sit here and pretend that badges are the sole motivation for WCG. Back on topic, I've spent far more CPU cycles on WCG than some of my friends with fewer machines. However, some of them simply chase points and won't do much WCG because of the points penalty. To each their own, but it would be nice to have a more even playing field so they would be more enticed to crunch here. However, I'm not holding my breath. Not because of a lack on any person's part, but because of all the variables to harmonize. And I'm still skeptical that a project admin might try to tweak their points allocation to draw a few extra crunchers their way. |
||
|
|
|