Top 100 - Why so low?

Message boards : Number crunching : Top 100 - Why so low?
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile BigWaveSurfer

Send message
Joined: 29 Nov 01
Posts: 175
Credit: 32,728,448
RAC: 14,267
United States
Message 1980280 - Posted: 14 Feb 2019, 13:41:57 UTC

I clicked on the top 100 participants on the BOINC page and was somewhat surprised to see only a few running Seti. Are we, the users here, really that small of a batch? Collatz Conjecture had a significant share of most of the top 100 users allotment. Is there a historical graphic that shows the active users for Seti for the last 5 or 10 years?
ID: 1980280 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 17795
Credit: 407,167,063
RAC: 143,291
United Kingdom
Message 1980281 - Posted: 14 Feb 2019, 13:43:51 UTC

Many projects "pay" far more per task (or per hour) than SETI.
In terms of number of users SETI is still one of the largest projects using BOINC.
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1980281 · Report as offensive
Cruncher-American Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor

Send message
Joined: 25 Mar 02
Posts: 1498
Credit: 338,591,048
RAC: 170,351
United States
Message 1980291 - Posted: 14 Feb 2019, 14:51:08 UTC

I can speak from personal experience a few years ago when I ran Collatz for a while. It gave WAY more credits on a daily basis than SETI does. That completely biases the Top 100 you were looking at.

The problem is that there is no metric for cross-project evaluation of work done. So no way of comparing credits given between projects. Also, there are users who work multiple projects at the same time, and that is a whole other can of worms.
ID: 1980291 · Report as offensive
Nick: ID 666
Volunteer tester

Send message
Joined: 18 May 99
Posts: 13056
Credit: 36,541,778
RAC: 21,016
United Kingdom
Message 1980296 - Posted: 14 Feb 2019, 15:06:19 UTC

The credits awarded on Seti main and Setri Beta don't even match.

If I set my computer to equal time for Main, Beta and Einstein, the RAC's are, approx, 2500, 4000, 100,000.
ID: 1980296 · Report as offensive
Profile BigWaveSurfer

Send message
Joined: 29 Nov 01
Posts: 175
Credit: 32,728,448
RAC: 14,267
United States
Message 1980302 - Posted: 14 Feb 2019, 15:45:27 UTC - in response to Message 1980281.  

Many projects "pay" far more per task (or per hour) than SETI.
In terms of number of users SETI is still one of the largest projects using BOINC.


Interesting, good to know! Thanks!
ID: 1980302 · Report as offensive
Profile Gone with the wind Crowdfunding Project Donor*Special Project $75 donor
Volunteer tester

Send message
Joined: 19 Nov 00
Posts: 41577
Credit: 41,999,167
RAC: 646
Message 1980333 - Posted: 14 Feb 2019, 17:51:44 UTC

The stated aim of Boinc in 2004 was that ALL projects should pay the same credits, and that crunchers should make their choice based purely upon the scientific merit of the individual projects. Unfortunately they couldn't, or didn't enforce that. At least the blatant cheating of Seti Classic was stopped.
ID: 1980333 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 11658
Credit: 174,075,840
RAC: 119,591
Australia
Message 1980446 - Posted: 15 Feb 2019, 7:12:44 UTC - in response to Message 1980291.  

The problem is that there is no metric for cross-project evaluation of work done.

Actually, the problem is that no one uses the metric for cross project evaluation- and that is the Cobblestone.
If everyone (particularly Seti- yes this project) actually awarded Credit for work done according to the Cobblestone definition, then cross project comparisons would be possible, and valid. And a very nice by-product of that would be Seti paying out a whole lot more for each WU processed than is presently the case.
Grant
Darwin NT
ID: 1980446 · Report as offensive
Profile Gone with the wind Crowdfunding Project Donor*Special Project $75 donor
Volunteer tester

Send message
Joined: 19 Nov 00
Posts: 41577
Credit: 41,999,167
RAC: 646
Message 1980454 - Posted: 15 Feb 2019, 7:50:15 UTC

That is a shame because Jeff deserves the prestige of it.

It has seemed to me that Seti has always somewhat undersold itself, even Eric calls it "an along for the ride" project as we don't have our own dish. In terms of credit Seti's view seems to be that they don't want credit hounds or fair weather friends "just here for the beer", they want long term stayers and supporters, But of course that long term loyalty can come at a disadvantage as we found with credit screw. Captive audience even.

But credits for work done are just figures on a bit of paper, you can't spend them like bitcoins. What they are useful for is comparing performance of individual systems for tweaking purposes.
ID: 1980454 · Report as offensive
rob smith Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer moderator
Volunteer tester

Send message
Joined: 7 Mar 03
Posts: 17795
Credit: 407,167,063
RAC: 143,291
United Kingdom
Message 1980456 - Posted: 15 Feb 2019, 7:58:34 UTC

...but only if they remain constant in terms of number given for a fixed parameter and not waggle around like [insert phrase]...
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?
ID: 1980456 · Report as offensive
Profile Gone with the wind Crowdfunding Project Donor*Special Project $75 donor
Volunteer tester

Send message
Joined: 19 Nov 00
Posts: 41577
Credit: 41,999,167
RAC: 646
Message 1980690 - Posted: 16 Feb 2019, 15:51:43 UTC

What they are useful for is comparing performance of individual systems for tweaking purposes.

..but only if they remain constant in terms of number given for a fixed parameter and not waggle around like [insert phrase]...

will "a xxxxxxxx in a shirtsleeve" do?
ID: 1980690 · Report as offensive
Profile Raistmer
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 16 Jun 01
Posts: 6115
Credit: 99,637,602
RAC: 44,441
Russia
Message 1982427 - Posted: 27 Feb 2019, 7:20:19 UTC

LoL, just as with elections. Important not how vote, but how count ;)
Credit-based metric is screwed. Even between AP and MB types of SETI work, not saying about inter-project comparison.
Even on earlier days when CreditScrew wasn't even in project my AMD GPU got more than order of magnitude more from MilkyWay than from SETI, credit-wise.
So, select top-100 and count share of that for SETI is about the same as count one from each state no matter how big state is. ;D
SETI apps news
We're not gonna fight them. We're gonna transcend them.
ID: 1982427 · Report as offensive

Message boards : Number crunching : Top 100 - Why so low?


 
©2019 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.