Ghost wu's back?

Message boards : Number crunching : Ghost wu's back?
Message board moderation

To post messages, you must log in.

Previous · 1 · 2

AuthorMessage
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1091607 - Posted: 30 Mar 2011, 12:41:10 UTC

I still have 110k in 'real' pending credits showing.....
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1091607 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34837
Credit: 261,360,520
RAC: 489
Australia
Message 1091615 - Posted: 30 Mar 2011, 13:37:24 UTC - in response to Message 1091607.  
Last modified: 30 Mar 2011, 13:48:56 UTC

I still have 110k in 'real' pending credits showing.....

But I seriously doubt that you'll get near all that 110K as D.A. is now hiding later pending credits with his new credit system, which seems very suspect of late, so that no one will hit him with Miep's rolling pin. :D

Cheers.
ID: 1091615 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1091619 - Posted: 30 Mar 2011, 13:54:36 UTC

I believe the new credit system was implemented around June. They had just left the code in the generate "pending credit". In the March 8th maintenance they put in the update that removed this obsolete feature. I know there were complaints from several people about them thinking the new system was granting them less credit. Which could be true. The theory is that less efficient processing will get less credit. So if your credit went down you were just having to much credit granted.

From looking over my daily RAC history for the year I'd say it had been pretty flat for me. With it between 45k and 55k most of the year staying around 50k. Except during very long outages or shutting-down/running machines. There are times when the RAC value seems to randomly just from 50k one day to 25k, then back to 50k the next day, but that sort of thing has occurred for the past several years if not longer.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1091619 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1091631 - Posted: 30 Mar 2011, 14:51:07 UTC
Last modified: 30 Mar 2011, 14:56:33 UTC

I've seen a couple of comments here regarding 'more or less efficient processing' affecting the granted credit.

Does that mean that on a given WU validated by a more efficient cruncher against a less efficient cruncher they would be awarded different credits for the same work?

Or would the less efficient cruncher penalize the more efficient cruncher by both getting awarded the lower of the two amounts?

The kitties wanna know.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1091631 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1091660 - Posted: 30 Mar 2011, 16:14:44 UTC - in response to Message 1091631.  

I've seen a couple of comments here regarding 'more or less efficient processing' affecting the granted credit.

Does that mean that on a given WU validated by a more efficient cruncher against a less efficient cruncher they would be awarded different credits for the same work?

Or would the less efficient cruncher penalize the more efficient cruncher by both getting awarded the lower of the two amounts?

The kitties wanna know.

I think there are rumors that it works both ways. In that whomever reports first sets the precedent. So a more efficient machine validating take a hit, but a less efficient machine gets a boost when validating. So a more efficient machine would want to always report work first.
[/rumor]
There was a lot going in the threads about the new credit system. That could be all rubbish.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1091660 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1091662 - Posted: 30 Mar 2011, 16:19:34 UTC - in response to Message 1091660.  

I've seen a couple of comments here regarding 'more or less efficient processing' affecting the granted credit.

Does that mean that on a given WU validated by a more efficient cruncher against a less efficient cruncher they would be awarded different credits for the same work?

Or would the less efficient cruncher penalize the more efficient cruncher by both getting awarded the lower of the two amounts?

The kitties wanna know.

I think there are rumors that it works both ways. In that whomever reports first sets the precedent. So a more efficient machine validating take a hit, but a less efficient machine gets a boost when validating. So a more efficient machine would want to always report work first.
[/rumor]
There was a lot going in the threads about the new credit system. That could be all rubbish.


Dunno......
In the past, the lowest credit claimed was the amount granted partly in order to ward off cheating. Perhaps the 'new credit' system cuts off avenues of cheating inherently. Dunno.

In the final analysis, I guess it does not matter that much to me, as I will be treated the same as anybody else on the project, give or take.

But it would be nice to know instead of having to say 'dunno'...LOL.

"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1091662 · Report as offensive
Profile James Sotherden
Avatar

Send message
Joined: 16 May 99
Posts: 10436
Credit: 110,373,059
RAC: 54
United States
Message 1091669 - Posted: 30 Mar 2011, 16:32:54 UTC
Last modified: 30 Mar 2011, 16:48:07 UTC

I was looking over my validated list and saw some work units that I got the small credit and some where I got the larger credit. If my CPU goes head to head with a GPU I get the smaller and vice versa. I didnt have time to see if Opp Apps gets the better of stock apps.

The CPU Versus GPU seemed to even out close to 50 50. But I didnt look at a big enough sample.

Im going to look after this in my spare time. Im thinking if My old P4 is getting the short end of the stick all the time I will pull it off line.

edit- After a quick glance Id say you get the credit of who ever is the fastest no matter if you finished first. Its how long your computer took for the task.
[/quote]

Old James
ID: 1091669 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1091680 - Posted: 30 Mar 2011, 16:53:11 UTC - in response to Message 1091669.  

I was looking over my validated list and saw some work units that I got the small credit and some where I got the larger credit. If my CPU goes head to head with a GPU I get the smaller and vice versa. I didnt have time to see if Opp Apps gets the better of stock apps.

The CPU Versus GPU seemed to even out close to 50 50. But I didnt look at a big enough sample.

Im going to look after this in my spare time. Im thinking if My old P4 is getting the short end of the stick all the time I will pull it off line.

edit- After a quick glance Id say you get the credit of who ever is the fastest no matter if you finished first. Its how long your computer took for the task.

I haven't seen any of my P4's change much over the past year. The fastest ones are normally in the 900-1100 RAC range. The only oddity I have seen if that the 3.6GHz machine has had higher RAC then the 3.8GHz machine consistently for several months. I have figured it must just be the AR of the work they are getting.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1091680 · Report as offensive
Previous · 1 · 2

Message boards : Number crunching : Ghost wu's back?


 
©2024 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.