Finally Decided To Set Up BOINC...

[ShotgunSteven]

I decided to take a break from DC after SETI Classic ended. I just now activated my new account, but realized I don't know a thing about BOINC. Does it use a screen saver like the old SETI? Is there a CLI version that is preferred, for efficiency?

Oh yeah, and did we beat out DSLR in the end results for Classic? I tried to find the stats but got a bunch of dead URLs.

 

[Zim Hosein]

Yes, BOINC has the same ability to show a screen saver

Welcome back to the TeAm ShotgunSteven :beer:

 

[Rattledagger]

The BOINC client contains 2 pieces:
1; BOINC "core" client, a cli that normally runs hidden, preferably insalled as service. Handles all uploads/downloads, that to crunch now and so on.
2; BOINC Manager, a gui that can use to monitor/control local or remote client. Atleast when installed as service, the Manager doesn't need to run at all.
There's also a cli-version of the Manager, giving you limited possibility to control BOINC.
Can actually also choose to use 3rd. party tools instead of the Manager, example BoincView that can monitor many computers at once.

When "attachs" to one or more projects, BOINC core client will automatically download Science-application(s). Some of the Science-applications does have a screensaver-part, but as long as you're not currently displaying the screensaver the science-application will perform like a cli.

For more info about BOINC, projects and so on, look on BOINC Wiki.

 

[Wolfsraider]

Originally posted by: ShotgunSteven
I decided to take a break from DC after SETI Classic ended. I just now activated my new account, but realized I don't know a thing about BOINC. Does it use a screen saver like the old SETI? Is there a CLI version that is preferred, for efficiency?

Oh yeah, and did we beat out DSLR in the end results for Classic? I tried to find the stats but got a bunch of dead URLs.

Yes, yes we did, even more so after "The cheaters results" from the different teams were removed

 

[ShotgunSteven]

Originally posted by: Wolfsraider

Originally posted by: ShotgunSteven
I decided to take a break from DC after SETI Classic ended. I just now activated my new account, but realized I don't know a thing about BOINC. Does it use a screen saver like the old SETI? Is there a CLI version that is preferred, for efficiency?

Oh yeah, and did we beat out DSLR in the end results for Classic? I tried to find the stats but got a bunch of dead URLs.

Yes, yes we did, even more so after "The cheaters results" from the different teams were removed

Sweet! Is there a page out there somewhere that shows final overall results?

EDIT: Just found this over on BBR's site.

#1: 24.336.008: Lost 61.782: OcUK - Overclockers UK
#2: 22.602.537: Lost 1.958.223: SETI.Germany
#3: 19.873.567: Lost 116.675: Team Ars Technica Lamb Chop
#4: 14.598.335: Lost nothing: TeAm AnandTech
#5: 14.508.591: Lost 5.179.373: SETI@Netherlands
#6: 14.151.915: Lost 93.324: BroadbandReports.com Team Starfire
#7: 11.318.470: Lost nothing: Hewlett-Packard.

 

[Rattledagger]

Originally posted by: ShotgunSteven
Sweet! Is there a page out there somewhere that shows final overall results?

EDIT: Just found this over on BBR's site.

Well, you don't need to go so long to find that table, 900 cheaters removed/adjusted

For more detailed listings, AFAIK Seti-Netherland still shows the "classic" stats, and SetiatWork possibly also still works...

 

[GLeeM]

Welcome back ShotgunSteven

 

[Soggysocks]

Originally posted by: ShotgunSteven

Originally posted by: Wolfsraider

Originally posted by: ShotgunSteven
I decided to take a break from DC after SETI Classic ended. I just now activated my new account, but realized I don't know a thing about BOINC. Does it use a screen saver like the old SETI? Is there a CLI version that is preferred, for efficiency?

Oh yeah, and did we beat out DSLR in the end results for Classic? I tried to find the stats but got a bunch of dead URLs.

Yes, yes we did, even more so after "The cheaters results" from the different teams were removed

Sweet! Is there a page out there somewhere that shows final overall results?

EDIT: Just found this over on BBR's site.

#1: 24.336.008: Lost 61.782: OcUK - Overclockers UK
#2: 22.602.537: Lost 1.958.223: SETI.Germany
#3: 19.873.567: Lost 116.675: Team Ars Technica Lamb Chop
#4: 14.598.335: Lost nothing: TeAm AnandTech
#5: 14.508.591: Lost 5.179.373: SETI@Netherlands
#6: 14.151.915: Lost 93.324: BroadbandReports.com Team Starfire
#7: 11.318.470: Lost nothing: Hewlett-Packard.

I suppose I now have to watch my back door....since you will ...:|.....no doubtedly be catching up to me :disgust:


But thats ........OK

Welcome back and lets get crunchin man

 

[panhead49]

hey steve!!!...good to see you .. been a while......i love those few sentences.....goes to show what a great bunch we have

quote:
#1: 24.336.008: Lost 61.782: OcUK - Overclockers UK
#2: 22.602.537: Lost 1.958.223: SETI.Germany
#3: 19.873.567: Lost 116.675: Team Ars Technica Lamb Chop
#4: 14.598.335: Lost nothing: TeAm AnandTech
#5: 14.508.591: Lost 5.179.373: SETI@Netherlands
#6: 14.151.915: Lost 93.324: BroadbandReports.com Team Starfire
#7: 11.318.470: Lost nothing: Hewlett-Packard.

 

[Assimilator1]

Hi Steve
Welcome back to The search

BOINC is ok ,only thing that annoyes me is that you can't set when to connect to berkely seperatly from the days cache you have ,like you could with SETIQ.

 

[Rattledagger]

Well, under BOINC you can set "Use network only between the hours of...".

As for SetiQueue, if my memory isn't too fuzzy, long time since used SetiQueue, the only difference between min and max in SetiQueue is how aggressively SetiQueue was trying to download work, there above min-queue you could slow-down according to "wu download gap" and so on, but normally you'll try to fill-up to max queue setting anyway...

 

[Assimilator1]

With SETIQ you could set your WU cache to 7days (or whatever) but you could still connect 24/7 if you wanted.
So if the servers went down, from that point you'd have a 7 day cache.
AFAIK you can't do that with BOINC, so it would depend when you last connected ,if the servers went down on your 6th day of 'cache' then you'd only have 1 day left.)

I don't see how limiting network hours useage would help?

 

[Rattledagger]

Assimilator1, you're not the 1st. that gets confused by the wording, "Connect every N days", in reality BOINC will normally ask for more work immediately when cached work drops below your cache-setting.

Meaning, after result-1 is finished and uploaded, and sometime during your crunching of result-2, you'll ask for more work.

The reason it doesn't always follow this is, BOINC bases everything after expected run-times, but in reality crunch-times variates some.

Example, if uses 1.5h on VHAR and 2h on "normal" wu, and uses 5-day cache-setting, if you've crunched through a bunch of VHAR the BOINC-client expects ever wu takes 1.5h. meaning 80-81 wu cached. But, immediately after you've finished a "normal" wu, BOINC-client expects wu takes 2h, meaning you've suddenly got 160 hours cached, or 6.666 days. Since 6.666 > 5, you'll crunch 1.666 days, 40h or 20 results before next time asks for more work...


Still, if single-project or 99%-resource-share or something, most of the time you'll ask for more work when you've finished 1-2 results, regardless of 0.2 or 5-days cache-setting.

 

[Orange Kid]

For more info about BOINC, projects and so on, look on BOINC Wiki.

I've always thought of this place as a wiki in itself

Too bad it had to come down to someone 'coining' a silly word for what was already happening in most good forums for people to take notice of another wonder of the internet

 

[Assimilator1]

Originally posted by: Rattledagger
Assimilator1, you're not the 1st. that gets confused by the wording, "Connect every N days", in reality BOINC will normally ask for more work immediately when cached work drops below your cache-setting.

Meaning, after result-1 is finished and uploaded, and sometime during your crunching of result-2, you'll ask for more work.

The reason it doesn't always follow this is, BOINC bases everything after expected run-times, but in reality crunch-times variates some.

Example, if uses 1.5h on VHAR and 2h on "normal" wu, and uses 5-day cache-setting, if you've crunched through a bunch of VHAR the BOINC-client expects ever wu takes 1.5h. meaning 80-81 wu cached. But, immediately after you've finished a "normal" wu, BOINC-client expects wu takes 2h, meaning you've suddenly got 160 hours cached, or 6.666 days. Since 6.666 > 5, you'll crunch 1.666 days, 40h or 20 results before next time asks for more work...


Still, if single-project or 99%-resource-share or something, most of the time you'll ask for more work when you've finished 1-2 results, regardless of 0.2 or 5-days cache-setting.

Ah very useful to know that ,thanks

And yes their wording is mis-leading