Posts by JLDun
log in
1) Message boards : Cafe MilkyWay : "A better way to simulate accretion of the supermassive black hole at the center of the Milky Way developed" (Message 66033)
Posted 23 Dec 2016 by Profile JLDun
http://m.phys.org/news/2016-12-simulate-accretion-supermassive-black-hole.html

Scientists at Princeton University and the U.S. Department of Energy's (DOE) Princeton Plasma Physics Laboratory (PPPL) have developed a rigorous new method for modeling the accretion disk that feeds the supermassive black hole at the center of our Milky Way galaxy. The paper, published online in December in the journal Physical Review Letters, provides a much-needed foundation for simulation of the extraordinary processes involved.
2) Message boards : Number crunching : 'Short' N-Body 1.40 WU; what kind of results? (Message 61826)
Posted 2 Jun 2014 by Profile JLDun
Task 749722331 (as a part of WU 56083573 had a run time of 396.02 second on my computer (shorter for the wingman, but the point would be the same). The StdErr, on mine, was:

<core_client_version>7.2.42</core_client_version>
<![CDATA[
<stderr_txt>
<search_application> milkyway_nbody 1.40 Windows x86_64 double OpenMP, Crlibm </search_application>
Using OpenMP 2 max threads on a system with 2 processors
Number of particles in bins is very small compared to total. (0 << 100000). Skipping distance calculation
<search_likelihood>-9999999.900000000400000</search_likelihood>
14:01:52 (137936): called boinc_finish

</stderr_txt>

What kind of work ends up being done by these types of work units? (Since I've seen it before, and I know it validates...)
3) Message boards : Number crunching : Permanently showing "Don't need" for MW when cache very low (Message 61523)
Posted 16 Apr 2014 by Profile JLDun
Are any of your other projects running in "High Priority"? (That'll affect it, if it's happening...)
4) Message boards : Number crunching : 'Incomplete' preference file(s) for Separation 1.28 WU's? (Message 61510)
Posted 11 Apr 2014 by Profile JLDun
For task 705141094, while it completed and validated, I noticed in the stderr output it contained the lines

<search_application> milkyway_separation 1.28 Windows x86_64 double </search_application>
Reading preferences ended prematurely
Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4'
Switching to Parameter File

Using SSE4.1 path

How big a 'thing' is that?

(I don't usually get Separation WU's, so I don't have a chance to see this that often.)
5) Message boards : MilkyWay@home Science : How to create a 'Team' in my account profile page? (Message 61325)
Posted 4 Mar 2014 by Profile JLDun
Team Page, to join or create a team.
6) Message boards : Number crunching : What Are These Prefs For? (Message 61157)
Posted 20 Feb 2014 by Profile JLDun
My opinion-until we hear something official: it looks, to me, like someone is setting up preferences ahead of time, in 'anticipation' of the day when a screen saver is produced; and the listed preferences relate to the expected visuals. [Particles=Show some (solar) bodies; Coarse=A group of Particles....]
7) Message boards : MilkyWay@home Science : Several of my "Separation v 1.28" WU's failing (Message 61083)
Posted 11 Feb 2014 by Profile JLDun
Oh. Thanks.
8) Message boards : MilkyWay@home Science : Several of my "Separation v 1.28" WU's failing (Message 61080)
Posted 11 Feb 2014 by Profile JLDun
Example: Task 664273903, which according to the BOINC manager 'died' after 2 seconds. All of the (recent) ones that died on my computer appear to be the same type of WU and all quit after 2 seconds.
9) Message boards : Cafe MilkyWay : Word Link (Message 8058)
Posted 30 Dec 2008 by Profile JLDun
Principle
10) Message boards : Cafe MilkyWay : How can i add video?? (Message 8057)
Posted 30 Dec 2008 by Profile JLDun
I think it wouldn't happen, to save bandwidth.
11) Message boards : Number crunching : Are we doing science yet? (Message 7887)
Posted 20 Dec 2008 by Profile JLDun
So currently I am seeing a 20x improvement in my crunch times. Is that due to the optimized app, shorter WUs or both?

Thanks,
Jim


Keeping in mind that, yes, the old and new apps are processing the workunits differently:

Old App (1.22): ~24h38m (DCF around 48.xxx-49.xxx)
Milksop app (v3): ~0h28m (DCF around 0.93x)

New app (0.7): ~2h50m-3h2m (DCF 0.17x-.19x)

And this is for a two year old Dell Laptop (Intel 1.4GHz Celeron M, MMX SSE SSE2)


[edit]Typo[/edit]
12) Message boards : Number crunching : New Work (Message 7613)
Posted 11 Dec 2008 by Profile JLDun
but I manually edited my client_state file and lowered the DCF to 0.25

Mine's sitting at around .19xx...

And we still have work...
13) Message boards : Number crunching : credit comparison to other projects (Message 7591)
Posted 10 Dec 2008 by Profile JLDun

Message from server: No work sent
Message from Server: There was work available, but none for you as you are a greedy sod.
14) Message boards : Application Code Discussion : fpops and memory bound estimation (Message 7497)
Posted 6 Dec 2008 by Profile JLDun
Accurate fpop ... estimations

Don't know how much this helps, but my laptop is running the 0.7 application in 9466.515625 seconds (~3:0x) with a DCF of 0.196707; I don't know about others but the estimate can be reduced to around 1/5th in my case.
(What's the DCF for the project right now?)
15) Message boards : Number crunching : New Work (Message 7471)
Posted 5 Dec 2008 by Profile JLDun
Yes, I must get out more.

'Out'?





Oh well bowled, sir!


"Hey, batter, batter, batter..."
16) Message boards : Number crunching : New Work (Message 7470)
Posted 5 Dec 2008 by Profile JLDun
Yes, I must get out more.

'Out'?





Oh well bowled, sir!


"Hey, batter, batter, batter..."
17) Message boards : Number crunching : Work Units for V1.22 (Message 7419)
Posted 4 Dec 2008 by Profile JLDun
If you got a 1.22 then it might be one of the few left in the queue (as an already-split workunit). Most everything else will be 0.4/0.5/0.6.
Go ahead and crunch it since the information can still be used- it's not as extensive as 0.x, but it's useful.
18) Message boards : Number crunching : post milkyway_windows_intelx86 problems here (Message 7417)
Posted 4 Dec 2008 by Profile JLDun
Should I detach and the reattach after this WU finishes or just wait?

Hmmm. For 0.6 the highest estimated I've seen is 23:xx, with actual completion of 3:xx on a 1.4 GHz laptop. Detach might help?
19) Message boards : Number crunching : New Work (Message 7416)
Posted 4 Dec 2008 by Profile JLDun
Yes, I must get out more.

'Out'?
20) Message boards : Number crunching : 20 workunit limit (Message 7415)
Posted 4 Dec 2008 by Profile JLDun
Last year Astro over at Seti got his old P1 (60 MHz no SSE or MMX) running on the MultiBeam units using a opt app (someone custom compiled it for his system).

I remember reading bits of that thread. Part of what I missed what that it was a custom app, though.
I guess this just goes to show the "Anonymous Application Platform" can be have far-reaching consequences...


Next 20

Main page · Your account · Message boards


Copyright © 2017 AstroInformatics Group