stephenbrooks.orgForumMuon1Q&Abackground client MUCH slower?
Username: Password:
Search site:
Subscribe to thread via RSS
wirthi [Free-DC]
2002-09-07 04:12:25
Hi,

when running the graphical client, a "workunit" is calculated in about twenty minutes on my pc (P4 1400Mhz), but when I run the background client, after 3 hours there is still no (new) result in results.txt.

I am running the newest client and put the top 20 results of the best250 file into the results.dat.  The process runs with about 97% cpu-time (and it's the only programm running on the computer).

Are there so big differences in the calculation time of different workunits?  Or does only the background client use the information of the results.dat (and thus need more time to complete a workunit)?

Thanke for your information,
Wirthi
Pascal
2002-09-07 04:41:52
Hi and welcome aboard, wirthi!

The difference between the work uníts is very great.

You may have a look into results.txt or results.dat.
Every second line starts with

2.654790 (58975 particles) [v4.21b] {B75C17BC}
or
0.006108 (29262 particles) [v4.21b] {B75C17BC}, for example.

The 2.654790 is the yield of the simulator.

The simulator has different genetic algorithms that optimize by mixing the parameters. 
Look a bit (maybe some hours, if you're very interested in the algoriths) on this board here.  smile

If you can speak German, look at www.rechenkraft.net there are many things explained in the muon board.  wink

___________________________
Member of www.rechenkraft.net - German Website about Distributed Computing Projects

1: Athlon TB-C, 1.2 GC/s, 256 MB DDR-RAM, Erazor x², ADSL-Flatrate, NIC Intel, Win 98 SE Mainboard MSI-6380 Rev.  1
2: Pentium III, 600 MC/s, 256 MB RAM, NIC Intel, Win 98 SE

[This message was edited by Pascal [www.rechenkraft.net] on 2002-Sep-07 at 12:20.]
wirthi [Free-DC]
2002-09-07 09:48:44
Hi,

thanks for the quick answer!  I am not _that_ new, I have run the client before (a month ago); Every workunit was completed in about 15-20 minutes then.  I didn't use any "best250" file then (just my own results).

When I installed the client yesterday, I installed the best 250 file (at least some of the entries).  I started the graphical client (just to ensure the client is working properly) and the first workunit was completed in about 15 minutes, as i assumed.  So i startet the backgroud client and - no new result for 3 hours.  I just startet to wonder about that.  In another thread I read that "better" results take longer time, that clarifies things a bit ....

[GERMAN] Und ich kenn rechenkraft.net, wie sonst würd ich all diese wunderbaren Projekte entdecken smile Gute Arbeit!  [/GERMAN]

Thanks again,
Wirthi
Bananeweizen
2002-09-08 23:09:31
Hi all !

I guess you happened to watch two different simulation types in the graphical and background client.  The 15 minutes run could have been a simulation of type random (where all parameters are randomized) while the 3 hours simulation was a crossover, mutation or an interpolation of the existing high results.
On my Ahtlon-1200 a current run can last 5 hours so your 3 hours may be ok too.

Ciao, Michael.

www.Rechenkraft.net - the most comprehensive german website about internet based distributed computing projects.
Stephen Brooks
2002-09-21 12:58:31
I just translated a page from rechenkraft.net with you posting on it using BabelFish.  Banana Wheat???  Well either that's a translation error, a type of German breakfast cereal, or just a particularly odd username.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
Bananeweizen
2002-09-21 17:13:56
quote:
Originally posted by Stephen Brooks:
Banana Wheat???  Well either that's a translation error, a type of German breakfast cereal, or just a particularly odd username.



*lol*

You are right with your translation but not completely right with the interpretation.  In that case "wheat" stands for "beer brewed from wheat" (don't know if that is a well known type of beer in other countries as well as in Germany).  By adding some banana juice to a big glas full of that beer you get my favorite drink. 
Most people can't imagine such a mixture to be drinkable at all but if I can get them to try it the majority states that it is of good taste.

On this picture you can see what it looks like (it's a picture from our website party some weeks ago): Picture from Party

Ciao, Michael.

www.Rechenkraft.net - the most comprehensive german website about internet based distributed computing projects.
HETTATLONGUN
2002-09-21 22:33:37
Hi, wirthi!

Yes.  Some of the simulations take a great deal longer to process than others.

It seems to have been introduced with muon1v421b.

With my AMD Athlon 900MHz CPU, some of the simulations can take 6 hours or more.  It doesn't make any difference in which version of the program I run, the GUI version, the background version or the commandline version.

The commandline version is probably the best.  It tells you what type of simulation is running, and you can gague progress by looking at the nanoseconds go by.  A completed simulation takes 300.00 nanoseconds.  Heh, that's nanoseconds to the particles!  In real time it can be hours and hours!  big grin

Ahh, morning....Nothing like a good cigar, a cup of coffee and the sound of hot shell casings hitting the deck....HETT[TEP]
[DPC]inquestos
2002-10-24 16:52:09
quote:
Originally posted by Bananeweizen:
Hi all !

I guess you happened to watch two different simulation types in the graphical and background client.  The 15 minutes run could have been a simulation of type random (where all parameters are randomized) while the 3 hours simulation was a crossover, mutation or an interpolation of the existing high results.
On my Ahtlon-1200 a current run can last 5 hours so your 3 hours may be ok too.

Ciao, Michael.




I'm new here (I haven't even send a single result), but I think it has nothing to do with the backgroundmode: it's just the best 250-results that makes the difference, in a possitive way though, check your results, you should see some good results in the stats.  I might be wrong...
[ARS]odessit
2002-10-24 19:40:28
quote:
Originally posted by Bananeweizen:
[By adding some banana juice to a big glas full of that beer you get my favorite drink. 

eek
Oh boy...When I came to US, the very first thing that I eate was a big box of bananas (expensive rarity at that time in former USSR) Since that time after I ate a box of bananas all by myself in 1 day, I can't even look at them, not to mention spending better half of the night on the john... roll eyes
but I do like German beer without the banana juce any time big grin

and to bring topik back on track - yes it takes longer time razz

ARS Team Atomic Milkshake
Unofficial Muon1 FAQ
[DPC] TeamBVD - Floppus
2002-10-26 14:57:47
Odessit; please name your brand of "german" beer? 

I suspect when closer examined, it proves to be either Belgian or Dutch... (otherwise it's NOT beer wink)

[DPC] TeamBVD
Floppus
http://62.195.224.44/stuff
Supp
2002-10-28 20:30:49
quote:
I suspect when closer examined, it proves to be either Belgian or Dutch... (otherwise it's NOT beer )


...even more probable it's Czech - otherwise it's NOT BEER big grin

rm -rf /
wirthi [Free-DC]
2002-11-01 13:41:32
You wouldn't dare to forget Austrian beer, would you?  big grin

Cheers!
Wirthi
Bill[Romulus2]
2002-11-01 16:35:58
I have never heard of banana juice eek Sometimes I think I should get out more and then something like this comes up big grin
: contact : - - -
E-mail: sbstrudel characterstephenbrooks.orgTwitter: stephenjbrooksMastodon: strudel charactersjbstrudel charactermstdn.io RSS feed

Site has had 25160430 accesses.