BOINC 4.09

Message boards : Number crunching : BOINC 4.09
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · 4 · Next

AuthorMessage
Profile Contact
Volunteer tester
Avatar

Send message
Joined: 16 Jan 00
Posts: 192
Credit: 2,249,004
RAC: 0
Canada
Message 26940 - Posted: 16 Sep 2004, 3:41:08 UTC
Last modified: 16 Sep 2004, 3:45:31 UTC

OK. Really fond of 4.08. Best Yet.
Hope 4.09 kills a few more bugs. Especially for dialup and slow puters.
Don't wanna try 4.09 till i have an unsolvable problem.
But, what say you!
ID: 26940 · Report as offensive
HachPi
Avatar

Send message
Joined: 2 Aug 99
Posts: 481
Credit: 21,807,425
RAC: 21
Belgium
Message 26960 - Posted: 16 Sep 2004, 5:34:47 UTC

If it ain't broke, don't try to fix it...

Greetz from Belgium ;-))




ID: 26960 · Report as offensive
ric
Volunteer tester
Avatar

Send message
Joined: 16 Jun 03
Posts: 482
Credit: 666,047
RAC: 0
Switzerland
Message 26965 - Posted: 16 Sep 2004, 7:20:27 UTC - in response to Message 26960.  
Last modified: 16 Sep 2004, 7:23:20 UTC

as usual:

-use to your own risk

-make first a backup of your boinc folder into an other folder

4.09 EXE

symbol files

Installed so far 3 clients (xp, not sp2) will wait a while to see behaviour.

so far NO more ssmmaalll ffrrreeezzzeeee
ID: 26965 · Report as offensive
ChinookFoehn

Send message
Joined: 18 Apr 02
Posts: 462
Credit: 24,039
RAC: 0
Message 26968 - Posted: 16 Sep 2004, 7:31:14 UTC - in response to Message 26965.  
Last modified: 18 Dec 2004, 6:38:50 UTC

ID: 26968 · Report as offensive
Profile Kevin N. Shapley
Volunteer tester
Avatar

Send message
Joined: 1 Jan 00
Posts: 100
Credit: 2,539,295
RAC: 0
United States
Message 26975 - Posted: 16 Sep 2004, 8:14:09 UTC

I have been using 4.05 on two AMD and one Intel boxes with no visible signs of any problems.

Am I lucky or is there something I am missing by not upgrading to newer versions of BOINC?

Seriously, WU's download, crunch, upload, then download some more and the only pesky thing is having to do a manual update to get the ready to reports out of the way.

Are the newer versions somehow better coded to do what they are required to do faster, better, or more efficiently?

Thank you for a serious answer in advance.


<a> [/url]
ID: 26975 · Report as offensive
Heffed
Volunteer tester

Send message
Joined: 19 Mar 02
Posts: 1856
Credit: 40,736
RAC: 0
United States
Message 26980 - Posted: 16 Sep 2004, 8:38:13 UTC - in response to Message 26975.  

> I have been using 4.05 on two AMD and one Intel boxes with no visible signs of
> any problems.
>
> Am I lucky or is there something I am missing by not upgrading to newer
> versions of BOINC?

No. 4.05 is the latest official publice release. Anything above that is alpha. If 4.05 works for you, (that's what I'm running) keep it until the next public release.

ID: 26980 · Report as offensive
Profile Kevin N. Shapley
Volunteer tester
Avatar

Send message
Joined: 1 Jan 00
Posts: 100
Credit: 2,539,295
RAC: 0
United States
Message 26983 - Posted: 16 Sep 2004, 8:54:23 UTC


@Heffed

Thanks, that is what my gut said, but you know how new shiny things can be..;)

ID: 26983 · Report as offensive
Profile mikey
Volunteer tester
Avatar

Send message
Joined: 17 Dec 99
Posts: 4215
Credit: 3,474,603
RAC: 0
United States
Message 26993 - Posted: 16 Sep 2004, 10:28:27 UTC - in response to Message 26975.  


> Seriously, WU's download, crunch, upload, then download some more and the
> only pesky thing is having to do a manual update to get the ready to reports
> out of the way.
>
The amount of time between reconnections to send those is adjustable from your account page. Mine was set at 2 days and I just changed it to 1 day and it seems to be working better, for me.

ID: 26993 · Report as offensive
STE\/E
Volunteer tester

Send message
Joined: 29 Mar 03
Posts: 1137
Credit: 5,334,063
RAC: 0
United States
Message 26995 - Posted: 16 Sep 2004, 10:45:04 UTC

A word of WARNING to those with P4 HT CPU's, I installed the 4.09 on 1 PC that I've been having trouble with running CPDN WU's and if nothing else it knocked the Piss out of my Benchmark Score.

I went from Mid 1700's with HT Enabled to Mid 1300's...Floating Point Speed
With HT Disabled I went from about 2050 to 1650...Floating Point Speed

So it looks about like a 400 Point loss with the v4.09 for HT CPU's or maybe it was just that PC in particular, I don't know. But I'm not putting it on another PC until somebody confirms this loss of Benchmark score...
ID: 26995 · Report as offensive
ric
Volunteer tester
Avatar

Send message
Joined: 16 Jun 03
Posts: 482
Credit: 666,047
RAC: 0
Switzerland
Message 26997 - Posted: 16 Sep 2004, 10:56:57 UTC - in response to Message 26968.  

> so far NO more ssmmaalll ffrrreeezzzeeee

Guten Morgen Herr Richard

>Bitte, was meinen Sie dabei?

nun ja nicht ganz einfach zu beschreiben.

Da sind noch finetuning Arbeiten im Gange welche etwas mit dem
"Kommunikations-Teil" des Clients zu tun haben, soweit ich verstanden habe.

einzelne User berichten, über Stabilitätsprobleme der offiziellen clients
ob die in der Masse zutrifft, lässt sich so nicht erhärten.

Bei mir wirkt sich dies so aus, manchmal, scheint der Cursor zu hängen,
schreibe weiter, jedoch die Characters are not displayed und Bruchteile einer Sekunde später wird der Bildschirm wieder gerefreshed und der "scheinbar"
kurzzeitige Hänger ist weg. Es spielt ME keine Rolle ob der Client einen Transfer ausübt oder nur rechnet.

Dieses Verhalten, so kann ich "für mich" berichten, zeigt der client 4.09 nicht mehr.
dies meinte ich mit
> so far NO more ssmmaalll ffrrreeezzzeeee
nicht ganz einfach, in Worte zu kleiden.

für mich ist einer der haupsächlichen Gründe einen higher level client zu verwenden, die stark verbesserte Managability, die entfernte Steuerung
des network access zum Beispiel finde ich einen Hit. Die allerdings nur über rpc. Aber dies ist eine persönliche Meinung von mir.

Ich denke, mit einer vorausgehenden Sicherstellung kann das Risiko, während eines Upgrades des Client Daten zu verlieren, doch deutlich reduziert werden.

Hatte bis Anhin noch keine Datenverluste mit diesem Vorgehen

friendly regard

ric




@Kevin
>Am I lucky or is there something I am missing by not upgrading to newer versions of BOINC?

This is a very personal choise.
There is some ++ to migrate, also some -- to do it.

Due I'm using a Boinc Add-on only based over rpc, I "have" to use a boinc client, able to serve information over this technique. Some more advanced remote procedure call are implemented in client 4.06 and above.
This is for ME a valid reason to give at least a try.

(don't tell it anybody else, but for me, I never had a "better" client")

2 of the clients a attached to cpdn and seti.

It looks like the "thingy" with the preemting is solved much better.

I do see my cpdn and seti clients, preemting (now called pausing) on a frequent manner.


The other thing, be lucky, if you system is not showing the possible
non linearity, perhaps you have it perhaps not, perhaps your system are to fast to recognize it as a "failure" and handle it themself.

If you don't see any problems on your side, there is basicaly not need to upgrade until official launch.


friendly

ric
ID: 26997 · Report as offensive
Ingleside
Volunteer developer

Send message
Joined: 4 Feb 03
Posts: 1546
Credit: 15,832,022
RAC: 13
Norway
Message 26999 - Posted: 16 Sep 2004, 11:18:36 UTC - in response to Message 26995.  

> A word of WARNING to those with P4 HT CPU's, I installed the 4.09 on 1 PC that
> I've been having trouble with running CPDN WU's and if nothing else it knocked
> the Piss out of my Benchmark Score.
>
> I went from Mid 1700's with HT Enabled to Mid 1300's...Floating Point Speed
> With HT Disabled I went from about 2050 to 1650...Floating Point Speed
>
> So it looks about like a 400 Point loss with the v4.09 for HT CPU's or maybe
> it was just that PC in particular, I don't know. But I'm not putting it on
> another PC until somebody confirms this loss of Benchmark score...
>
>

The benchmark has undergone some changes so it now actually does the different tests and isn't optimized-away by the compiler. This of course gives lower benchmark-scores, but this is fixable by changing cobblestone-factor again.
ID: 26999 · Report as offensive
STE\/E
Volunteer tester

Send message
Joined: 29 Mar 03
Posts: 1137
Credit: 5,334,063
RAC: 0
United States
Message 27001 - Posted: 16 Sep 2004, 11:24:04 UTC

The benchmark has undergone some changes so it now actually does the different tests and isn't optimized-away by the compiler. This of course gives lower benchmark-scores, but this is fixable by changing cobblestone-factor again.
==========

It's also fixable by going back to the v4.05 which I did and got my 400 points back...hehe ;)

ID: 27001 · Report as offensive
Profile Paul D. Buck
Volunteer tester

Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 27015 - Posted: 16 Sep 2004, 12:15:31 UTC - in response to Message 26999.  

Ingleside,


> The benchmark has undergone some changes so it now actually does the different
> tests and isn't optimized-away by the compiler. This of course gives lower
> benchmark-scores, but this is fixable by changing cobblestone-factor again.

Yes, but this might get my experience with work done and credit claimed and granted. My fastest and HT processors are asking for more credit than the other participant's computers. Besides, it is about the science...
<p>
For BOINC Documentaion: Click Me!


ID: 27015 · Report as offensive
Profile PyroFox
Avatar

Send message
Joined: 5 Apr 03
Posts: 155
Credit: 213,891
RAC: 0
Canada
Message 27019 - Posted: 16 Sep 2004, 12:29:39 UTC

or IS it??

..



of course it is :P who doesn't want to find that elusive signal...

*i don't see any haaaaaannnddss!*

;)
-Fox
[/url]
ID: 27019 · Report as offensive
Profile Patrick_

Send message
Joined: 3 Apr 99
Posts: 30
Credit: 9,657
RAC: 0
United States
Message 27030 - Posted: 16 Sep 2004, 13:10:45 UTC

So far so good! Running much better than 4.08, no freezes so far, running XP SP2. The CPU benchmarks took a little longer than normal, but nothing out of the ordinary.
[url=http://setiweb.ssl.berkeley.edu/]
ID: 27030 · Report as offensive
Profile xi3piscium
Volunteer tester
Avatar

Send message
Joined: 17 Aug 99
Posts: 287
Credit: 26,674
RAC: 0
China
Message 27041 - Posted: 16 Sep 2004, 14:23:11 UTC
Last modified: 16 Sep 2004, 15:29:49 UTC

4.09 for nix.....anyone know where?

<img>
ID: 27041 · Report as offensive
SpaceRat
Avatar

Send message
Joined: 17 Feb 02
Posts: 47
Credit: 799,501
RAC: 0
Germany
Message 27044 - Posted: 16 Sep 2004, 14:30:45 UTC - in response to Message 27041.  

> 4.09 for nix.....anyone know where?

Here is the Link

Boinc 4.09




TEAMLESS??
Drop by ours and have a look SETI Synergy
ID: 27044 · Report as offensive
webmaster10

Send message
Joined: 17 Jun 99
Posts: 4
Credit: 64,474
RAC: 0
United States
Message 27052 - Posted: 16 Sep 2004, 15:15:57 UTC

Loaded 4.09 last night on Win98 SE, connection wireless. Run Seti & LHC. Smooth, did benchmarks w/o burping, work units running right along, uploaded completed units fine.

By jove, I think they've got it.
ID: 27052 · Report as offensive
Nuwanda
Avatar

Send message
Joined: 2 Aug 03
Posts: 71
Credit: 1,337,642
RAC: 0
India
Message 27053 - Posted: 16 Sep 2004, 15:28:28 UTC - in response to Message 26960.  

> If it ain't broke, don't try to fix it...
>
> Greetz from Belgium ;-))
>
>
>


that is so true. no problems on my AMD here, so unless i do encounter any (or am required to), im sticking to 4.05
----------------------------
Team X-BTF S@H Forums
S@h Berkeley's Staff Friends Club © member
<img>
ID: 27053 · Report as offensive
Profile Patrick_

Send message
Joined: 3 Apr 99
Posts: 30
Credit: 9,657
RAC: 0
United States
Message 27064 - Posted: 16 Sep 2004, 16:19:53 UTC

I can't upload now... either 4.09 or server is down.
[url=http://setiweb.ssl.berkeley.edu/]
ID: 27064 · Report as offensive
1 · 2 · 3 · 4 · Next

Message boards : Number crunching : BOINC 4.09


 
©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.