climateprediction.net home page
Posts by old_user132880

Posts by old_user132880

1) Questions and Answers : Windows : Computation error (Message 24220)
Posted 6 Sep 2006 by old_user132880
Post:
John

You\'re trying to run the new Coupled Ocean models with an obsolete version of BOINC. They need a 5.* version, and you still have 4.19

And make sure that you uninstall 4.19 before trying to upgrade, or you\'ll probably end up with split versions. The 5.* versions are completely different to the 4.* versions.



I do daily backups, and my most recent model crashed in the same exact spot as it did before. So....not the best of models then?
2) Message boards : Number crunching : How far along are you??? (Message 23676)
Posted 20 Jul 2006 by old_user132880
Post:
After my first hadcm errored out at abt 47%..I am well into the second one and another is running as well (14.2% on one and 2.44 on another).

Is it possible I just got a bad unit? I know stuff like this has happened before....
3) Questions and Answers : Windows : Comments for \'Generic solutions to models\' sticky (Message 23584)
Posted 11 Jul 2006 by old_user132880
Post:
Hmmm after 45% or so, my hadcm crashed....The PC wasn\'t being used any differently and I do have BOINC excluded from Norton.....
4) Questions and Answers : Wish list : reduce disk usage. (Message 21739)
Posted 30 Mar 2006 by old_user132880
Post:
is waiting for a dual quad core mobo and the quad cpus to roll out.

I\'m simply waiting now for a twin dual core mobo.....

May be waiting awhile.
5) Questions and Answers : Preferences : Fine Tuning a P4 Dual core (Message 21677)
Posted 28 Mar 2006 by old_user132880
Post:
Be wary of the overclocking. This project is very sensitive to it and may turn up errors even when all other checks tell you the computer is ok.


Ok, am not experiencing problems, but that doesn\'t answer my question:

I see someone with a straight P4 churning out units @ 800 hours to my dual cores 1100... 2.8 Gghtz, o/c\'d to just under 3.....
6) Questions and Answers : Getting started : how long???? (Message 21676)
Posted 28 Mar 2006 by old_user132880
Post:
Your P4 is registered as 2.0GHz, and 700 hours sounds about right to me. CPDN results are very computationally intensive, but they trickle results back to the scheduler every 10802 timesteps. There are 259284 timesteps for each of the 3 phases, so trickles are returned every 1.39% or so.
<br><a href=\"http://www.teampicard.net/\"><img src=\"http://www.teampicard.net/images/picardmini.gif\"></a><a href=\"http://climateapps2.oucs.ox.ac.uk/cpdnboinc/team_display.php?teamid=3\">Join us here</a>


Ok so with a P4D and 1.024 gigs of DDR2 pc4300 RAM, why is my real time calcs at 1100 hours? What can I do to tweak this....?
7) Questions and Answers : Getting started : Multiple unrecoverable errors for sulphur project (Message 21675)
Posted 28 Mar 2006 by old_user132880
Post:
Well, I was suffereing the same thing and quit using the screen saver graphics. Sincethen, zero miscues. I updated my drivers from ATI tothe latest, and it\'s a new vid card, but that didn\'t help.....

So I don\'t use the \"view graphics\" option.
8) Questions and Answers : Windows : Very long calculation time ? (Message 21674)
Posted 28 Mar 2006 by old_user132880
Post:
running a P4D and 1.024 megs of RAM, although the estimates, and I stress \"estimates\", are of around 1500 hours, I have experienced a rality of 1000 to 1100 hours.......

So don\'t let the estimates scare you, and whatever portions you finish are of value to the researchers....
9) Questions and Answers : Windows : Am I wasting mine & CPDN\'s time? (Message 21673)
Posted 28 Mar 2006 by old_user132880
Post:
One of your models failed with: exit code 1073807364

While we\'ve never been able to pin this down to an exact cause, it\'s something to do with graphics, possible a Windows conflict.
Some people have fixed it by updating the driver for their graphics card from the card maker\'s web site.
One person recently reported that he was using Excel at the time of the failure.

In your previous post you said that your computer was overclocked. The climate program is huge, complex, and touchy, on computers that are not 100% stable.
Overclocking is NOT recommended.

And BOINC version 4.43 is a bit old, and should be upgraded to the latest 5.2.13 version.



I\'d have to look again, but iirc, I am o/c\'d at 5 or 10% on my P4D...and I simply quit using the graphics. After that, to date, I have experienced no more WU crashes. I do remember you telling me a couple crashed because they were \"bad\" WU\'s.....
10) Questions and Answers : Windows : Am I wasting mine & CPDN\'s time? (Message 21672)
Posted 28 Mar 2006 by old_user132880
Post:
Co2..you mean you haven\'t made a solar array/storage battery system? :shocked:
11) Questions and Answers : Windows : Update openGL drivers against climate pred. crashes ? Corr\'ed (I hope) (Message 21529)
Posted 23 Mar 2006 by old_user132880
Post:
After the first three WU\'s crashed on me while using the graphics/screen saver, even though two apparently were just bad WU\'s, I am nearing the first 100% completeion (should be today)...perhaps onkly coincidentally because I no longer use the graphics.

I no longer use the graphics for that and a few other reasons. Even with a 512 meg vid card, and the most up-to-date drivers, I notice a marked slow down. Cuttingthe XpPro swap file to zero seems to improve this a tad, and I see physical RAM useage increase (1.024 gigs at the moment). I run a slight overclock (10%...3.2 GHtz roughly) and that also seems to slightly improve some other minor issues I was seeing connected to BOINC.

A friend is running an AMD 64 bit cpu mobo and I am running a P4D and there appears little advantage to the AMD inthe real world. We run the same WU\'s (Seti, CPDN and Einstein), same amount of RAM, wirtually same clock speed cpu\'s.....yet his AMD will finish less than an hour ahead of me. So if anyone\'s wondering, for the extra $100 or so for the AMD....he doesn\'t think it was worth it.
12) Questions and Answers : Preferences : Fine Tuning a P4 Dual core (Message 20929)
Posted 1 Mar 2006 by old_user132880
Post:
Any suggestions for ramping up a P4 Dual core to crunch WU\'s a bit more rapidly?

I am running my P4D @ 5% overclock ( a safe amount, remains in heat generation safe parametres, which makes it about 3 GHtz, with 1.024 gigs of DDR2 RAM.

Any suggestions would be greatly appreciated.
13) Questions and Answers : Windows : Screensaver (Message 20926)
Posted 1 Mar 2006 by old_user132880
Post:
I\'d not run the graphics, really. I, as well as others, was having a problem with CPDN WU\'s crsahing while running the globe graphics. Since I quit running it at all, I have had no problems with crashes.

Crashes, in spite of running a 512 meg video card, 1.024 gigs of DDR2 RAM and a P4 dual-core processor...and all XP Pro and driver updates.

So, in my nsho, I recommend against the CPDN graphics.
14) Questions and Answers : Windows : Crashed WU retsarts.... (Message 20208)
Posted 14 Feb 2006 by old_user132880
Post:
You can only restart from a backup made prior to the crash.

Your 3rd model, even though it crashed, did so after the end of phase 1, which, as you can see by the graphs on the model page, uploaded successfully. (Click on the P1s at the bottom of the page.)
So the extra data contained in this phase 1 upload will be valuable to the researchers for use in the start of the coupled ocean model.

There was a bad batch of data sets issued in December/January which failed after phase 1, so it wasn\'t your fault.

Upgrading to the latest graphics card drivers may solve the display problems.
But Suspend and Exit BOINC, then backup the BOINC folder first.



I don\'t feel it\'s anything to do with my graphics drivers as I am using a very new 256 meg pci-e card and accordingto the Sapphire website I do have the newest drivers.

At any rate, listening to all who have had crashed models that seem to be vid card related, and due to the fact that it slows calculation down, I prefer not to use the graphic representation at all. Better safe than sorry. ;)

Maybe I am overlooking things, but is Crunch3R helpful in speeding up the process? Or, are there quality drawbacks to doing such a thing? I\'d think there\'d be some drawbacvks, because if it was overall beneficial, BOINC wouldprobably have incorporated such things into the program.

I am running SETI and Einstein as well, and have already overclocked 5%, a safe amount. looking at the stats for higher overclocking would concern me as the higher voltages could potentially start degrading pn junctions within the cpu. Heat isn\'ta problem with a Thermaltake liquid cooler and \"exotic\" fluid...I am running at abt 38 degrees celsius and 5% o/c.
15) Questions and Answers : Windows : Crashed WU retsarts.... (Message 20188)
Posted 14 Feb 2006 by old_user132880
Post:
Is there any way to restart crashed WU\'s? I think the problem I was having was due to use of the graphicdisplay and have quit using it after 3 crashes @ the 20% mark.
16) Questions and Answers : Windows : Delete account (Message 20020)
Posted 7 Feb 2006 by old_user132880
Post:
You can detach from a project in BOINC, but deletion of an account? Unlikely for some obvious reasons....
17) Questions and Answers : Windows : Will ClimatePrediction burn out my CPU? (Message 20019)
Posted 7 Feb 2006 by old_user132880
Post:
+Most oft he newer mobo\'s have temp sensors for the board and the cpu. It merely takes having the appropriate software to monitor these signals.

For example, with my P5LD2 -DL, AI Booster not only tells me cpu temp (which for the PentiumD, I run at abt 58 degrees C, avg.), and mobo temp, but case fan speed as well as power suppply fan speed. Italso tells me the core voltage and I can obtain other readings.

Because of being able to keep an eye on these readings, I know when I start running 60 degrees C avg, it\'s time to clean out the case, specifically the cpu heatsink.

Now then, towers obviously have an advantage over laptops in volume to dissapate heat and choices for auxilliary cooling that laptops don\'t, so it\'s vital they have some one to observe temp trends.

While it\'s unlikely you\'ll fry your cpu while not seriously overclocking it, the cooler you can keep the electronics, the more efficiently they\'ll operate.
18) Questions and Answers : Windows : sulphur model program crashes repeatedly (Message 20018)
Posted 7 Feb 2006 by old_user132880
Post:
I had the same problem repeatedly.
On one of my machines i have Boinc installed as a service and on the other its standard.
On the first one I have tried to reinstall Boinc and reset all files, and nothing seems to help.
On the last one it helped to reinstall Boinc.
I never use the graphics due to the fact that crashes Einstein projects.


Hmmm...I had the problem that climatepredictor WU\'s would crash and nothing else would (Einstein, SETI)....
19) Questions and Answers : Windows : sulphur model program crashes repeatedly (Message 19696)
Posted 27 Jan 2006 by old_user132880
Post:
Are you running the graphics with it? If so, don\'t!
20) Questions and Answers : Windows : How do I compute faster? (Message 19695)
Posted 27 Jan 2006 by old_user132880
Post:
I have 4.096 gigs of the fastest DDR2 RAM there is, and am running a 3.2 gig P4D. According to the estimated time, it\'ll take months to finish. However, after 12 days of running I am at slightly over 20% completion.

Initially, I had problems with models crashing, but I was running the graphics too. Sincethen, I don\'t run the graphics and this seems to cure a pandemic problem.


Next 20

©2024 climateprediction.net