Dues the sulpher model have the same number of timesteps to a Trickle as the older models?

How many Trickle's to a model?

Any chance of a time estimate for a P4 2.4 Gig system finishing one? I run it about 15 hours a day (other programs the other 9 hours). Can give it more time if needed to finish by Jan. 31, 2006

Cheers
Ray Brown
<A>Rays-Place.com</A>

<A> BOINC WIKI</A>

<IMG SRC="http://www.boincsynergy.com/images/stats/comb-1842.jpg">

There's some info in the <a href="http://boinc-doc.net/boinc-wiki/index.php?title=Sulphur_Cycle"> BOINC Wiki,</a> and <a href="http://www.climateprediction.net/board/viewtopic.php?t=3159"> here</a> in the community forum.

Ok well my P3 800MHz machine just downloaded a sulphur_cycle workunit. It immediately went into EDF and now no other projects are running. Hrmm. I also notice this workunit has an earlier deadline. Now it will run just this CPDN workunit for 6 months without downloading from the other projects. Should I continue running it even though it may not make the deadline or just abort it? If I do abort it is there any way to guarantee getting a regular hadsum_4.13 workunit with a one year deadline? I currently have a estimated time of 4710 hours and the deadline is 1/28/2006. Any info would be appreciated.

&gt; Ok well my P3 800MHz machine just downloaded a sulphur_cycle workunit. It
&gt; immediately went into EDF and now no other projects are running. Hrmm. I also
&gt; notice this workunit has an earlier deadline. Now it will run just this CPDN
&gt; workunit for 6 months without downloading from the other projects. Should I
&gt; continue running it even though it may not make the deadline or just abort it?
&gt; If I do abort it is there any way to guarantee getting a regular hadsum_4.13
&gt; workunit with a one year deadline? I currently have a estimated time of 4710
&gt; hours and the deadline is 1/28/2006. Any info would be appreciated.

Well Thyme Lawn stated on another board that computers with HD requirements too low for sulphur are now being given hadsm3 slab models. So possibly there are other limits in effect as well now, such as processor speed (just speculation). When did you download the new unit on the P3? You could try a project reset and see if it downloads another sulphur. If it does or does not, post back here. If it does, if I were you, I would suspend climateprediction.net and let it work on other projects. Running sulphur on a P3 800 MHz is not a good use of computer time on that generation computer.

I downloaded it about 5 hours ago. I can't figure out how it thinks my p3 800 can handle a workunit like that. So I should abort the workunit then?

&gt; I downloaded it about 5 hours ago. I can't figure out how it thinks my p3 800
&gt; can handle a workunit like that. So I should abort the workunit then?

What version of BOINC are you running?

You can do a project reset. It will download a new model. If it is sulphur, and if you are running BOINC 4.4x, then set the climateprediction.net project to get no new work, then abort the sulphur model. It should proceed on. There may be a way of forcing it to get a new slab model (hadsm 4.13) by playing with your disk space preferences, but I'm not sure what to set it to for force it to get a slab.

&gt; There may be a way of forcing it to get a new slab model (hadsm 4.13) by
&gt; playing with your disk space preferences, but I'm not sure what to set it to
&gt; for force it to get a slab.

Work out how much disk space is currently being used by BOINC, add 700MB and setting that value as your disk space use no more than preference should hopefully work.

If it doesn't download anything try increasing the value by increments of 100MB until you get a hadsm3 model.

You shouldn't download another sulphur model, but if that happens discard it as geophi suggested and try subtracting 100MB from the value until you're sent a hadsm3 model.
<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>

apparently as TL has discovered, the disk limit is the deciding factor, although supposedly the BOINC scheduler should take into account your PC speed etc, i.e. if it will exceed the 6 month limit etc (since by then we'll hopefully have the coupled model out!). perhaps I need to upgrade the BOINC server software, but it seems it's not being as "rigorous" on the CPU time calculation as we'd like it to be. Also the new BOINC clients scheduling seem a little odd with how they'll stop communication (and cancel trickles).

The trickle interval is the same as hadsm3, it's basically the 3-phase hadsm3 (same # of timesteps per phase i.e. 259,248) with two extra phases for the sulphur at "pre" and "post" industrial levels.

&gt; apparently as TL has discovered, the disk limit is the deciding factor,
&gt; although supposedly the BOINC scheduler should take into account your PC speed
&gt; etc, i.e. if it will exceed the 6 month limit etc (since by then we'll
&gt; hopefully have the coupled model out!). perhaps I need to upgrade the BOINC
&gt; server software, but it seems it's not being as "rigorous" on the CPU time
&gt; calculation as we'd like it to be. Also the new BOINC clients scheduling seem
&gt; a little odd with how they'll stop communication (and cancel trickles).
&gt;
&gt; The trickle interval is the same as hadsm3, it's basically the 3-phase hadsm3
&gt; (same # of timesteps per phase i.e. 259,248) with two extra phases for the
&gt; sulphur at "pre" and "post" industrial levels.
&gt;
&gt;

Carl
When the new models are out in Feb. 06 will we be able to keep working on Sulpher models still in process like we can the Slab models now or have to start the new ones?

Seems like letting people finish up the sulpher models would be of more use, but I am not the one going over the results so can not say for sure. My P4 2.4 is estimating another 101 days, so I have some time for other projects along with the sulpher model.
Cheers
Ray
Ray Brown
<A HREF="http://www.rays-place.com">Rays-Place.com</A>

<A HREF="http://boinc-doc.net/boinc-wiki/index.php"> BOINC WIKI</A>

<IMG SRC="http://www.boincsynergy.com/images/stats/comb-1842.jpg">