Message boards :
Number crunching :
Succesful completion & upload... :-)
Message board moderation
Author | Message |
---|---|
Send message Joined: 5 Aug 04 Posts: 186 Credit: 1,612,182 RAC: 0 |
http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=296 Uploaded automaticaly without a bobble - interesting to see that BOINC didn't get another model until only a few hours before this one completed. Just over 19 days to complete the model - slightly longer than under Classic CPDN, but not a huge amount. <a href="http://www.nmvs.dsl.pipex.com/"><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=6&team=off&trans=off"></a> |
Send message Joined: 5 Aug 04 Posts: 30 Credit: 39,745 RAC: 0 |
Thanks Nick and great to see; every trickle and every graph and more importantly confirmation of a successful upload before the Public Launch date. <img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=11"> <i> UK4CP @ www.uk4cp.co.uk (United Kingdom Group) Celeron 2.6GHz XP Pro SP1 768MB RAM<i> |
Send message Joined: 7 Aug 04 Posts: 187 Credit: 44,163 RAC: 0 |
I'm looking forward to one as well in 12 hours. :-) <a><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=18"></a> |
Send message Joined: 10 Aug 04 Posts: 94 Credit: 309,849 RAC: 0 |
Question: Does the cpdn app upload any of the work data at the completion of a phase or at the completion of the entire work unit? <img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=35&team=off"><a href="http://mysite.wanadoo-members.co.uk/thefinalfrontear/index.html">Team Link<a> |
Send message Joined: 5 Aug 04 Posts: 30 Credit: 422,225 RAC: 0 |
After phase 1 I could see the graphs of temperature und rain on the website, so I think the client uploads data at the end of each phase. |
Send message Joined: 5 Aug 04 Posts: 21 Credit: 9,084,238 RAC: 0 |
> Uploaded automaticaly without a bobble Nick, Good to see your result uploaded ok. I'm still on phase 1, so you've blazed past me for that WU (:-)) Happy crunchin', Martin |
Send message Joined: 7 Aug 04 Posts: 2168 Credit: 64,543,482 RAC: 6,686 |
> Question: Does the cpdn app upload any of the work data at the completion of a > phase or at the completion of the entire work unit? > > If it's like the classic client, it uploads a relatively small subset of data at the end of phases 1 and 2 (at least enough to create phase temp/precip graphs), then a large amount of data at the end of the run. |
Send message Joined: 5 Aug 04 Posts: 907 Credit: 299,864 RAC: 0 |
cool uk_nick, a nice stable run too. I guess people finishing soon may want to wait and get the launch version which should fix some of the quirks of the alpha & beta test. |
Send message Joined: 7 Aug 04 Posts: 2168 Credit: 64,543,482 RAC: 6,686 |
> cool uk_nick, a nice stable run too. I guess people finishing soon may want > to wait and get the launch version which should fix some of the quirks of the > alpha & beta test. > > Carl, Is there going to be a list of fixes in the official version? Thanks. George |
Send message Joined: 5 Aug 04 Posts: 907 Credit: 299,864 RAC: 0 |
> Carl, Is there going to be a list of fixes in the official version? Thanks. I'm not sure I have enough time to write all that! ;-) But seriously, here is what has been fixed off of the top of my head: <ol> <li>better suspend/resume behaviour with BOINC, and you can now unload the app & memory on a "suspend" based on your web-page prefs <li>graphics have been tidied up a bit, e.g. cloud & snow layers, and you can "peel" off layers in case you want to view the different layers in detail <li>graphics should perform better (i.e. lower CPU, esp on "field" scenes such as temp/precip/pressure, cloud/show scenes are still a bit cpu heavy by their nature) <li>should behave better on restarts <li>the "rewind a model-day/month/year" (depending on how far you've made it) to recalculate should work again <li>added more diagnostic fields output that get uploaded, so CPDN/BOINC HadSM3 runs provide much more information than the original Windows-only/CPDN runs -- such as regional means for about 100 fields, convective cloud leve fields, rel. humidity, etc. </OL> |
Send message Joined: 5 Aug 04 Posts: 390 Credit: 2,475,242 RAC: 0 |
> I'm not sure I have enough time to write all that! ;-) > But seriously, here is what has been fixed off of the top of my head: Thank for quick list, Carl. And congrats to Nick; i wish my model haven't crashed (i should have been finished about 2 days ago). Nice and clear graphs - i really like them. <img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=56&trans=off"> |
Send message Joined: 5 Aug 04 Posts: 21 Credit: 9,084,238 RAC: 0 |
[...] > But seriously, here is what has been fixed off of the top of my head: [...] Carl, Do you want the present Beta runs to continue with their present models, or should we all reload with the updated 'official' release? (Are the Beta model results useful?) Aside: Aswel as the global temperature charts, can charts be plotted for global min/max temps, and similarly for global min/ave/max windspeed and pressure? Regards, Martin |
Send message Joined: 7 Aug 04 Posts: 187 Credit: 44,163 RAC: 0 |
Mine is completed as well. <a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=721">Result.</a> A couple odd things though... Is credit granted twice? It appears as though I was given the incremental credits as trickles were reported, then upon completion, given them again? Is this what was intended? Not that I'm complaining, but was wondering if this was supposed to happen. My accounts page says I have over 12,000 credits now, (with over 1,000 RAC) with the completion of this single WU. Also, my FX-53 keeps going way down on Carls CPU reference list. There are apparently 3, FX-53s running the project now, and one of them seems to appear near the top where it would be expected, then will go way down the list for a few days. From looking at the benchmarks, I can see this is my machine. It is currently down there now, listing an Avg_sec_per_TS of 8.468, even though it hit 1.77s the whole run. (with a few 1.76s) It also shows an estimated completion of 76.23 days, although the run was completed in 16. Any idea what is going on with this? <a><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=18"></a> |
Send message Joined: 7 Aug 04 Posts: 2168 Credit: 64,543,482 RAC: 6,686 |
> Mine is completed as well. > Also, my FX-53 keeps going way down on Carls CPU reference list. There are > apparently 3, FX-53s running the project now, and one of them seems to appear > near the top where it would be expected, then will go way down the list for a > few days. From looking at the benchmarks, I can see this is my machine. > > It is currently down there now, listing an Avg_sec_per_TS of 8.468, even > though it hit 1.77s the whole run. (with a few 1.76s) It also shows an > estimated completion of 76.23 days, although the run was completed in 16. Any > idea what is going on with this? > This is what happened to me when it restarted the same model/work unit. Make sure the name of your work unit has changed to a new model. |
Send message Joined: 7 Aug 04 Posts: 187 Credit: 44,163 RAC: 0 |
> This is what happened to me when it restarted the same model/work unit. Make > sure the name of your work unit has changed to a new model. Thanks, geophi. This behavior occured when processing the previous model. The current model is indeed different than the previous. I was wondering if it might have something to do with the frequency the trickles were reported. Since I am currently attached to 4 other BOINC projects that aren't CC 4.xx capable, I disable BOINC access for much of the time in order to quit pestering the other projects with a client they don't support. I enable network access every so often to submit trickles, so there are often more than one at a time. I guess maybe only Carl can say whether that is the case or not? <a><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=18"></a> |
Send message Joined: 5 Aug 04 Posts: 907 Credit: 299,864 RAC: 0 |
Hi heffed, credits should be fixed now. I overlooked that BOINC will basically add credit again at the end of a run, but it should be fine now. I'm not sure about the cpu.html page, it was pretty basic, but I see that AMD FX53's have really moved up? I'll probably "pretty up" and double-check the cpu page when I get a chance, it was obviously something I threw together quickly for novelty value. |
Send message Joined: 7 Aug 04 Posts: 187 Credit: 44,163 RAC: 0 |
> Hi heffed, credits should be fixed now. I overlooked that BOINC will > basically add credit again at the end of a run, but it should be fine now. > I'm not sure about the cpu.html page, it was pretty basic, but I see that AMD > FX53's have really moved up? I'll probably "pretty up" and double-check the > cpu page when I get a chance, it was obviously something I threw together > quickly for novelty value. Thanks Carl. :) |
Send message Joined: 5 Aug 04 Posts: 186 Credit: 1,612,182 RAC: 0 |
It was fixed at one point Carl, but it seems to be screwed up again..? http://climateapps2.oucs.ox.ac.uk/cpdnboinc/top_hosts.php http://climateapps2.oucs.ox.ac.uk/cpdnboinc/show_host_detail.php?hostid=140 <a href="http://www.nmvs.dsl.pipex.com/"><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=6&team=off&trans=off"></a> |
Send message Joined: 7 Aug 04 Posts: 187 Credit: 44,163 RAC: 0 |
> It was fixed at one point Carl, but it seems to be screwed up again..? I see the same thing Nick. My main account page has the proper credit, but the host detail list still shows inflated credit. <a><img src="http://boinc.mundayweb.com/cpdn/stats.php?userID=18"></a> |
©2024 climateprediction.net