I normally don't bother looking at BOINC, it does it's joyful thing in the background & every so often I look to see how far along climateprediction is going since it takes a month to complete a WU.

But I digress; I was used to seeing times of usually an hour or less to complete a WU. Now I'm seeing times of 7-8 hours.

To wit: 3/2/2006 11:20:23 PM|rosetta@home|Computation for result PRODUCTION_ABINITIO_RANDOMPARALLELANTIPARALLEL_1cc8A_310_307_2 finished

took 7h 57m to complete and indeed it looks like all the results for Rosetta are taking a very long time.

Used to be the searching would have wild gyrations for a very short while & then settle into a gentle rocking situation. Now the wild gyrations seen in the searching & accepted fields are quite prolonged and the accepted energy cross point is doing a fantastic amount of searching.

As the BOINC program is the newest for Win XP, nothing has changed but the WUs themselves.

My computer is quite a bit faster than most and I imagine it would take far longer with most desktops. Something appears to have changed and this is curious.

Gary

EDIT: I should have added that this seems to be the way with all of the latest Rosetta WUs.

---

Hi Dr. Gary

Read this thread for more info.

https://boinc.bakerlab.org/rosetta/forum_thread.php?id=669#10375

Anders n

---

Well if you did bother looking at BOINC you would see why. The default CPU time for WUs was set to 8 hours some time ago if you read the messages....if you want to change it then you will have to change your prefs....

---

Interesting.

I've lived (and worked) under the "Inform before you perform" method.

I run BOINC and the programs it supports as a courtesy to the scientists looking for new discoveries, it is not a formost aspect to my life and as the computer is often running 24/7 I do not see Boinc as I hide icons that otherwise take up space when their program is running.

If the software is changed or the methodology has changed and it is up to me to make a modification, an email to me would have alerted me. If one spammer working from his living room can send out hundreds of thousands of emails several times a day, one bulk emailing informing users of the change would not have been hard to do.

Thank you for your replies.

Gary

---

If the software is changed or the methodology has changed and it is up to me to make a modification, an email to me would have alerted me.

You don't need to do anything. The default value will work just fine.

If one spammer working from his living room can send out hundreds of thousands of emails several times a day, one bulk emailing informing users of the change would not have been hard to do.

That sounds like a good way to get on all the lists of spammers that the ISPs use.

If one spammer working from his living room can send out hundreds of thousands of emails several times a day, one bulk emailing informing users of the change would not have been hard to do.

That sounds like a good way to get on all the lists of spammers that the ISPs use.
[/quote]

I disagree.

Rosetta already has the email addresses of it's volunteers. Sending directly to them would not be adding to a spam list. My daily Kim Komando is not listed as spam nor are any of the listservers I receive email from.

---

...and if you DID spend time on the Boards of the projects you support you would see that whenever a DC project DOES send out a bulk e-mail to its members literally hundreds of complaint flood in say just that about being spammed. DC projects can NEVER win, they are damned if they do and damned if they don't!

If one spammer working from his living room can send out hundreds of thousands of emails several times a day, one bulk emailing informing users of the change would not have been hard to do.

First, your original post was dated March 3rd. If you checked the news on the front page, you would see that a default runtime was initiated to alleviate several bugs in the software (Most notably - stuck workunits, 1% bug.. etc), and free up communication time on the network.

As the rate of work has varied at times from between 1 hour to 14 hours on even VERY fast systems, the change to force work to a stable run time was a key component to several bugfixes. Since the runtime was at times more and at times less than 8 hours, a public notification would not be, and is not nessacery.

The fact that you are noticing this distinct change is indicative of our success in fixing several bugs within the software, and as indicated WHEN YOU JOINED the project, applications will change without notice, as this is one of the underlying features of the BOINC system. Such changes to the science application can and WILL affect the runtime of work. For a further explanation, you can consult the BOINC-WIKI, as I will not extrapolate further on key features of the BOINC core program itself.

Rosetta is not the only project that has varied the runtime of it's workunits. LHC workunit runtimes have changed significantly several times in the past. Even the SETI program is about to complete a major change to it's application which will increase computation runtime from ~1 hour to about 30+ hours. Nobody sent out a mass email from LHC, and nobody will be sending out a mass email from the SETI department either. This is how BOINC works. BOINC will intelligently decrease the number of workunits that you download from each project to compensate for the increased runtime. Your computer will always compute according to your resource share specifications. As such - The amount of work your BOINCMGR requests will be altered as soon as it realizes that the runtime for the rosetta project averages a different length. This means that the impact on user's work committance is completely negligible.

I hope that this explains some portions of BOINC that you are obviously unfamiliar with. So that you are not caught off guard by other functions of the BOINC infrastructure, I suggest that you read through the BOINC-WIKI, or the next time you attach to the project - read the about section.

If the software is changed or the methodology has changed and it is up to me to make a modification, an email to me would have alerted me.

If this is how your needs are arranged then there is no reason to have BOINC on your system. Period. This is not how BOINC is designed, nor will it ever be. The project managers will have a need to change the science applications at a moments notice to make bugfixes, changes to the scientific code, switching algorythims, goals, etc... - Not to mention, this action is embossed in the about section on every project I have ever attached to. If you have not taken the time to read the about section on a website you are downloading software from, of which you are allowing to run on your computer 24/7 without your intervention.... I fear for you.

---

If the software is changed or the methodology has changed and it is up to me to make a modification, an email to me would have alerted me. If one spammer working from his living room can send out hundreds of thousands of emails several times a day, one bulk emailing informing users of the change would not have been hard to do.

not everyone wants an email, if you want to stay up to date with what's going on, get an RSS reader and subscribe to the various project news feeds out there

---

Want to search the BOINC Wiki, BOINCstats, or various BOINC forums from within firefox? Try the BOINC related Firefox Search Plugins