...
Yet, despite a big increase in # hosts which has exceeded 110.000, we're still at "only" 22 TeraFLOPS, roughly where we were in Jan-06 (and I don't know how these TFLOPS # is calculated, if it comes from BOINC benchmarks it could be highly overstated).

Some food for thought IMHO.

There is a very suspicious connection between the teraflops reflected for the project and the "Credits last 24 hours" number. It turns out that the project tersflops are actually credits divided by 100,000. It also looks as though the credits used are the awarded credits, not the claimed credits. So the number is way off of any real estimate of ACTUAL teraflops being applied to the the project by donated computers. It is reflective of the teraflops that are successfully used by the project (IE including all errors, failed reporting and user aborted WorkUnits). It would be a much more accurate number if the claimed credit was used instead of the awarded credits, but You can understand why it is done the way it is.

---

Moderator9
ROSETTA@home FAQ
Moderator Contact

I knowi am rather new here. But has anyone in the know ever thought of maybe contacting Sony on line entertainment? After all I play thier Everquest 2 game and i know for a fact they have about 100K users world wide if you could talk them into just posting a link on the signin Advertisment page you might get some bites. Even a 1% try out from SOE players would be 1000 to 2000 computers. Alos all these machines used by the players are 2.3 ghz or above cpus with 1 gig ram to make the game run clean enough . So they would have no problems with these WUs and would accomplish them in good time. Not sure if this is any help at all just a suggestion. After all SOE is always looking for tax write off. but again this would take someone that could explain what is done here.

Charlie

---

Well I guess sony won't do anything directly, as it's not something that they will make money out of. However, advertizing in game sounds like a good idea, especially if you're involved in a large guild.

Same goes for World of Warcraft, or any other online activity, by the way.

- trib'

---

OK, I've just posted the draft copy of the tell-a-friend about Rosetta letter. I am hopeful that this will help stimulate some ideas for the "brief lay language explanation of Roseta@home and its importance" that Dr. Baker requested in this thread. I'm thinking this needs to boil all the way down to a one paragraph explaination.

If you have comments or concerns about using this letter for the tell-a-friend function, please place them in the thread referenced above.

If you'd like to borrow from that content to make a message more tailored to the BBC crowd, please revise at-will and post your results here in the BBC thread.

Crunch more Rosetta!

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

Two days and not even a nibble of attempting a BBC message...

let me have a go at it... Keep in mind the target here is specifically this BBC audience. They've got a PC, they've got internet, and they probably are already participating in the BBC climate experiment. So they have some interest in science, computing, and volunteering their time to help science.

I tried to keep it very short and pointed. I think that's about all the room they're likely to give you off the BBC page. Hack it up, rewrite it, take another approach, but post some more ideas for what this message should say!
==== brief lay language explanation of Roseta@home and its importance ====
As with the BBC climate experiments, Rosetta@home relies on the help of thousands of computers like yours. However, rather than modeling weather, Dr. David Baker and his University of Washington team are developing models of proteins which will someday be used to protect us all from cancer, HIV and other deadly diseases. Participating in the project is informative, fun, and truly helps to advance the science of good health.

Here's the link for more information at the Rosetta@home website. Please consider joining the team today:
https://boinc.bakerlab.org/rosetta/moreinfo2.html
(this landing page would be where you would compare R@H to climate, and point out R@H runs on Mac, and uses less disk space, and have WUs where YOU control how long they run). The rest of the details and reference links would be very similar to the proposed landing page for the tell-a-friend function.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

---

Just trying to help take advantage of the link we've been offered. I'm not exactly clear where it will be. ...we can always go for more, once the message is defined.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

There really is no interaction between the BOINC CPDN and Either Rosetta or Ralph. In fact they may be more compatible in some ways than other projects...application swapping and other elements of the BOINC environment work well between the projects. The only issue is that the BBC CPDN is not using a standard BOINC package.

I've not done it, so it's a bit unclear. If BBC uses it's own BOINC installation, how do the two interoperate? I mean if I fire up two BOINCs do they just both run at low priority? Or do they both constantly think the PC is in use? If they both run, it would seem the combined memory footprint would be large enough to thrash even 1GB machines. Would the suggestion be to have one run from noon to midnight and the other run from midnight to noon?

Yes, I think a new FAQs thread for BBC migrants would be great. Try to keep it to <10 Q&As, and then link to the general R@H FAQs from there too.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

However, rather than modeling weather, Dr. David Baker and his University of Washington team are developing models of proteins which will someday be used to protect us all from cancer, HIV and other deadly diseases.

Maybe revise this to say...

However, rather than computing weather patterns to model climate, Dr. David Baker and his University of Washington team are computing atomic structures to model proteins. These protein models will someday be used to protect us all from cancer, HIV and other deadly diseases.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

There really is no interaction between the BOINC CPDN and Either Rosetta or Ralph. In fact they may be more compatible in some ways than other projects...application swapping and other elements of the BOINC environment work well between the projects. The only issue is that the BBC CPDN is not using a standard BOINC package.

I've not done it, so it's a bit unclear. If BBC uses it's own BOINC installation, how do the two interoperate? I mean if I fire up two BOINCs do they just both run at low priority? Or do they both constantly think the PC is in use? If they both run, it would seem the combined memory footprint would be large enough to thrash even 1GB machines. Would the suggestion be to have one run from noon to midnight and the other run from midnight to noon?

Yes, I think a new FAQs thread for BBC migrants would be great. Try to keep it to <10 Q&As, and then link to the general R@H FAQs from there too.

I summerised it above https://boinc.bakerlab.org/rosetta/forum_thread.php?id=1396#13903

Basically if you have BOINC (normal) installed it is not a problem, but then they probably already know that.
If they use the customised BBC-CCE boinc client they need to uninstall it and install a normal one. That's if I remeber rightly. There is a way to convert over to the normal boinc client without loosing current model progress but it's a bit more involved. I think you may just install boinc on top of the bbc-boinc and hope for the best. would need to check up on that.

But there's no problem with them both running togther other than memory usage and each task will take twice as long to finish :)

---

Team mauisun.org

Thanks for all your input and help.

Here is a recent message sent to us in response to some information we gave them about R@h.

"The BBC producer told me she is concerned that people not decide to participate in Rosetta@home over the Climate Change project. Can you please send me a bit of language re who (or rather what computers) canÕt do the BBC project but could do Rosetta@home that BBC could use on their website?"


From the input below, it seems like there isn't much of a reason from a technical standpoint why one might choose R@h over BBC (i.e. if you can run BBC you can run R@h and visa versa). However for the following reasons one may want to choose R@h.

1. The science. One may want to help health related issues rather than climate.

2. The length of work units. One may want to finish work units sooner. Users can set the run time for R@h work units from 1-24hours, as opposed to longer run times (quoted from the BBC site "A fast machine (on all the time) will complete the experiment in 3-4 months. A slow machine could take 6-9 months or even longer.").

3. User's may want scientific results and feedback sooner. R@h posts results of top predictions and will participate in CASP7 which is a community wide experiment held every two years with the goal to assess progress in protein structure prediction. The experiment will start soon and run through this summer.

any other reasons?

Can someone volunteer to help make the landing page? Feet1st's is a great start http://www.geocities.com/feet1st/TellAFriendMore.html. We need to address BBC's question above.

Thanks!

R@H uses considerably less disk space.
The BOINC Wiki shows BBC Climate using about 600MB, and R@H using around 191MB. My R@H presently shows 35MB, with 5 WUs on deck, but that doesn't really include any space for BOINC itself.

Rosetta runs on Mac, climate doesn't. Both support Linux and Windows.

With much debate about R@H current, and future memory reqs, I believe that both projects state 512MB recommended. So, I think we call it a draw on memory.

Also with much debate, Climate requires broadband. Rosetta CAN be done on dial-up. But may take more than an hour to get everything downloaded to get started. Doable, but you want to know before you begin the endeavor.

I'll need help! to make a BBC landing page. I need help with the following:

I need feedback on the landing page we created for the tell-a-friend function, and ways it should be changed to work for the BBC interested click-throughs. So the premise in this case is that someone's just read a very brief paragraph about R@H (similar to tell-a-friend), and they want more info.
Review the current tell-a-friend landing page here:
http://www.geocities.com/feet1st/TellAFriendMore.html
and, if possible, with your feedback that "It needs details on...", if you can provide the details, or links that already have details, that will be very, very helpful.

Mod9 mentioned some specific FAQs that could be gathered together?

Does anyone know where to find links on the various ways to install the BBC app and how to tell which installation method you've presently got? Am I correct that I can run it though an "actual BOINC", through a "stripped down BOINC", and as a "standalone"?

Does anyone know where to find info. on how to convert the BBC BOINC installation into a standard one, without detaching from the BBC project?

There was mention that BBC killed all the outstanding WUs and had to build them all again. Is that true? Please reference some doc on the subject.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

Thanks for all your input and help.

Here is a recent message sent to us in response to some information we gave them about R@h.

"The BBC producer told me she is concerned that people not decide to participate in Rosetta@home over the Climate Change project. Can you please send me a bit of language re who (or rather what computers) cannot do the BBC project but could do Rosetta@home that BBC could use on their website?"


From the input below, it seems like there isn't much of a reason from a technical standpoint why one might choose R@h over BBC (i.e. if you can run BBC you can run R@h and visa versa). However for the following reasons one may want to choose R@h.

1. The science. One may want to help health related issues rather than climate.

2. The length of work units. One may want to finish work units sooner. Users can set the run time for R@h work units from 1-24hours, as opposed to longer run times (quoted from the BBC site "A fast machine (on all the time) will complete the experiment in 3-4 months. A slow machine could take 6-9 months or even longer.").

3. User's may want scientific results and feedback sooner. R@h posts results of top predictions and will participate in CASP7 which is a community wide experiment held every two years with the goal to assess progress in protein structure prediction. The experiment will start soon and run through this summer.

any other reasons?

Can someone volunteer to help make the landing page? Feet1st's is a great start http://www.geocities.com/feet1st/TellAFriendMore.html. We need to address BBC's question above.

Thanks!

David,

CPDN no longer supports Mac AT ALL. They unceremoniously killed the running Mac models, and when people went to get new models, there was simply a message that there is no model for the Mac platform any longer.

As luck would have it the Mac is a very stable Rosetta Machine. A big part of this message should be devoted to pulling at least that audience to Rosetta. A lot of the CPDN Macs were high end dual and quad processors.

The New models, are very memory intensive, and require high speed processors. Many people attach to CPDN and after a few hundred hours of running (and failed models) they begin to realize their machine is not progressing fast enough to complete even a single model on time. Even with a dedicated fast machine with plenty of memory a CPDN model can take 3 month to complete.

There are a LOT of machines that do not fit the CPDN project that would do very well here.

---

Moderator9
ROSETTA@home FAQ
Moderator Contact

---

Best UFO Resources
Wikipedia R@h
How-To: Join Distributed Computing projects that benefit humanity

Can someone volunteer to help make the landing page? Feet1st's is a great start http://www.geocities.com/feet1st/TellAFriendMore.html. We need to address BBC's question above.

David, perhaps I misunderstood. Are you asking for the project comparisions to by placed on a landing page? Or are you asking for two things, project comparisons for BBC to put on their page, and a BBC-oriented landing page at Rosetta?

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

Thanks for all your input and help.

Here is a recent message sent to us in response to some information we gave them about R@h.

"The BBC producer told me she is concerned that people not decide to participate in Rosetta@home over the Climate Change project. Can you please send me a bit of language re who (or rather what computers) cannot do the BBC project but could do Rosetta@home that BBC could use on their website?"


From the input below, it seems like there isn't much of a reason from a technical standpoint why one might choose R@h over BBC (i.e. if you can run BBC you can run R@h and visa versa). However for the following reasons one may want to choose R@h.

1. The science. One may want to help health related issues rather than climate.

2. The length of work units. One may want to finish work units sooner. Users can set the run time for R@h work units from 1-24hours, as opposed to longer run times (quoted from the BBC site "A fast machine (on all the time) will complete the experiment in 3-4 months. A slow machine could take 6-9 months or even longer.").

3. User's may want scientific results and feedback sooner. R@h posts results of top predictions and will participate in CASP7 which is a community wide experiment held every two years with the goal to assess progress in protein structure prediction. The experiment will start soon and run through this summer.

any other reasons?

Can someone volunteer to help make the landing page? Feet1st's is a great start http://www.geocities.com/feet1st/TellAFriendMore.html. We need to address BBC's question above.

Thanks!

David,

CPDN no longer supports Mac AT ALL. They unceremoniously killed the running Mac models, and when people went to get new models, there was simply a message that there is no model for the Mac platform any longer.

As luck would have it the Mac is a very stable Rosetta Machine. A big part of this message should be devoted to pulling at least that audience to Rosetta. A lot of the CPDN Macs were high end dual and quad processors.

The New models, are very memory intensive, and require high speed processors. Many people attach to CPDN and after a few hundred hours of running (and failed models) they begin to realize their machine is not progressing fast enough to complete even a single model on time. Even with a dedicated fast machine with plenty of memory a CPDN model can take 3 month to complete.

There are a LOT of machines that do not fit the CPDN project that would do very well here.

I agree Mac OS-X client PC's will be you main catch on this (though I'm not sure you'll get the ex-CPDN people as we are talking from BBC-CCE. People involved directly with CPDN probably have looked elsewhere already. Remember only very late, well after registering and if you have any problem (even then a person still may not) get put throuh to CPDN main project.
We are talking directly BBC-CCE people and not the larger CPDN, though word will get around.

I think if you target MacOS-X people (even better if you can get an Intel client, see Ralph suggestions) you then get aadvertisement and curiosity generated)

---

Team mauisun.org

Not only the Macs, what about people who want to have thier computers run say 8 hours a day?
They're not going to complete the BBC-CPDN run in time for its deadline... but we can set 8hr runs here.

---

Not only the Macs, what about people who want to have thier computers run say 8 hours a day?
They're not going to complete the BBC-CPDN run in time for its deadline... but we can set 8hr runs here.

Actual completion is not so necessary for BBC-CCE since any partial task is also good to them. Though I do understand what your saying.

The MacOS-X target does not impede on their own target audience, anything else does, even people who will not finish a full task there may still start a task and give them some scintific information. So I say, be smart target very honestly what they cannot support :-)

---

Team mauisun.org

---

Moderator9
ROSETTA@home FAQ
Moderator Contact

Fully agree -ish :-)

The laptop's exact statment applies here too. I think it is a standard BOINC site safety clause (we have a similar one here at Rosetta.

You're right on the contradiction with the 'partial part', though I think they're more on about literally just downloading and seeing what it looks like so nothing is reported and no data is sent back. As opposed to 20% or so of the experiment getting sent back. I don't know enough.

Also Model Crashing is by design (They say so, but that maybe on the CPDN's site?)

Rosetta@Home also do not support older Win9x/ME OS's. It says so at the same place it mentions the laptops.
https://boinc.bakerlab.org/rosetta/rah_requirements.php

EDIT/ Though over at CPDN it states there are physically problems running under BOINC-CPDN http://climateapps2.oucs.ox.ac.uk/cpdnboinc/forum_thread.php?id=3678&PHPSESSID=5f997f3e92ddef555606015f3d0856f1 I don't think we have any reported problems ?


But all in all, the BBC are doing us a favour, so be nice back.

Pitch the MacOS-PPC's as we all agree (or seem to)

You could pitch the slower spec PC's as well 500MHz to 1500MHz since our spec page says they're ok to use :-)

Either way we need to get on with it :-)

---

Team mauisun.org