Page 1 of 1

G4 processor future, can we have a statement, please?

PostPosted: Mon Jan 28, 2013 10:03 am
by iBozz
Other threads and replies on this august and highly esteemed forum have indicated that G4 processing has ceased. Indeed, my two G4 Macs have been idle since December.

I am happy to leave those two Macs waiting for F@H work if there will be some, but if G4 processing is indeed dead then I can turn them to other distributed computing use.

I emailed Pande but received a standard reply suggesting that the answer to any query is sought on here. Fine, i can understand that the volume of emails must be staggeringly high and that this forum is the best place for information and help.

May I suggest, however, that those who control the Folding project let us know, with a definitive statement, whether G4 processing is indeed dead, please?

It is remarkably frustrating not to know whether to keep waiting or move on and donate my spare G4 capacity to some other project.

And if a statement has been made then may I apologise for overlooking it and ask that someone points me in the right direction?

Thanks in advance.

Re: G4 processor future, can we have a statement, please?

PostPosted: Mon Jan 28, 2013 1:06 pm
by Macaholic
iBozz wrote:And if a statement has been made then may I apologise for overlooking it and ask that someone points me in the right direction?

Thanks in advance.


From Dr. Pande's blog;

JANUARY 06, 2010
Update on Power PC OSX support

We have been going through various shortages for Power PC OSX WUs, due to the new cores not supporting PPC machines. While we expect that we may have new PPC WUs in the future, we want to give a heads up that we are depreciating PPC OSX support in Folding@home. We would love to support as many OS's as possible, but the overall compute power from the PPC group is small enough, and the efforts to support it large enough, that we have decided that it is better for us to put our resources into improving support for the other OS's we are supporting.

Thanks to all of you who have faithfully supported FAH with the processing that you have donated!

Re: G4 processor future, can we have a statement, please?

PostPosted: Mon Jan 28, 2013 1:21 pm
by iBozz
Thanks, Macaholic, I hadn't seen that, at least not recently. I do remember it, vaguely, now you draw my attention to it.

But that's three years ago and I have been Folding on G4s quite happily since then, with only one break - swiftly filled - until now.

If this is indeed the end of the G4 road, as seems likely, then a definitive and authoritative update to that Blog would be very much appreciated (preferably flagged on here as well, we don't all read Blogs) and, I'm sure, set at rest the minds of those of us currently in suspended animation!

Re: G4 processor future, can we have a statement, please?

PostPosted: Mon Jan 28, 2013 1:34 pm
by bruce
What you're probably saying is that during that 3 years, useful work was being performed by servers and G4s that were quite happily doing so. Also, Stanford didn't need to spend a lot of time/money on support because nothing important broke....but eventually something broke that wasn't worth fixing.

Re: G4 processor future, can we have a statement, please?

PostPosted: Mon Jan 28, 2013 4:54 pm
by Joe_H
In this case nothing had broken yet on the server for PPC OS X clients, assignment from the remaining server to OS X/PPC was turned off to resolve a conflict in providing work. Intel OS X systems can be set to fold with an uniprocessor slot that works fine with WU's that use the A4 processing core. However the assignment servers did not distinguish those A4 WS's from the WS with uniprocessor core 78 WU's available for PPC. Intel systems would go into a loop trying to download and use a PPC core that would not function on them. So what broke was the ability to only supply these PPC WU's to just the systems that could process them.

Re: G4 processor future, can we have a statement, please?

PostPosted: Mon Jan 28, 2013 5:26 pm
by iBozz
I'm still unsure!

Are G4 PPC machines now dead or merely in hibernation?

Will there be any more units for G4 Macs - yes or no?

Thanks.

Re: G4 processor future, can we have a statement, please?

PostPosted: Mon Jan 28, 2013 5:43 pm
by bruce
The Assignment Server code "broke" and it's both very complex and very critical to FAH. Apparently cost of resolving that "conflict" is exactly what he was talking about. Somebody is probably still trying to resolve issues like this one.

Re: G4 processor future, can we have a statement, please?

PostPosted: Tue Jan 29, 2013 12:55 pm
by iBozz
Thanks, still not a definitive answer!

The above, and linked posts, imply - or at least I infer therefrom - that this is a temporary, nevertheless major, issue that will be resolved in due course.

And again I infer that PPC G4 units will become available when the technical team have managed to resolve the issue.

Fine, if that's the case then I can quite happily leave my two G4s polling for units until eventually successful.

Thanks for the advice, but please let me know if I am making a complete idiot of myself (something at which I am a consummate expert, having had lots of practice!) and am totally misunderstanding things!

Re: G4 processor future, can we have a statement, please?

PostPosted: Tue Jan 29, 2013 11:20 pm
by mmonnin
Until a PG member answers I'm not sure you'll get a definitive answer, especially one with a time on when you'll get some new WUs. :(

Re: G4 processor future, can we have a statement, please?

PostPosted: Wed Jan 30, 2013 9:17 am
by iBozz
Thanks, mmonnin, it's an answer from PG that I am hoping to get! The advice from The Usual Suspects has, as ever, been excellent but a definitive statement would be very much appreciated.

Last time a shortage of Units was queried, over a year ago, Vijay himself came on and told us that the matter would be looked into and, within a very short while, we were off Folding again.

It's almost as if Vijay had some influence over the Project! :lol:

If someone from the Project could let us know then we could decide how to best use our still fully functional, albeit slow and old , G4s!


Re: Shortage of WUs for PPC (v5) clients {For Deprecated Cli
by VijayPande ยป Tue Oct 25, 2011 11:11 pm

We'll take a look and see what we can do here.
Prof. Vijay Pande, PhD
Departments of Chemistry, Structural Biology, and Computer Science
Chair, Biophysics
Director, Folding@home Distributed Computing Project
Stanford University

VijayPande
Pande Group Member

Posts: 2456
Joined: Fri Nov 30, 2007 7:25 am
Location: Stanford

Re: G4 processor future, can we have a statement, please?

PostPosted: Thu Jan 31, 2013 7:12 am
by Rattledagger
iBozz wrote:I am happy to leave those two Macs waiting for F@H work if there will be some, but if G4 processing is indeed dead then I can turn them to other distributed computing use.

I emailed Pande but received a standard reply suggesting that the answer to any query is sought on here. Fine, i can understand that the volume of emails must be staggeringly high and that this forum is the best place for information and help.

May I suggest, however, that those who control the Folding project let us know, with a definitive statement, whether G4 processing is indeed dead, please?

It is remarkably frustrating not to know whether to keep waiting or move on and donate my spare G4 capacity to some other project.

Not a definite answer on Mac-work from FAH, but it's easy to run a BOINC-project but still asking for FAH-work and crunch FAH-wu immediately if you get any. Since you needs an older BOINC-client to work on PPC, after installation you'll need to:
1: Make (or edit if already present) a file called cc_config.xml, located in your BOINC data-directory (exact location should be displayed on start of BOINC-client).
Use Notepad or similar.
2: cc_config.xml should include atleast:
Code: Select all
<cc_config>
   <options>
      <exclusive_app>The_Name_of_the_FAH-PPC-CORE.EXE</exclusive_app>
   </options>
</cc_config>

(Mac probably won't use an .exe-extension, but if where are any, remember to include it.)
Note, atleast with older BOINC-clients, the app-name is CaseSENsitive.

3: Save the cc_config. If you're running BOINC Manager (GUI), one choise in advanced view is to "read config file", choose this option.
If not using Manager, re-boot computer should also work.
4: Use a small cache, so you won't sit with much work in case FAH has any.

It's possible to include multiple <exclusive_app> and also <exclusive_gpu_app> (not applicable for Mac/PPC), just include multiple lines if desirable.

The effect of <exclusive_app> is, the moment a FAH-wu is downloaded and the FAH-core starts running, BOINC will pause crunching (if you've not selected to "run always" in BOINC), and continue BOINC after FAH-core has finished.