Change in BA requirements

Do you know the difference between a good and great leader. A good leader will learn his limitations and focus on what he excels at. A great leader understands his limitations and learns to overcome them.
Obvisouly, VJ has way more on his plate then he can handle. I don't understand why he doesn't delegate management to an actual manager. I have nothing against grad students, having been one myself. However, they have their dissertation to worry about. I know they hired a folding realtions person, but that's someone who's just going to tell us the FaH ship isn't sinking.
I know FaH is aircraft carrier that takes miles to turn. In this case, a great leader would delegate management to a competent person. Am I crazy or are we witnessing hubris in action?
 
One last thought for you BA folders...perhaps someone saw an answer to this that I missed:

I guess the BA folders still have 11 months to make a choice on how they want to allocate their equipment.

I have to say, someone asked in the thread rather directly to DR. KASSON...
since the project has been finished (paper written) are the current WU just adding additional background data, or are they looking at something new?
Click to expand...

I didn't see any answer to this question.
If I was paying electricity to run BA now...I would want that question answered.
Adding granularity and confirming data is great...but if they are just maintaining the work units for PR...then I would be looking to move on more quickly.

JMHO
 
Umm.

I know they hired a folding realtions person
Click to expand...

Pardon my curiosity but what is said person's name and where can announcement be found?
 
I knew they were shutting down BA. I didn't know they weren't any new BA projects. Is this correct? If so, I'm setting all my slots to finish and going to BOINC tomorrow.
 
I saw the question asked...I did not see an answer.
I am not saying that means anything...not getting an answer is not unusual.

I would ask in the thread again...but snap... it is closed.
And I am banned, so there is that too. ;)
 
Emmanon said:
I knew they were shutting down BA. I didn't know they weren't any new BA projects. Is this correct? If so, I'm setting all my slots to finish and going to BOINC tomorrow.
Click to expand...

Welcome to the Commandos. Feel free to stop in the BOINC or DC-Vault threads. :D
 
I'll take that as a yes. I'll have to suffer through a few days of folding withdrawl.
 
Don't take what I said as implying anything FAH....I have pretty well avoided FAH up to this point...
 
Nope...I was a BOINC'er that was looking into possibly running FAH right before the BA change announcement. :D
 
Commandos is a BOINC team? Does it have a team number? Was it founded by former folders?
 
I guess the BA folders still have 11 months to make a choice on how they want to allocate their equipment.
Click to expand...
That is what I am doing, posting the BA end in my mind and educating myself about options. As it stands now, two things need to occur if I am still to be with F@H after next January: 1) Pande Group must proactively engage their donors better, to be demonstrated by best effort roadmaps and transparency, and 2) there needs to be a systematic, logical points system, backed up by clear explanations of points = value of science accomplished. Yeah, I know, nothing new that we haven't already stated in the last few weeks - just my personal conditions for staying with the project.

In the meantime, I'll be looking at other projects, ready to hit the ground running, should it come to that.
 
Linden said:
Yeah, I know, nothing new that we haven't already stated in the last few years - just my personal conditions for staying with the project.
Click to expand...

I fixed that for you. Really, this shit has been going on for more than a decade. It's way past time to move on.
 
Emmanon said:
You're not a folder who left for BOINC?
Click to expand...
Nope, Gilthanis wasn't, but I am - was here for the first 4 billion [H] F@H milestones. I took off about 4 years ago after a combination of a previous FF situation like this and the ugliness of a Chimp Challenge (2009, I think?), then got drawn back for the fun of the Overtake Default challenge. But I just wound down my F@H late last week and moved my GTX680 back onto BOINC.
 
It's hard to quit now. I'm 7th place in the top 10 producers; soon to be 6th. I'll hit the hundred million mark in a week or so. There aren't a lot of Boinc projects that use GPUs and x86 cores. Most are math projects that are good science. My mom is a cancer survivor, so I lean towards medical projects. Did I overlook a medical gpu and x86 project? I'm running Linux, but I'd go back to Windows if needed:(
 
POEM will have GPU again hopefully within the next month. It typically has both GPU (both AMD and NVidia) and CPU. If you have NVidia cards, you can run GPU only at GPUGrid. Those are the only two BIO/Med projects with GPU capabilities for now. However, the glorious part of BOINC is that you can mix and match projects under the same client. So, if you have NVidia GPU's right now, you can run GPUGrid and POEM both to utilize your processors and video cards.

Now here is where it gets even more beautiful. If you have AMD cards, you could still run them at FAH and then keep you CPU's busy on a BOINC project. In BOINC, there are a lot more BIO options for the CPU.

Most of our FAH brethren seem to prefer WCG. (They are CPU only right now and probably for a long while) Some have also supported SIMAP, Docking, and Rosetta. These aren't all of the bio/medical projects, but some of the more well known ones. They too are only CPU.

This list gives a pretty good description of most of the projects. http://hardforum.com/showthread.php?t=1801908
 
Last edited:
I've been running MCM lately on WCG on a few CPU boxes; my father (not a cancer survivor, unfortunately) received treatment at the hospital running the project many years ago.

http://www.worldcommunitygrid.org/research/mcm1/overview.do

don't know about the requirements; I'm running Win7/8 64 and Ubuntu 13.10; boinc installation was painless; check out Gilthanis' handy guides and go to town!

Louis
 
Louis_Wu said:
I've been running MCM lately on WCG on a few CPU boxes; my father (not a cancer survivor, unfortunately) received treatment at the hospital running the project many years ago.

http://www.worldcommunitygrid.org/research/mcm1/overview.do

don't know about the requirements; I'm running Win7/8 64 and Ubuntu 13.10; boinc installation was painless; check out Gilthanis' handy guides and go to town!

Louis
Click to expand...

System requirements for WCG are found here: http://www.worldcommunitygrid.org/help/viewTopic.do?shortName=minimumreq
 
mdk777 said:
One last thought for you BA folders...perhaps someone saw an answer to this that I missed:

I guess the BA folders still have 11 months to make a choice on how they want to allocate their equipment.

I have to say, someone asked in the thread rather directly to DR. KASSON...

I didn't see any answer to this question.
If I was paying electricity to run BA now...I would want that question answered.
Adding granularity and confirming data is great...but if they are just maintaining the work units for PR...then I would be looking to move on more quickly.

JMHO
Click to expand...

mdk777, can you provide a link to the post this question was asked to kasson ? I did a text search in the 47 pages of the FF's Change in BA requirements thread, and could not find any occurence regarding a BA paper already written ?

it's actually pretty important to me since my plan was to fold BA until May 1st, 2014 then re-assess what to do with my 2P... if the influenza paper is already written and submitted to peer review I guess I could have a better use of my electricity
 
IIRC a paper has been written on the 6900/6901 projects but I don't think a paper has been written on any of the others
 
Nathan_P said:
IIRC a paper has been written on the 6900/6901 projects but I don't think a paper has been written on any of the others
Click to expand...

yes, I found references to this older one when I was checking the output of BA. After reading mdk777's post, I was wondering whether I missed some information regarding a 810x paper already submitted...
 
Thanks. I didn't notice that WCG supported GPUs. I'll probably stick with FaH a little longer. I want to hit another milestone or two. Then I going to the dark side, at least they have decent management. WCG has good management right? I don't want to jump ship to sign up with another cluster fuck.
 
WCG supports GPU's but don't have any sub projects that use them as of this time. They had one for HCC that finished about 2.5 years worth of work in about 6 months. The project scientists were the same as the current MCM work being done. WCG techs have stated that there isn't any known GPU sub projects in the pipeline in the foreseeable future. They are ready when the scientists are ready.

The WCG staff are pretty good. I will warn that their policy is to not give dates to anything because it just causes more problem within the donors. So, it is always a wait and see in regards to most improvements or launches and such. The Community Advisors or CA's are pretty active, but a few are starting to get a little too comfortable with their Wit and Banter. Sekerob is very knowledgeable and is a wealth of information, but sometimes fails to be polite when trying to be witty.

There is also a very active community at WCG. So, along with that you get the good and the bad. I will forwarn to just ignore astrolabe. if he ever pipes up. He is kind of a dick but every project has at least one.

WCG also has work for the ARM processor under Android. Currently the only work available is Fight Aids At Home, but when SN2S comes back up it too will be an option.
 
Crap, I need a Bio project for my 10 GPUs. I've invested a lot of moeny in my them:( I suppose I can setup 2 quad SLI rigs to watch porn in 4k;)
 
Emmanon said:
Crap, I need a Bio project for my 10 GPUs. I've invested a lot of moeny in my them:( I suppose I can setup 2 quad SLI rigs to watch porn in 4k;)
Click to expand...
GPUGrid is Bio and it seems to always have work and pays quite well ;) so if other projects do not have work you can always run it, then when the other projects have work you can switch back and forth if you wish. That is one of the pluses to boinc projects vs F@H is choice. http://www.gpugrid.net/

As far as smp or bigadv ending neither is going to end bigadv is just leaving the donor support world. Hopefully the new points system will have enough of an incentive to keep the MP rigs running on smp. I believe they will but who knows. PG has some work cut out for it at this time in the donor support and PR end of things and hopefully they get it worked out soon. I see the FF itself is currently getting very little traffic and I see there daily ppd productions dropping so maybe that will change some attitudes towards donors.

Any way I am currently siting on the fence watching and currently running more boinc than F@H, I am still a believer in the F@H program, I just do not believe in the current treatment / management of donors at PG. There current attitude is that some will leave and some will stay there will always be others to replace those that leave. (which is correct) but with a attitude and management change towards donors and the way the program is set up F@H could be so much more. :(
 
Though I have some very few projects done on BOINC to test basic setup I don't want give up hope on F@H yet.
Vijay recently spend more effort on communication using the RSS which I see as one kind of evidence that he understood and action to mitigate/quickfix some of our concerns. Some of the news are even just as heads up like actual work on OS X GPU support with Apple. But that's good even to read those chunks of info.
I still hope that Vijavy reach out to HR and get one dedicted (part time) communication specialist/management support. In the long run that's what he REALLY need and to get time back for himself and for the scientific work. Knowing that hiring can be a lengthy process I hope in few weeks we get feedback/result on progress.

As per FF: really quite over there recently ... Battlefield comes in mind or maybe just super bowl weekend.

Anyway: the purpose of the project is great and good to kick some cancer cell's butt.

Not give up on the hope that some of you high performer remain active and also represent the project in a positive way.

I really feel bad & sad on how all that tuned out :mad:
 
some of us will still be here folding, the rest will crunch away. Either way science should get done.

And a year is a long time
 
OK, I'm going to stay until BA shuts down.
Compared to regular x86 WUs, do you get more points for folding Beta WUs?
 
Emmanon, you should certainly find GPUGrid to be compatible with those cards. If you are interested in BOINC, I would connect to GPUGrid and any other CPU based project that you like. You could support them both at the same time.

My suggestions for BOINC beginners with the options you mention:
GPUGrid for GPU's (duh...lol)
POEM for CPU and soon GPU again
WCG CPU
SIMAP CPU
Rosetta CPU
Docking CPU
WUProp CPU (this project is non-CPU intensive and thus resides mostly in memory) It can run alongside all other work units and quite honestly there is no reason not to run it. Their website is also really helpful for planning hardware performance at different projects and their specific apps.
 
Emmanon said:
OK, I'm going to stay until BA shuts down.
Compared to regular x86 WUs, do you get more points for folding Beta WUs?
Click to expand...

There is no point bonus for folding beta WUs, be they CPU or GPU. Never has been, and I doubt there ever will be. Usually it's the opposite, where projects just entering beta have been poorly benchmarked and earn low PPD until enough beta testers say that it's low compared to what they're used to seeing on their hardware. I think this is a function of the benchmark hardware being rather old. AFAIK the benchmark CPU is still an i5 750 and the benchmark GPU is a GTX 460, which can lead to point scaling issues on certain projects. VJ did mention in his end-of-BA post that they may update the benchmark hardware this year, which I think is overdue.

The one thing you may have heard contrary to the previous paragraph is that Kepler and Southern Islands cards perform much better with the beta flag. This is because a weighting quirk of the assignment servers has been causing the beta flag to get nothing but core 17 WUs almost uninterrupted for the last several months, and on new, especially high-end, GPUs, core 17 flat-out smokes the older cores 15 and 16 in terms of PPD. Core 17 was released to full FAH a couple of months ago, and for a while now the only core 17 project being sent out to everyone has been out of beta as well. There was briefly a new core 17 project that was beta-only, but various issues have caused it to not pop up again.
 
mdk777, can you provide a link to the post this question was asked to kasson ? I did a text search in the 47 pages of the FF's Change in BA requirements thread, and could not find any occurence regarding a BA paper already written ?

it's actually pretty important to me since my plan was to fold BA until May 1st, 2014 then re-assess what to do with my 2P... if the influenza paper is already written and submitted to peer review I guess I could have a better use of my electricity
Click to expand...

Sorry, I looked again and I can't find it either.

It may have been in a different thread...there was a previous thread on the number of remaining BA WU available on the servers before the announcement of increasing the core count requirement, so it might have been there.

I thought it was a good question, and it stuck in the back of my mind during the subsequent debate...

I'm really not trying to lead anyone astray here, someone might have a better understanding of the projects than I (as a previous poster indicated here)
 
It looks like I'm going to have to split my GPUs and CPUs between two projects.
 
Found the way that thread was closed really offensive. It's like a part-time rarely-seen and unliked barkeeper walking up to a group of the regulars and telling them to change the subject.
 
You must log in or register to reply here.
Back
Top