Formula BOINC 2022 - Sprint admin change

This section is where you can post topics related to Challenges from other websites, such as Formula-BOINC, SETI-Germany and BOINCstats, as well as project challenges such as from WCG, PrimeGrid et al.

As a Team, we really want to engage as many team members with these challenges, as this gives us a great way of attracting more members to our Team...plus we get to show our strengths as a Team by (hopefully) doing well in each one.

PS: I have moved lots of topics into this section where they are relevant...as you can see, we've always been happy to take part in Challenges...and long may that continue ;-)
Post Reply
UBT - Timbo
UBT Forum Admin
Posts: 9673
Joined: Mon Mar 13, 2006 12:00 am
Location: NW Midlands
Contact:

Formula BOINC 2022 - Sprint admin change

Post by UBT - Timbo »

Hi all

Just a heads-up for you all.

The Formula BOINC Sprints this year have had a troublesome time, as some stats have not been correctly collected or displayed and this has caused some issues for FB members.

As such, Sebastien has now allowed me to take on an administration role within FB so I will in future be "managing" each of the Sprints.

This is not a difficult task as it boils down to randomly selecting a project for each Sprint period and then keeping my fingers crossed that the "back office" calculations and the display of stats works as it should.

The 11th through to 22nd Sprint projects have already been chosen at random by yours truly.

I guess the total amount of time required to do this administration, will be measured in minutes...as I only have to monitor the chosen project just a day or two before the official start time and then ensure that the project chosen is actually online and providing tasks to crunch - if it isn't I can change to the "back-up" randomly chosen project. ;-)

Simple really.

regards
Tim
UBT - PennyQ
Active UBT Contributor 15+ yrs
Posts: 351
Joined: Mon Apr 17, 2006 1:00 am
Location: South Bucks

Re: Formula BOINC 2022 - Sprint admin change

Post by UBT - PennyQ »

Tim,
I hope Sebastien has a ready supply of HobNobs for you! :lol: :lol: :lol:
Penny
The impossible we do today; miracles take a little longer

Image
stephenhunter
Active UBT Contributor
Posts: 125
Joined: Wed Feb 16, 2022 9:00 am
Location: North Yorkshire

Re: Formula BOINC 2022 - Sprint admin change

Post by stephenhunter »

:lol: :lol:
Image
UBT - Timbo
UBT Forum Admin
Posts: 9673
Joined: Mon Mar 13, 2006 12:00 am
Location: NW Midlands
Contact:

Re: Formula BOINC 2022 - Sprint admin change

Post by UBT - Timbo »

UBT - PennyQ wrote: Sun Jul 17, 2022 10:14 pm Tim,
I hope Sebastien has a ready supply of HobNobs for you! :lol: :lol: :lol:
Penny
Hi Penny

Indeed !!

I wonder if McVities sell HobNobs in France? (where Sebastien is based) - maybe they are called Le HobNobs over there?

Though with all the red tape on imports since Brexit, I wonder if they would pass through Customs without being confiscated by officials who are short of a bicky to dunk in their tea :-)

Mind you, I'd settle for a couple of bottles of Beaujolais Nouveau at the appropriate time (3rd Thursday of November) - in fact when I lived in Kent, I got free tickets for the newly opened Channel Tunnel and I did a day trip to France, found a nice emporium selling BN at a cheap price - bought a case of 12 bottles (and some cheeses, some French beer and a few other tasty morsels) and it was the tasiest BN I've ever had...happy days ;-)

regards
Tim
UBT - Timbo
UBT Forum Admin
Posts: 9673
Joined: Mon Mar 13, 2006 12:00 am
Location: NW Midlands
Contact:

Re: Formula BOINC 2022 - Sprint admin change

Post by UBT - Timbo »

Hi all

So, the last 2 Sprints have all taken place successfully and there's been no gripes or moans from any of the Formula BOINC community...so it looks like the Sprints will run properly (under my "control") until the end of the year.

So, if you fancy joining in the fun please visit this website page and you will find out which project is selected for each of the rest of this years Sprints.

https://formula-boinc.org/sprint_v2.py? ... en&sprint=

Note: The Sprints are on a summer break, (just like the F1 season) and will restart on 25th August at 21:00 UTC when the chosen project will also be announced.

regards
Tim
stephenhunter
Active UBT Contributor
Posts: 125
Joined: Wed Feb 16, 2022 9:00 am
Location: North Yorkshire

Re: Formula BOINC 2022 - Sprint admin change

Post by stephenhunter »

I'm trying with SRBase, but no tasks are downloading :cry:
Image
stephenhunter
Active UBT Contributor
Posts: 125
Joined: Wed Feb 16, 2022 9:00 am
Location: North Yorkshire

Re: Formula BOINC 2022 - Sprint admin change

Post by stephenhunter »

Got some now. Had to go onto the SRBase website and tick the right projects :shock:
Image
UBT - Timbo
UBT Forum Admin
Posts: 9673
Joined: Mon Mar 13, 2006 12:00 am
Location: NW Midlands
Contact:

Re: Formula BOINC 2022 - Sprint admin change

Post by UBT - Timbo »

Hi all

The Sprint season for FB 2022 has now ended and UBT colelcted a total of 6 FB points through out the 22 Sprint events.

And we've earned just 14 points (so far) on the Marathon - though this does continue until the end of the year...so we could earn some more FB points, if anyone is interested in making some extra effort? But I doubt it will propel us into the top half of the table.

Needless to say, this has not been a successful year on these Challenges, as UBT is currently languishing in 24th (out of 27) places.

see here:

https://formula-boinc.org/index_v2.py?y ... team=&thx=

The admin at FB, Sebastien, is currently working on developing the website to incorporate a few updates, which will hopefully generate more interest...so, I would hope that UBT can do better next year...but that will depend on each members financial situation as well as their preference for crunching specific projects.

I guess we will have to wait and see how much enthusiasm there is, to do better in 2023?

regards
Tim
djg2008
Active UBT Contributor
Posts: 27
Joined: Wed Sep 07, 2022 8:20 pm
Location: Essex

Re: Formula BOINC 2022 - Sprint admin change

Post by djg2008 »

I'm not really a sporting person, so understanding Formula One is not my area. I noticed one thing about these challenges. They seem to be all about the place. We need a unified approach to making challenges easy to find, all in one place, etc.

I stopped running BOINC back in 2004 when on SETI, I discovered my old account (albeit the CPID is not the same, so I can't seem to join the old data to my BAM data (even though it's not a lot and very old)). I only found out about this again this year, and I'd say given my situation - I have put a great deal of effort into running BOINC efficiently as I possibly can project wise - given the 18 year gap to make up for!

I also realise that I am not an advanced user, still a beginner really. Therefore I don't feel it is my place to throw suggestions, typically most people don't agree with my ideas anyway. I will keep an eye on challenges. I fear a lot of them are simply means to crunch certain projects/subprojects in rotation. I prefer to keep projects->subprojects in accordance to what I want to assist/uncover. Challenges takes me away from that to a degree - and specifically doing those last 2 CPU intensive challenges on PrimeGrid was not something I wanted to try and do (given I only have one quite good computer to do it on, the other is mostly incapable on many projects like LHC, Rosetta, PrimeGrid[CPU&GPU-Intensives]).

It would be better I think if there was an alternative to challenges, using team/country/world standing measurements. I place nicely at the moment in PrimeGrid, NumberFields and Combined in the world and UK and Team (this will change over time). Also discoveries should weigh in too. But sadly this is something BOINC would need to code in as far as I am aware. It would be nice to see some proper Android development too, allowing more tasks not just WUProp/Universe/TN-Grid.
Folding@Home
Image

BOINC
Image
UBT - Timbo
UBT Forum Admin
Posts: 9673
Joined: Mon Mar 13, 2006 12:00 am
Location: NW Midlands
Contact:

Re: Formula BOINC 2022 - Sprint admin change

Post by UBT - Timbo »

djg2008 wrote: Fri Nov 25, 2022 7:29 pm I'm not really a sporting person, so understanding Formula One is not my area. I noticed one thing about these challenges. They seem to be all about the place. We need a unified approach to making challenges easy to find, all in one place, etc.
Hi djg

As a long term cruncher (going back to the pre-BOINC days of SETI@home) I can hopefully give you some opinions on the subjects you have raised.

Firstly, most projects themselves, create their own challenges, so that teams can compete against each other, and usually in celebration of some specific project-based reason. And projects organise the results into specific webpages so you can see how you are doing individually, as well as on a team-basis. And the whole idea of this is to generate more data for the project to use.
djg2008 wrote: Fri Nov 25, 2022 7:29 pm I stopped running BOINC back in 2004 when on SETI, I discovered my old account (albeit the CPID is not the same, so I can't seem to join the old data to my BAM data (even though it's not a lot and very old)). I only found out about this again this year, and I'd say given my situation - I have put a great deal of effort into running BOINC efficiently as I possibly can project wise - given the 18 year gap to make up for!
The CPID is based on your email address...and hence as long as you use the *same* email address for each BOINC project you have joined, then all your stats will be collated together. Of course, SETI@home has now shut down and so it may be that they are not exporting any more stats data and hence aligning your old SETI account with your new CPID might not be possible. If you can log into your SETI account using your *original* email address and password, then you can change the email address on your account to your current email address, and your SETI stats might then be "picked up" by your BOINC Manager account and also with the stats dsplayed on various project websites (as all your credits are also collected by every project so you can log in on any BOINC project and see all of your stats).
djg2008 wrote: Fri Nov 25, 2022 7:29 pm I also realise that I am not an advanced user, still a beginner really. Therefore I don't feel it is my place to throw suggestions, typically most people don't agree with my ideas anyway. I will keep an eye on challenges. I fear a lot of them are simply means to crunch certain projects/subprojects in rotation. I prefer to keep projects->subprojects in accordance to what I want to assist/uncover. Challenges takes me away from that to a degree - and specifically doing those last 2 CPU intensive challenges on PrimeGrid was not something I wanted to try and do (given I only have one quite good computer to do it on, the other is mostly incapable on many projects like LHC, Rosetta, PrimeGrid[CPU&GPU-Intensives]).
The biggest issue in recent years is that project applications are now getting more "data intensive" and hence to run BOINC tasks on some projects relies on you having recent hardware (with newer CPUs and newer GPUs). This relies on people investing in their host machines and for some this is not cost effective, if their basic computer is good enough for every other task they want to do on it.

And the idea of Challenges (either organised by the projects or by other teams, such as SETI.Germany with their "Pentathlon" or L'AF with "Formula BOINC", or indeed BOINCstats with their own Challenges) is simply to give many projects a huge boost over a short period of time, which can help their research, whilst at the same time giving a competitive edge to the whole idea of crunching data and, of course, some "team bonding" when numerous team members are all crunching at the same time.
djg2008 wrote: Fri Nov 25, 2022 7:29 pm It would be better I think if there was an alternative to challenges, using team/country/world standing measurements. I place nicely at the moment in PrimeGrid, NumberFields and Combined in the world and UK and Team (this will change over time). Also discoveries should weigh in too. But sadly this is something BOINC would need to code in as far as I am aware. It would be nice to see some proper Android development too, allowing more tasks not just WUProp/Universe/TN-Grid.
BOINCstats does this to some degree, as it collects members, teams, and country stats and displays them accordngly.

https://classic.boincstats.com/en/stats ... tStatsInfo

Android development is down to the projects and if they can afford the cost of re-coding WIndows/Linux/Mac tasks so to create specific applications that will run on Android...and of course the Android OS has changed a lot in previous years so, even if an app was created it might have to continuously updated to reflect how Andoid OS works...and of course as mentioned above, some project data needs quite powerful CPUs and on some mobile devices using Android, they simply do not have the "horse power".

Hope the above helps

regards
Tim
djg2008
Active UBT Contributor
Posts: 27
Joined: Wed Sep 07, 2022 8:20 pm
Location: Essex

Re: Formula BOINC 2022 - Sprint admin change

Post by djg2008 »

All I can say on SETI classic is that maybe it uses an *ooold* Yahoo email; but the email address actually on the Seti page right now is my current email (and username). The cpid is different. I did change my screen name (this week I think) to djg2008 to make it inline with the rest of my Boinc-Manager accounts. Still the CPIDs are mismatched (the old one starts 'df36...' and my present accounts start '498c...'

I am learning to keep my eyes out for the challenges. I should be more free to participate - especially after Wieferich and Wall-Sun-Sun ends. I say that, but I am only getting WW to 100,000,000 for Double Bronze, then changing subprojects before the Numberfields one begins. I will boost my potential with The Science Cloud droplets during that challenge to make up for previous lack of input, also to push NFS@home beyond 2,000,000.

I guess my gripe with the Android thing was more or less related to the fact that if BOINC was widely available and supported on the OS, it is something a lot more people would contribute to overall. But yeah I do understand there are challenges involved. I think a lot of it comes down to awareness of Berkley's volunteer computing. If it was more popular, Android might receive more attention (and perhaps iOS for that matter). It would make more sense with computers able to be on such small chips these days - albeit not powerful enough ones.

I just feel like if everyone globally knew about the whole thing, it may engage more people to say 'ohh I'll give that a try, that sounds like a great use of computing'. The project is less simple than factory-setting-up a Samsung phone, and most people use phones/laptops/consoles to game rather than data crunch. PC-knowledgeable people were quick to jump on the Bitcoin-Mining bandwagon and more likely overlooked the Volunteer-Research possibilities; leaving us with the dedicated base of people who stuck with it/returned.

It can be a bit irritating, as the Bitcoin stuff just made computing more expensive - and the planet less green - whilst not really assisting science as impactfully as it could have. I guess the Blockchain is important - and actually I shouldn't moan - it seems Blockchain may contain a lot of criminal laundering, so I am glad I stayed away from the whole concept!
Folding@Home
Image

BOINC
Image
UBT - Timbo
UBT Forum Admin
Posts: 9673
Joined: Mon Mar 13, 2006 12:00 am
Location: NW Midlands
Contact:

Re: Formula BOINC 2022 - Sprint admin change

Post by UBT - Timbo »

djg2008 wrote: Sat Nov 26, 2022 1:54 pm All I can say on SETI classic is that maybe it uses an *ooold* Yahoo email; but the email address actually on the Seti page right now is my current email (and username). The cpid is different. I did change my screen name (this week I think) to djg2008 to make it inline with the rest of my Boinc-Manager accounts. Still the CPIDs are mismatched (the old one starts 'df36...' and my present accounts start '498c...'
Hi djg

There is a difference between "Classic SETI@home" which was a stand-alone application, that started as a screensaver and just crunched data one task at a time - and the later BOINC-based SETI@home - so, you cannot merge "stats" from the "stand-alone version", with the later BOINC version.

But of course if we're just talking about the BOINC version, then the CPID's should align, but it will take a few days to happen...and it relies on SETI "updating" their XML data export file, so that 3rd party stats websites (like BONCstats) can actually compile all the credits for each different CPID. The username doesn't really make any difference, except when searching the stats for your own project credits.
djg2008 wrote: Sat Nov 26, 2022 1:54 pm I am learning to keep my eyes out for the challenges. I should be more free to participate - especially after Wieferich and Wall-Sun-Sun ends. I say that, but I am only getting WW to 100,000,000 for Double Bronze, then changing subprojects before the Numberfields one begins. I will boost my potential with The Science Cloud droplets during that challenge to make up for previous lack of input, also to push NFS@home beyond 2,000,000.

I guess my gripe with the Android thing was more or less related to the fact that if BOINC was widely available and supported on the OS, it is something a lot more people would contribute to overall. But yeah I do understand there are challenges involved. I think a lot of it comes down to awareness of Berkley's volunteer computing. If it was more popular, Android might receive more attention (and perhaps iOS for that matter). It would make more sense with computers able to be on such small chips these days - albeit not powerful enough ones.

I just feel like if everyone globally knew about the whole thing, it may engage more people to say 'ohh I'll give that a try, that sounds like a great use of computing'. The project is less simple than factory-setting-up a Samsung phone, and most people use phones/laptops/consoles to game rather than data crunch. PC-knowledgeable people were quick to jump on the Bitcoin-Mining bandwagon and more likely overlooked the Volunteer-Research possibilities; leaving us with the dedicated base of people who stuck with it/returned.

It can be a bit irritating, as the Bitcoin stuff just made computing more expensive - and the planet less green - whilst not really assisting science as impactfully as it could have. I guess the Blockchain is important - and actually I shouldn't moan - it seems Blockchain may contain a lot of criminal laundering, so I am glad I stayed away from the whole concept!
Thanks for your feedback on this...at the end of the day, we are all volunteers and crunching any data is time consuming, and costly...in terms of both power and hardware. And I guess it might be more "offputting" for "newbies" to get involved, when they see that some members have millions of credits and the newbies give up after crunching a few tasks and find out they are only awarded maybe a few tens of credits...and they realise it will take them a very long time to get into the top 100,000, even the top 1,000,000 ranked places.

I think it is easiest to crunch data for those projects that a) you have a high level of personal interest in helping them and/or b) if you want to achieve specific milestones with various projects (and once obtained, you move onto another project).

regards
Tim
djg2008
Active UBT Contributor
Posts: 27
Joined: Wed Sep 07, 2022 8:20 pm
Location: Essex

Re: Formula BOINC 2022 - Sprint admin change

Post by djg2008 »

UBT - Timbo wrote: Sat Nov 26, 2022 5:08 pm Hi djg

There is a difference between "Classic SETI@home" which was a stand-alone application, that started as a screensaver and just crunched data one task at a time - and the later BOINC-based SETI@home - so, you cannot merge "stats" from the "stand-alone version", with the later BOINC version.

But of course if we're just talking about the BOINC version, then the CPID's should align, but it will take a few days to happen...and it relies on SETI "updating" their XML data export file, so that 3rd party stats websites (like BONCstats) can actually compile all the credits for each different CPID. The username doesn't really make any difference, except when searching the stats for your own project credits.
Ah right, and yes I have zero credits for new SETI. Apparently I only managed to rack 41 classic work units anyway, which seems pitiful really. Maybe I did more on the Folding@Home project on Playstation 3.

I guess I'm not overly concerned about the CPID aligning or not! It's hibernated for now anyway. :)
UBT - Timbo wrote: Sat Nov 26, 2022 5:08 pm Thanks for your feedback on this...at the end of the day, we are all volunteers and crunching any data is time consuming, and costly...in terms of both power and hardware. And I guess it might be more "offputting" for "newbies" to get involved, when they see that some members have millions of credits and the newbies give up after crunching a few tasks and find out they are only awarded maybe a few tens of credits...and they realise it will take them a very long time to get into the top 100,000, even the top 1,000,000 ranked places.

I think it is easiest to crunch data for those projects that a) you have a high level of personal interest in helping them and/or b) if you want to achieve specific milestones with various projects (and once obtained, you move onto another project).

regards
Tim
I understand what you mean about newbies being enticed but then realizing it's probably not worth the effort.

One is LHC@Home - lacking any generosity at all in credit gains, as are some other projects. Given how computers vary vastly, and some of these projects have Heavy system requirements - you'd think they would reconsider their work/credit calculations based on dynamic data and not static data. I am not really sure what the solution could be, but I think that WUprop could do something there with all the data it has gathered. Obviously I don't know the nuances of BOINC enough to be certain.

One idea could be like a 'WUprop exchange', where different applications are rated what their credit should award based on the system that submits the completed work unit. Also validation errors should award a little credit - it still had to be computed even if it failed!

GFN17-MEGA is a prime candidate for that - on my 660 graphics card. It seems to error out a Lot on this machine. Instead of keeping on, I instead decided to switch my GPU to a different subproject that won't error out/invalidate at all (GFN16 for example, or WW) on this rig. As far as software goes, BOINC is intermediate - and it takes a while to get there.

This is why I'd say BOINC has that feel of being an Alpha project that moved to Beta and has been in Beta for about 20 years, lol. When this changes, we should see more contributions! Unless of course quantum computing can remove the need for us to crunch if QC's can eventually do it all instead (much faster than our best rigs can now).
Folding@Home
Image

BOINC
Image
Post Reply