• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Formula-BOINC 2015

I will point out that currently Linux is faster at Universe@home as he is working on the optimizations.
 
Last edited:
I just threw 6 cores of my MacBook (i7-3720QM) at Wanless Mersenne. Should help a little there. Not sure I can leave this running 24x7, though. I may also have to cut it back to only 4 cores. This thing isn't really designed for crunching.
 
Yeah.. I have 1 core from a VM doing WEP-M+2 right now as well. Since they intentionally boycott Windows it isn't something I tend to run very often...
 
As a new team member, I'm not sure I understand everything, but I looked at the Formula-Boinc website and I think I'm fairly up to speed. I do have some questions tho.

1) Of the I projects listed in my intro message, where would I be able to do the most good for the team?
2) Would it be better for me to join a new project or two to help the team more?
3) This does last the entire year, correct?
4) Last one. Since I joined after the first of the year, I can still help, right?

Thanx!
 
1. Comology, Constellation, and MindModeling (the first two are also DC-Vault projects). MM doesn't always have work, so make sure to have another as a backup if you focus on it.
2. Most of our team will be hitting WCG, so if you want to push our position with FB, other projects would benefit more. The way to see where things are needed is to basically compare where we currently stand and the resources needed to get to the next place. You have to juggle things like project popularity and what hardware they support. So, clearly a GPU project isn't going to be much good if you have only CPU's to bring. That sort of thing. Projects like WEP-M+2 (wanless) isn't very popular so it is easier for us to get scoring position. Projects like QCN is tough because you have to have more devices that have capable sensors. So, OC'ing a system wont help. Lots of factors to consider. I also recommend bouncing a few just to keep things fresh.
3. Yes. This is a yearly challenge. So, each year it resets. This gives new teams a chance to shine whereas the DC-Vault is ongoing and never resets.
4. Yes. This is based on over all team production, so you can start and stop as many projects through the year as you like.
 
Last edited:
I have been trying for weeks (or at least it seems like it) to contact the Constellation server but cannot get a response. I've tried web, request for WUs, and ping. They all fail. I'd be happy to run some work from there if I could only contact them. Anyone else having similar problems?

I joined WEP-M+2 (wanless) only to be told by BOINC it doesn't support Windows 64-bit. Looks to me like that is something that might ought to be put on the home page of the project seeing as there are a few computers running that kind of OS out there.

I tried running VM on my i7 machine once and it didn't work well at all. I don't know what was wrong, but it seemed like something sure wasn't right with it.
 
Constellation has had server issues lately. So odds are you aren't dong anything wrong. Just hang in there. Wanless has an apps page like most projects and has also discussed not supporting windows in their forums. Yes it is only because he is trying to force people to support linux and mac os
 
I thought something was up with Constellation as I have completed several tasks from there. But most projects put out a notice of some kind if they can when they are having trouble. Guess they couldn't.

I figured out what you said about wanless. Even saw your posts on there about the same. I just think it could be made more visible I guess. Either that or I missed it.
 
We've now taken 3rd place in BURP which is the glass ceiling there unless someone with a lot of resources wants to knock it out of the ballpark. I'm gonna keep running it for a bit to try to build up some padding and then possibly switch to YAFU where we are close to taking 3rd place as well. Both of these projects are good for multi-socket systems with small amounts of RAM. Just don't expect a lot of points from YAFU as it uses CreditNew...
 
4. Yes. This is based on over all team production, so you can start and stop as many projects through the year as you like.
Click to expand...
How does Formula BOINC compile and monitor teams' production? Do they scan each BOINC project's site? Do they scan/cooperate with BOINCStats-BAM!?

Simply put, if I register my BOINC projects at BAM! or in the BOINC Manager with H teams, dies H get credit in BF? Also, we have more than one team name - HardOCP and [H]ard|OCP. Do both names count for BF?

I wanna play, tell me the rules. :cool:
 
They can pull stats from the projects like the stats sites can. Anyone can pull them actually. They know both teams because last year I poked and prodded (and slightly shamed) them until they made a fix to associate the correct team names. All you have to do is be attached to the project you want to push and be associated with our team. FB will take it from there. They basically look at how many points each team has at the beginning of the cycle and then just tally from there.

And just to be clear...by associating yourself with our team at BOINCStats (BAM), it will update your status at the project too. So, anyone pulling stats from the project will know you are associated with your team...
 
Tonight I kicked up my output for some of the "easy point" projects: dDM, Sztaki, Convector, Edges, Lattice. I have also been killing it over at Wanless (with only 6 cores of an i7-3720QM in my Macbook) and we've gained 9 spots there in the last week. At this rate, we will be in the top 10 there in the next week.

If everyone could give just a small work share to these projects (as low as 25), we could get in, and stay in, the top 10 of each all year. In some, we would probably be in the top 3 all year. Those are relatively easy points we are leaving on the table now. Please chip in just a little on these lesser-known projects - for the team. Position 5 is in reach. Beyond that would be almost impossible, unless we suddenly had a huge influx of members. I would be ecstatic with 5th. Let's do this!
 
Last edited:
BURP is another good one to devote a little time to... The work is sporadic, so it is a good project to share time other other projects.
 
"The team, the team, the team..." as spoken by the late, great Bo Schembechler
 
Just some insight... The Lattice Project can be tough to get work at times. The work units can be monsters to run both in run time and ram requirements. Make sure it is a box that can handle it. Convector sometimes goes dry as well.

I had good experience last year when I did almost a million points at dDM. EDGES and SZTAKI are good projects, but EDGES still has its issues too.

I may create a list tomorrow of projects that rarely have work and so should be on ones list in hopes to snag work when they appear.

And we are now 2nd for Yafu.
 
... or Gen George S. Patton, if you're not a Wolverines fan... :D
 
The WU's at Lattice for the last few months have been small - 1-2 hours, at most. The work has also been steady - I had a small work share set for it before tonight and would see work on my rigs fairly consistently since about the time they started sending these smaller WU's.

Update: I just checked my history and it appears they have been sending some 5-6 hour WU's lately. Those are the largest I've seen. All of the recent tasks from 1-6 hours used ~16MB of memory each.
 
Last edited:
Additionally, I haven't received any of those giant WU's at all since the same timeframe. I used to get those multi-day monsters previously.
 
Since we basically have 4 types of crunchers on the team, I will give a quick break down as it will probably be necessary to create at least two lists for FB.

1. DC'ers that only want to crunch one project to attempt to be the #1 team somewhere. Typically this has been FAH and is also kinda flowing over to WCG which is a much tougher place to play in.

2. DC'ers that want to push our standing at DC-Vault as that has been typically the large challenge we have supported in the past and also supports projects outside of BOINC. (FAH is one of them)

3. DC'ers whom want to push Formula-BOINC and thus have a wide selection of BOINC only projects.

4. DC'ers that just do their own thing.

For this purpose I'm mainly going to focus on 2 & 3 as they are really our target markets. The reason that the 2nd category matters is that those pushing our position at DC-Vault can help us with FB at the same time. The following projects are in both and thus any contribution helps us in both:
Asteroids
Collatz
Constellation
Cosmology
Einstein
Enigma
GPUGrid
Leiden Classic
MalariaControl
Milkyway
NFS
NumberFields
POEM
Primaboinca
PrimeGrid
Rosetta
SAT
SETI
SZTAKI
theSkyNet POGS
WCG
YoYo
You will find that many of these projects are also the more popular ones and thus will probably be harder to score in.

Now, for those wanting to just push FB, here is a list of projects that tend to have low work available, lots of down time, etc... And thus if you really want to make any dent there, you need to have rigs attached for when work is issued.

ABC
AlmereGrid (I recommend avoiding the project for the most part)
CAS
Climate (not ClimatePrediction)
CONVECTOR (it has been up and down lately)
DistrRTgen (pretty much a dead project)
IBERCIVIS
MindModeling (runs out often and doesn't always have work for all OS's)
Najmanovich Research Group
OPTIMA
Pirates
Renderfarm.fi (pretty much a dead project)
RNA World
SlinCA (server status page is not correct)
Spatiotemporal Quality of Service (This project is probably dead by now)
The Lattice Project (has had work lately, but who knows if it will continue consistently)
Volpex
 
Last edited:
Here are some other factors to weigh in on...

The following projects need virtualbox, JAVA, or Visual C++ Redistributable Packages for Visual Studio 2013. installed on the system:
Atlas - VB
RNA World - VB
vLHC - VB
distributed data mining - JAVA
Gerasim - Visual C++ Redistributable Packages for Visual Studio 2013 to do GPU work units

I only listed these because they have work available. There may be a few others, but unless they get work, they aren't worth breaking down.

The virtualbox projects tend to require a lot more memory and thus would be a tougher project to do. However, they would also probably be some easier targets for scoring due to the difficulties.
 
BURP has been running pretty steady for me for a while. The work units have also been solid. This project seems to score better points than Yafu too, but I have not done any in depth analysis to verify it.

Yafu works well with MT work horses like the FAH users are used to. However, the points at this project sucks as it uses the terrible Credit New point system. Last I checked Quad core systems were the sweet spot for points. Since this is a test bed project, I would not drop heavy hitter systems on it unless you are just terribly short on RAM. It would actually reward you better to break a multi socket system down into smaller VM's than to run all cores on these work units.

WEP-M+2 also known as Wanless is a Linux/Mac only project. The admin is on a crusade to convince more people to use Linux and thus refuses to publish a Windows app. The project is not very popular and would be easy scoring for those with Linux rigs or those willing to run a Linux VM.

Quake Catcher Network is a project whose days are numbered. However, until it actually is shut down, we need as many devices attached as possible. Anyone with an Android phone or tablet should at least run this project as it is non-CPU intensive. Some devices don't work, but many of them do. Since it is basically a number of devices game, OC'ing or high core counts don't compute here. Since it is non-CPU intensive this means those worried about heat on their device shouldn't be so concerned. Of course there is other supported hardware, but the bottom line is we need every device we can muster running it.

MindModeling constantly runs low on work. So it would be nice to have multiple people with this attached to grab work when it is there.

CPU's should probably be focused on CPU only projects. POEM for example runs low on GPU work, but when it has it, it blows the CPU points out of the water.
 
A couple of notes:

1) ABC, while technically still a live project, hasn't had work in many months. The admins were posting on their forum, but haven't done so in a few months now. I think this project is dead at this point.

2) DistrRTgen, while technically still a live project, hasn't had work in many months. The admins have been completely non-responsive. I think this project is dead at this point.

3) OPTIMA had a small amount of work back in Oct, but nothing since. It appears they only have small batches of work once or twice per year.
 
Would a project like MindModeling, which has work for Linux more often than other OS's, take advantage of the VirtualBox version of BOINC?
 
I think every project could take advantage of VirtualBox and have more reliable results. The problem is that your user base dwindles since most systems out there still have 4GB or less of memory.
 
I mean by me running it on my end.

Could I run more tasks by running VirtualBox on projects like MindModeling?

Also, could I run more projects?

Or do the projects have to be written to take advantage of VirtualBox?
 
The projects choose to use VB. So, there are a few projects that require it either for all work or for certain work units. Currently the following either have VB apps or require VB for all of their apps:
Atlas
Climate
GPUGrid (only when work is available for the app)
RNA World (only some apps but they seem to be the only ones with work anyways)
vLHC

There are a few more from LHC that aren't on BOINCStats, but that doesn't apply here. Also, you could make your own VM's in virtualbox and set up BOINC on them. This would mean the same box is requesting work more often and therefore could.
1. cache more work than the typical 10 days as the system will be essentially doubling up.
2. get more work than a single box if work units dry up quick because they will be asking more often.
3. could be set to a specific project and let run dry while the host BOINC client chews up regular projects. This would guarantee the client is hungry and asking for work more likely when work is available.
4. let's you bipass any daily limit by having essentially a second or third box filling the same quotas.
 
We are now tied for 8th place with Gridcoin even though they have us ranked at 9 now. We are also one point away from 7th place, so it is pretty close right now.
 
The were supposed to add Universe@home today, but it hasn't yet showed up. They also haven't made and firm open decisions on whether the Test project will stay. I did PM Phil about the situation since other BETA/Alpha projects aren't included. He forwarded it on, but as expected there is nothing announced in a timely manner.
 
We are currently in 8th place at Gerasim and when I'm done hitting my goal, we should be borderline 7th. Positions above that will take a much larger contribution.
Lattice Project hasn't had work in a while, so still waiting for that to come back.
I still plan on putting ~500k into Leiden Classical.
I only have about 7k left for my goal at LHC but they are without work ATM.
Constellation still has server issues and was out of work when I last checked. That project will be near impossible to do anything with.
I plan on hitting CSG one more time for testing but that will certainly put me over my personal goal.
I am also ~80k from my personal goal for SRBase
 
I'm currently working on my personal goals at WCG, Poem, and PrimeGrid. Planning to hit Gerasim, CSG, and Distributed Data Mining next.
 
Looks like we finished in 8th place. Not bad considering how large some of the teams we are competing with are.
 
You must log in or register to reply here.
Back
Top