They've also taken the no.1 spot on NFS from us.
Formula Boinc 2016
- Thread starter hanluc
- Start date
If at all possible, run only the 16e work units, they pay the better due to their memory requirements, which is just under 1 gig per thread.
Last edited:
Exactly that, no need to abort anything if you don't want to, http://escatter11.fullerton.edu/nfs/prefs.php?subset=project edit preferences and select only the lasievef & lasieve5f, it was my understanding that lasievef was for windows and lasieve5f was linux but i believe both platforms run each of them just fine, maybe Carlos could explain.
If you have just enough memory make sure you change the default boinc client memory usage from 50% in the options/comp preferences/memory setting.
Last edited:
http://stats.free-dc.org/stats.php?page=team&proj=nfs&team=1116
Oh nice production, I wasn't expecting that
andy & Oz too joining in 
dDM servers have crapped out again, can't get new work or send the ones I've completed.
Thanks
Did you see Bowmore from TSBTI thought Scole was thier big hitter.
Little digging... Bowmore is Bryan from seti.usa he swapped teams for a week and dumped 4m credits on NFS for TSBT.
The main reason they did so well recently on NFS was occasionally some quick work units are dished out which complete several times quicker than the normal ones but give the same amount of credit and they managed to catch them, though they do both have quite a few cores at their disposal ...but let's not big them up too much as I'm sure they're reading this, don't want it to go to their heads
Boinc Pentathlon should start at the beginning of May. Only two competitions pending on Boincstats, GPU Grid & WCG but we can make our own if there is a project you (or anyone else) had in mind?
We are back to 1st on NFS
You've hit 1m already
those quick works units are currently being sent out, they are around x4 quicker than the ones previous, it's worth increasing your cache in the boinc client to grab as many as you can, maximum seems to be 1046 WUs per machine.TSBT are filling their boots too.
I'm going to have me a few of those bad boys
Not quite as fast on Ubuntu as Windows is.
It also looks like 100 * number of cores is the limit, so not sure what sort of machine you are running to get 1046 !
I have a 10.46 core machine
I have not noticed any difference between Linux and Windows, it's hard to tell when the work unit runtimes vary so much themselves. It used to be the case that Linux was a good 30% quicker a while back.
This seems right, my 8 threaded machines are getting 800 each, but the limit does seem to be a 1000 and something, don't think it's exactly 1046.
Unlike you I had a valid reason, I hid them for my stealth attack for top spot on NFS, I didn't want the opposition to see what I was up to, not that they were taking any notice
You can see my linux machine, it's currently ranked the top host on NFS
It's an iniquitous situation, they get to read everything we discuss, we get read nothing they do.
I have an idea which will net us a few points and there isn't a thing TSBT can do in any way to stop us
If we all jumped on NFS they wouldn't be able to stop us anyway, unless more SUSA muscle came in.
You can set you memory requirements in the boinc client to something like 80% you'll run some tasks but others will say "waiting for memory" or you could change you CPU usage limits to run a limited percentage of cores you have.
Got your email and replied. They have hidden groups on their site, one of them being the FB team, I'm registered there and could ask to join but I've a feeling they'll say no
Chris, are your machines running the multi-threaded 'camb_boinc2docker' work units on cosmology? if so are your CPUs being utilised 100% in the task manager?
We're on course for losing the top spot on numberfields, our portfolio of no.1 spots is rapidly shrinking.
sent you an email hanluc when you get chance to reply
streetlight is having his new cable laid on Tuesday to his garage ,so im sure he'll get a few projects back to number 1 with the amount of systems he has lol
He's done a lot of work on numberfields but for Gridcoin, great if he wants to join us but if mining is his thing then fair enough.
sent you an email hanluc when you get chance to reply
Replied, all this secrecy. I have sent 3 messages via trust to 3 dons/mods past week about whether a created 'social group' could be made private, the content being only visible to its members, not one of them have replied
You can always go and join their forum with a funny name and do a bit of work. They don't look up your kilt to check if you're scottish, trust me
Brilliant idea, I'll sign up as Chuck MacDonald, drop a few points for them, in on the FB discussion and report back here!
Thanks Cadmunkey, I would say Asteroids@home as we've slipped a lot of points from there recently or Citizen Science grid, we haven't scored many points on that.
I don't think there's any point now, it would just slightly delay the inevitable, better use elsewhere I think.
With regards to keeping things private, how about text encryption? I'm not sure if it would even be permitted here though.
https://encipher.it/download
password: hello (as a test)
https://encipher.it/download
password: hello (as a test)
-----BEGIN PGP MESSAGE-----
Version: GnuPG v2.0.19 (MingW32)
jA0ECQMCe2nCKIt7/6Wu0mABxNzXfz/IS6Ib2axjcbrDB+QHhvkL4oloHqqujJO1
UpY8qfqidSDtMwp50v0ntVGiaslkaOSB0kT34aHcOQf+Wma0v7LomKjKMXOWnEQG
ciKuXuKFqeQsHwCgXbm6OeY=
=xPZj
-----END PGP MESSAGE-----
Version: GnuPG v2.0.19 (MingW32)
jA0ECQMCe2nCKIt7/6Wu0mABxNzXfz/IS6Ib2axjcbrDB+QHhvkL4oloHqqujJO1
UpY8qfqidSDtMwp50v0ntVGiaslkaOSB0kT34aHcOQf+Wma0v7LomKjKMXOWnEQG
ciKuXuKFqeQsHwCgXbm6OeY=
=xPZj
-----END PGP MESSAGE-----
... taunting us with our own encryption method, oh the cheek of it!
Climate prediction has been axed from the stats, dropping us 25pts. I can only guess it's to do with the scoring as FreeDC is showing all of today's scores as minuses. Don't if it's a temporary or permanent exclusion, no news on the FB site. Scores between 1st & 2nd now very close.