FormulaBOINC

I was actually able to release anything I had yesterday afternoon using Teamviewer on my phone via a dodgy cell connection. No fun at all, but it worked, eventually. One Haswell quad is not responding to connection requests, though, and nobody will be able to get to it until Monday morning, too late to be of any help.

 

Luckily the old Westmeres seem to chew these WUs up pretty well, looks like they might be making a significant contribution after all.

@Markfw , I got rid of the L5639s and found a deal on some X5670s, gave the ol X8DTL a nice boost!

 

Decent change in output (divide these numbers by 7 for BOINC points).

Mark was missed during the PG race (4-15 and 4-16).

 

I thought it had X5660's ?

 

The X5660s were in the SR-2, I have those clocked at 3.4. I guess the Supermicro with the L5639s was one of the bonus items, and it's doing really well with the X5670s in it!

 

Even better than the last pic.

 

I knew it! The kilts came off now.



(Covering my eyes and looking the other way.)

 

The 1st String has entered the fray!

 

WHY OCuK, WHY? What has been seen, canNOT be unseen!




Fellas, this race started in the upper 60s F (daytime). It's 91F / 32.8C and climbing today. Expect diminished output from me. Nearly as 'warm' tomorrow too.

 

Almost 2 full days in. We look to be in decent shape at the moment, but don't forget OCuK's comeback in the last Sprint. I have seen at least 3 decent bunkers get released from them, but I'm betting there is a MOAB out there still, from their side. So no slacking!

Now that several of them had to release early, to get more tasks no doubt, here's a little snippet of info I've been trying to keep track of as time permits. It shows current score (1900UTC update), and the points produced for the last ten hours. Not enough of a gain by OCuK to make a difference these last 26 hours, which is why I'm extra certain they have 2-4 bunkers that haven't reported in yet. Big ones. Combined, it's going to be another Mother Of All Bunkers!

1275k---204 Anandtech
1111k---194 OCN
984k---- 229 OCuK

 

I had to turn off my 32 core Opteron box. Its only does 5k ppd, but it puts out TONS of heat, and its getting warm here too ! Next wed will be 75. Right now with all the windows open, and 62f outside, its 80f in the house !

 

Here at my place in central Europe, a cloudless night has broken, and a minimum of -2°C / 28°F has been forecast. Max had been 12°C / 54°F today.

Needless to say, all of my machines are already running; no more cores to add to the race here.

 

91F outside today, having to shut down. Just came home from work, inside the house was 85F. Most of my systems running 85C-100C. Looks like I will have to take the summer off. Can't have the electricity usage up once I start irrigating the lawn. Don't want a $600 utility bill.

 

One of the first BOINC Pentathlon projects will be WCG OpenZika. So if we concentrate on that, a few pending WUs may spill over to four days from now.

 

OpenZika performs validation almost only on WUs submitted to "new" machines. Once a machine becomes trusted by WCG/OpenZika, quorum and replication is set to one in most subsequent WUs. "Veteran" machines only occasionally get OpenZika WUs that require verification or validation.

 

Right now, OcUK passed OCN. Each of them have 70 % as many points as we have. But this means little; we know these guys show a tendency to upload their work late.

*) Edit: Our own user xii5ku has picked up that bad habit of theirs and deferred 250 k until yesterday. When I meet him, I shall have a word with him about sportsmanship.

 

We're just trying to stay ahead of OCN. Third is looking good for you guys and deservedly so.

TBH We just want this sprint to be over so we can get on with the job of bunkering for the pentathlon.

Looking at last years pentathlon stats I notice we were one place above you in the overall table, hopefully we can have another good tussle again this year

 

WCG tells me that no work is available for OpenZika, Smash Childhood Cancer, and Mapping Cancer Markers.

 

A 50 k points firecracker from UK went poof.

 

Oh crap. 50% of my machines are now idle, the other 50% will run out in 3 hours or so. Let's hope OcUK has the same problem.

 

I'm at Outsmart Ebola Together for now. Nice small tasks, quorum=1 mostly. Only the one machine which had not been working WCG before got a boatload of quorum=2 OET tasks.

 

All of mine still have work. Is this over in 4 hours ?

 

yep, 4 hours