DC Vault 2

phoenicis said:
Clearly my google fu isn't what it used to be. I haven't tried it but all that looks much more straightforward. Just a couple of things though, sudo can be omitted with HP Cloud as you login as root and I did try the apt-get install ia32-libs libstdc++6 freeglut3 approach but it didn't solve the GFAM problem. I dunno, maybe I fluffed the installation.
Click to expand...

Hmmmm... Interesting. Never tried running that in project in BOINC in Ubuntu yet. This wiki I found about a year ago so BOINC has progressed quite a bit since then so I guess there could be problems with that particular project. I got it to work with Cosmology@Home without a hitch though.

phoenicis said:
Speaking for myself, I reckon the gap to 3rd is too big for just the three of us so I'm going to let some rigs run for a few days in aid of buffer building but the majority of cores are already moving on to pastures new. Feels odd looking at HFM for bigadv tpfs again.
Click to expand...

Sounds good. I was planning on doing pretty much the same, and maybe shifting even more to SoB. Feel free to join me if you would like:D
 
Thanks to MJ12 for the "boost" in NFS. :) Its always nice to have some more processing power on our side!
 
Nice 30T score for the day in SOB mmmmmdonuts !
I hope to be back to help in a few weeks from now.
 
Those running Cosmology@home should check their boxes and probably choose a different project for a while. The results server is refusing uploads and reports ... Error reported by file upload server: Server is out of disk space.
 
Looks like the Cosmology drought has brought the hammer down on ABC. :D Well done guys! I also heard a rumor that dmaine wasnt the only one putting up points in Muon1 today. ;)
 
Oh joy, I get to be passed by Phoenicis in another project! ;)

The help is certainly welcomed, though. Haha
 
Briliu said:
Oh joy, I get to be passed by Phoenicis in another project! ;)

The help is certainly welcomed, though. Haha
Click to expand...

I've seemed to notice this trend around here as well. So far I am managing on holding him off in SoB, but my 1 day head start in ABC was no match for him.
 
I have him beat on average until tomorrow thought. He's putting some huge numbers on Muon though
 
Collateral damage is an unfortunate yet sometimes necessary consequence of our offensive operations. Your sacrifices shall always be honoured in these [H]allowed halls.
Sorry, I've been spending too much time working on RM Commando bases lately. I’ve somehow managed to avoid saying ‘hey, I’m a Commando too’ for fear of them smashing my face in.

Moving swiftly to the positive, the team seems to be filling out it’s ranks nicely with us now carrying out concerted efforts on several fronts. Nice going guys!


Having been spotted dipping my toe into the Seventeen or Bust water, I’d appreciate thoughts from the SoB crunchers as to whether I’ve got it set up correctly.

I’m using v26.6 of the executable running on a 980X Vista box OCed to 3.8 GHz. Since it’s also running GPUGrid, there are only 5 workers running across 10 threads (CPUs to use is set to 2) with about 75% total CPU utilisation. Daytime/Nighttime stage 2 memory is set to 8192MB which is two thirds of the total available. I’m getting iterations of about 0.03 sec. which, from reading earlier posts from donuts/Razor, doesn’t sound too great for multithreading but I’m really not sure.

I haven’t had time to look into the scoring system but I seem to have appeared within the stats without having completed a task. Points being awarded against interim checkpoints?
 
Last edited:
Phoenics, I love you. Worry not :D Every time you pass me it makes me want to spend money! Come march we should be getting new workstations at my work. HOPEFULLY some SB hex cores. I can really put the afterburners on over the weekends at that point! ;D
 
phoenicis said:
Having been spotted dipping my toe into the Seventeen or Bust water, I’d appreciate thoughts from the SoB crunchers as to whether I’ve got it set up correctly.

I’m using v26.6 of the executable running on a 980X Vista box OCed to 3.8 GHz. Since it’s also running GPUGrid, there are only 5 workers running across 10 threads (CPUs to use is set to 2) with about 75% total CPU utilisation. Daytime/Nighttime stage 2 memory is set to 8192MB which is two thirds of the total available. I’m getting iterations of about 0.03 sec. which, from reading earlier posts from donuts/Razor, doesn’t sound too great for multithreading but I’m really not sure.

I haven’t had time to look into the scoring system but I seem to have appeared within the stats without having completed a task. Points being awarded against interim checkpoints?
Click to expand...

Great to have you in SoB.

Your time per iteration looks good. My E5649 @ 2.53 GHz is running around 0.038 - 0.042 seconds per iteration and my old i7 920 at 3.3 GHz was at 0.032 seconds. (These were both v26.6) so your 980x looks like it is around where it should be.

As far as the memory allocation goes is I only allocate 1GB but thats because I also tend to run BOINC on the same machine as well, or other really RAM intensive programs. I honestly have never noticed a difference between the default of 8 MB and a larger value. Supposedly it only matters for a few hours on a LL/P-1 test.

The Prime 95 client by default updates once a day unless you communicate with it manually or complete a WU.

And yes the points are awarded on an checkpoint basis. You do not actually have to complete the WU to keep the points, which is kind of unfortunate unless they some how mark in the message they are sending to the server where you are mathematically in case the WU doesn't get completed. This would prevent redundant work for the partial completions for these very long WU's. (I am not sure about the last part though) Your score looks reasonable based on my i7 920, which was getting 4.0-4.4T a day if I remember correctly.

So yes it looks like you set it up correctly:)
 
Last edited:
phoenicis - Like it or not your a Commando now. Don't think for a second that your not. :)

So i got a BOINC notice today that GPUGrid now supports AMD GPUs. Interested i went and looked at the thread: http://www.gpugrid.net/forum_thread.php?id=2839

Unless I am completely losing it, it seems that GPUGrid is using AMD GPUs under a separate project called 'donateathome' to mine Bitcoins as a donation to the project. Kind of a neat idea but yet it seems a little strange at the same time. Thoughts? Good idea?
 
metallicafan said:
phoenicis - Like it or not your a Commando now. Don't think for a second that your not. :)

So i got a BOINC notice today that GPUGrid now supports AMD GPUs. Interested i went and looked at the thread: http://www.gpugrid.net/forum_thread.php?id=2839

Unless I am completely losing it, it seems that GPUGrid is using AMD GPUs under a separate project called 'donateathome' to mine Bitcoins as a donation to the project. Kind of a neat idea but yet it seems a little strange at the same time. Thoughts? Good idea?
Click to expand...
I think donating our hardware time to GPUGrid for the science is great but donating our computer time just to raise Bitcoins for there project is not why I crunch.
 
Briliu said:
Phoenics, I love you. Worry not :D Every time you pass me it makes me want to spend money! Come march we should be getting new workstations at my work. HOPEFULLY some SB hex cores. I can really put the afterburners on over the weekends at that point! ;D
Click to expand...

And acquiring more hardware is the appropriately [H]ard response to being mowed. Look forward to seeing those afterburners :cool:

mmmmmdonuts said:
So yes it looks like you set it up correctly:)
Click to expand...

I'll stop fretting. Many thanks for the help :)

metallicafan said:
So i got a BOINC notice today that GPUGrid now supports AMD GPUs. Interested i went and looked at the thread: http://www.gpugrid.net/forum_thread.php?id=2839

Unless I am completely losing it, it seems that GPUGrid is using AMD GPUs under a separate project called 'donateathome' to mine Bitcoins as a donation to the project. Kind of a neat idea but yet it seems a little strange at the same time. Thoughts? Good idea?
Click to expand...

Well spotted. I understand why they're doing it and I've no wish to diss anybody mining Bitcoins, it just isn't for me. If it came to a choice, I'd rather donate cash as it cuts out the middle man or middle computer in this case. Somebody has to crunch the science, be it distributed or not, but funding doesn't have to include additional energy consumption.
 
Is Docking@home no longer a target? I see that I am about the only user in it, but I thought it was mentioned recently?
 
Yes, suggestions for a newb are appreciated. I've just been doing the WCG stuff. I have a fantastic new board from R-Type that needs to be abused. With the 2600k @ 4200 and 8 GB. Please give me a target for 8 threads....
 
hiodie said:
Is Docking@home no longer a target? I see that I am about the only user in it, but I thought it was mentioned recently?
Click to expand...
Yes we still need to do some in Docking - I have had problems in the past with Intel CPU's in Docking and only have 1 AMD CPU.
 
0ldways said:
Yes, suggestions for a newb are appreciated. I've just been doing the WCG stuff. I have a fantastic new board from R-Type that needs to be abused. With the 2600k @ 4200 and 8 GB. Please give me a target for 8 threads....
Click to expand...
The 2600K would be great for the current push we are having in ABC.
Be sure to join Team [H]ard|OCP
 
hiodie said:
Is Docking@home no longer a target? I see that I am about the only user in it, but I thought it was mentioned recently?
Click to expand...

I will send a little AMD Docking help your way in about 12 hours and stay until F@H gets some stuff fixed. I am tired of wasting my electricity.
 
hiodie said:
Is Docking@home no longer a target? I see that I am about the only user in it, but I thought it was mentioned recently?
Click to expand...

0ldways said:
Yes, suggestions for a newb are appreciated. I've just been doing the WCG stuff. I have a fantastic new board from R-Type that needs to be abused. With the 2600k @ 4200 and 8 GB. Please give me a target for 8 threads....
Click to expand...

Yeah, we still have plenty of work to do in Docking. After a few more days on ABC I'll probably switch back over to Docking. I ran about 200K worth of Docking after the Christmas Challenge was over with an i7 860 and a 2500K without any problems but ???? Two CPUs on the project for a month isn't exactly any all inclusive sample size. :) Let us know if you have problems. Good to have you crunching with us!
 
I’ve not had any problems crunching Docking with Intel chips of the Sandy Bridge and Westmere/Gulftown variety. I have read on their forum that there had been problems with Wolfdale (Core2Duo) chips.

With reference to the same project, I seem to have a rocket in my rear view mirror! Just as I was getting over Razor and DooKey stomping all over me on Poem, it’s happening again. This latest travesty will of course call for a thoroughly disproportionate response at the weekend :D

Very nicely done KMac :)
 
Welcome aboard acewolf, we are grateful for your help in Yoyo.
Speaking of Yoyo, I need to get back in there someday soon and catch up to everyone who blasted by me in the last Yoyo push.
 
Is it just me or does scoring across all projects seem really high today? I have almost double the points that i would normally have in several different projects. And when I look at team graphs on FreeDC it seems that a lot of teams took a very upward jump today. Did i miss changes tot he scoring system or something else?

Also if anyone sees any crunching stats on the new AMD 7700 i would be interested. Seems like we could use another AMD GPU. :)
 
You are looking at free-dc.org stats, aren't you?
His power was off for a few hours yesterday and I think it affected his update time, so yesterday was like 20 hours and today is like 28 hours, but my numbers are made up and for illustration only.
 
Wieferich@Home has now been removed from the DC Vault. Despite a TON of time devoted by Sazan this was still one of our worst projects and a very hard project to progress in. So this has moved us to within about 1100 Vault points of the Knights!
 
After a brief hiatus I'm throwing my systems on Docking. We should be able to make a few spots with the effort I see us throwing into it.
 
I've started the switch over to Docking.

+ For anybody wishing to play around with HP Cloud computing who missed the boat last time round, I received an email this morning talking about the usual 5 'pre-qualified' referrals if they put “@hpcloud” in the Twitter field on the account application before the 22nd of Feb. If it's anything like last time, that just means they're handing out accounts again regardless of whether it's a referral or not.

Briliu said:
Haha Phoenicis. How's it feel to be on the other end of the stick? ;)
Click to expand...

Et tu, Brute?;)
 
It has been far too long and I am finally back! Living in Cali is expensive but, since my electric bill was only $60 last month I am definitely going to start crunching again.

I see everyone has been hitting docking@home hard so I will jump on that bandwagon :)
 
etothe3ric said:
It has been far too long and I am finally back! Living in Cali is expensive but, since my electric bill was only $60 last month I am definitely going to start crunching again.

I see everyone has been hitting docking@home hard so I will jump on that bandwagon :)
Click to expand...
Hi Eric - Good to hear from you! Other than the expense how do you like living in Cali?

DooKey said:
Is it just me or is Docking not sending out any more work?
Click to expand...
Looks like they are out of work (at least for the moment).
http://docking.cis.udel.edu/status/

EDIT: Looks like Docking now has more work available. :)
 
Last edited:
etothe3ric said:
It has been far too long and I am finally back! Living in Cali is expensive but, since my electric bill was only $60 last month I am definitely going to start crunching again.

I see everyone has been hitting docking@home hard so I will jump on that bandwagon :)
Click to expand...
Nice to have you back Eric!
I haven't seen an electric bill under $200 for a couple years and my electric company raised it by 5% as of January 2nd - like I don't pay enough already.
 
I have received an invitation code for Team [H]ard|OCP account creations from Neurona's administrator.
If you would like to participate in this project please Private Message me and I will share it with you.

Info about Neurona:
"Neurona@Home is a BOINC-based project with the main aim of simulating the behavior of a large assembly of cellular automata neurons connected in a complex network.

WARNING
Please be aware that, due to the nature of the network being simulated, this project requires a HUGE amount of RAM, between 3 and 6 Gigabytes per Work Unit, and quite likely this quantity will increase in the next months as the investigation goes further. You will need a suitable 64 bit Linux or Windows PC (due to the RAM requirements there is no 32 bit worker available). Because of this, subscription to this project is made only by invitation code upon request of the volunteer."
 
New version of Prime 95 is out. It takes advantage of AVX on SB processors. No improvements on non SB platforms, but I am seeing a 20+% increase in speed on 17 or bust thus far. (from v26.6)

I am even finding stability on my 2600k from the v26.6, so I don't think you will get a much better stress tester than this.
 
Congrats to phoenicis for achieving 100 Million Combined BOINC points! That was quick! nice work sir. Cheers.
 
You must log in or register to reply here.
Back
Top