DC Vault 2

its strange, with auto for muon, my 2600k shows about max of 95% cpu usage. you would think the gpugrid could slide in there for some cycles. i have a queue filled with primegrid now so im going to wait to try gpu grid again.
 
We'll get there eventually :)

I saw that we're pretty damn close to 2 spots in Milkyway so i fired up my new 6970 on it. HOLY CRAP is it significantly better than my 4870 (for less power too)!

Also, for the next few months I'll have a lot more access to the hex core at work, so my muon points should go up a wee bit ;) (This weekend I was able to use the hex and the 12 so there should be a bit of a spike lol).

Fight on, guys! Awesome push in Muon :D
 
dmaine said:
Noticed something interesting today, since i have been doing Muon1, my GPU grid stats dropped considerably. I checked gpu usage and was down to 35% GPU usage while running Muon1, when i stopped Muon1, GPU usage went back to around 98%. Never had a issue with GPU usage and any other cpu crunching. I switched to Primegrid gpu only and Muon1 and my gpu usage is back up to 98%. any one seen anything like this, it seems to be only Gpugrid and Muon1 together don't like each other.
Click to expand...

I’ve not seen this but I’m only running Muon1 on 10 out of 12 threads. I’ve given the GPU (560Ti) it’s own thread using the swan_sync environmental variable. Despite this, with or without Muon running, it’s only reaching mid 80s GPU usage so I’m somewhat envious of the 98% you’re able to get.

metallicafan said:
Props to the whole team on RNAWorld! You guys just cracked the Top 20. :) A special shout out to phoenicis - I saw a couple days of 100K points in RNA World. That's [H]ard if I've ever seen it. Cheers!
Click to expand...

Thanks metallicafan. A fantastic team effort by all :)

mmmmmdonuts said:
Jeez, dmaine, way to mow over me in a matter of weeks in dpad. On another note great job everyone on RNA World. My queue in RNA World is filled with nothing but the XXL units. I guess everyone must be turning them back. One XXL unit I have is 208 hours in and it is still churning.
Click to expand...

I'm one of your wingmen on a particular Intron unit. I don’t think I’m going to make the deadline as it's been crunching for almost 16 days and will timeout on the 31st. With the RNA message board server down it’s not looking good. No idea why they give a 20 day deadline on units that typically take 30 days.

With respect to Muon1 mowing. Beep, beep mon ami ;)
 
Apparently you're supposed to ask for an extension, though with the forums down I guess you're out of luck.
 
We're three days away from moving up a spot in Wieferich@home. That should be some nice points. I'm hoping I don't lose power today as Irene blows through, but hopefully it won't be too bad.

Nice work on Muon guys.
 
Great work on Wieferich@home Sazan! I see we moved up a spot and gained another 200 vault points! We keep gaining Vault points like we have the past month or so and the 7th spot wont be all that far off. :)
 
Still busy with OGR-27 here. I noticed we raising in the rankings almost 1 rank per day currently. Team HardOCP now at #71 and i personally made my way into the Top 100 producers in the overall stats :)
 
Once I get my SR-2 back up and running I'll throw it into the fray.
 
I missed the W@H Vault update yesterday, but I knew it would probably be a lot of points. I'm going to keep running until I get to 500K personal points which should give us a bit of a cushion, but I probably won't try to overtake anyone else. It's just too slow and painful.

We're about a week from an overtake in PSP, which will be huge. Thanks to Razor for his help there. Thanks also to everyone crunching Muon1, and we should make some big gains there in a month or so. TheMTZ continues to crush OGR. :)

My SR-2 running RNA World will be offline next week while I install some new hardware and do some rewiring. It's my main desktop so I want to get the upgrades done quickly, but I don't know if that's realistic. I might fire up another box in the meantime.
 
Reinforcements eagerly anticipated Dookey. Especially when they're of the SR-2 variety.

A hearty well done to SazanEyes and TheMTZ on your Wieferich@home and OGR-27 acheivements.

@ all - Any thoughts on a new thread and team target project(s) for September?
 
phoenicis said:
Reinforcements eagerly anticipated Dookey. Especially when they're of the SR-2 variety.

A hearty well done to SazanEyes and TheMTZ on your Wieferich@home and OGR-27 acheivements.

@ all - Any thoughts on a new thread and team target project(s) for September?
Click to expand...

No problemo. Just let me know which project to run.
 
DooKey said:
No problemo. Just let me know which project to run.
Click to expand...

:cool: I'm a bit of DC Vault tactical noob so I was fishing for suggestions or one of Sazan's helpful priority lists. I'm currently leaning towards an all rig assault on Muon1 to reduce the overtake timescales there. It also happens to be a project that really interests me. The fly in the ointment is whether I can easily get it running via Wine on my 4P Linux only rig at the weekend.
 
Last edited:
Another one to look at is GIMPS, i was on it for a while , we have a lot of ground to cover to gain spots but could be worth it. But it supports linux. One of the problems is it takes so damn long to run, didnt want to leave my machines running something that could crash and die and lose a months worth of work. In Muon, we're 34 days from 11th place and 64 days from 9th , some big gains there.
 
Seems like we have some good milestones coming up in the coming months. :) September is going to be fun.

I will actually be offline for next week. I'll be shutting down a couple rigs tomorrow and my last one on Sunday. Most of next week I'll be hiking in the Rockies. :cool: So unfortunately my production will be 0 for 4-5 days but I'll be back strong running Muon again soon.

Edit: Sazan - How was Irene in your area? Hopefully you came out ok?
 
Last edited:
I never lost power from Irene (or the Great Virginia Earthquake of 2011 :) either). My lights flickered a few times but I have a UPS on every computer.

GIMPS is probably the #1 project, and we've lost ground there recently. If you run double-check work, it should only take a week instead of a month, and if you're really impatient, the trial factoring work is even quicker. They credit you for amount of CPU time so it doesn't really matter which you run, but I believe frial factoring doesn't support hyperthreading so you may only be using 50% of your CPU. I still haven't fixed my i7 that was running GIMPS, but I'll get back to it in the fall.

Muon and Prime Sierpinski Project are also good choices. Basically look at anything below 9000 in the Vault score, which are all non-BOINC projects I believe. That should be the top 10. W@H is too slow to make a lot of progress, though. (I forgot how silly their scoring system is until I started running it again.)
 
Thanks for the suggestions guys :)

I'll start to slowly move over to Muon1 while reading up on GIMPS and PSP in the interim. The initial play with Muon via Wine wasn't successful - I'll give it another try once the RNA queue empties which might take a while given that I was granted a deadline extension on 2 never ending Intron units. Aborting them would somehow seem impolite after such a quick response to the extension request.
 
A quiet Sunday so I’ve had some time to play around with things. The SR2s have cleared the RNA queue including one of the Intron units which, on its own, gave the pleasant surprise of 48K points:eek:. I set them off running Muon on 24 threads each only to be met by a CPU utilisation of 60%. Each box has been switched to 4 instances with 6 threads which has bumped the CPU number up to 95% - much better.

I managed to get Muon going via Wine on the Opteron rig but it was operating dog slow. Not sure what I was doing wrong but I quickly ran out of patience so the machine is now crunching GIMPS. I’ve returned my first results and now have 4 points – yippee:D.

Thanks to kenlo and sbrug for continuing to battle away at GPUGrid. We’re making slow but steady progress up the ranks and are managing to keep our distance from the DPCs.
 
phoenicis - Fantastic job on Muon. Over 2 million points today alone.

I will probably be joining you on GIMPS very very soon.
 
Good job on Muon, phoenicis. I had the same CPU utilization issues on my SR-2, so I just run Muon on my lower-end hardware and run something else (like GIMPS) on the SR-2.

I'm dealing with some leaks in my roof right now, but luckily they're not near any of the computers. :eek:
 
SazanEyes - would you want to make a run at Rechenkraft in PSP?
That would give us the 1k DC-Vault points (plus the Muon1 points) we need to pass them in the DC-Vault for 7th place?
I have had 4 cores on PSP but I can x4 that for 16 cores which will help reduce the 2 months the stats currently show it would take to pass them.
 
I'm going to keep the SR-2 on PSP, but I just fired up a hex yesterday so I'll add that. It's finally starting to cool off a bit, but maybe that's just the rain.
 
By my figuring we should pass them in about month and if the Hex is added then even sooner.:cool:
 
Well im back. Unfortunately the hiking trip was a bit of a bust as it rained most of the time. :( I came back with only hiking a few miles and only about 150 pics on the DSLR. Oh well.

Got the i7 back on Muon. I'll add the other quad next week.

The Northbridge fan on my old Opteron 170 dual core is about to die. Its making a horrible whine. This seems like a good excuse to buy a new MB/CPU combo. Anyone heard anything about the new AMD chips? I thought i heard they were going to be released in Sept? They are 8 core chips right?
 
I added the hex on PRP just now, so that's three hexes pounding away (and the new one is my fastest at 4GHz). I'm probably going to start back on Muon soon too. I still need to fix the i7 that had been running GIMPS, plus the UPS that is hopefully just a bad battery. Maybe I can work on that over the weekend.
 
metallicafan said:
Well im back. Unfortunately the hiking trip was a bit of a bust as it rained most of the time. :( I came back with only hiking a few miles and only about 150 pics on the DSLR. Oh well.

Got the i7 back on Muon. I'll add the other quad next week.

The Northbridge fan on my old Opteron 170 dual core is about to die. Its making a horrible whine. This seems like a good excuse to buy a new MB/CPU combo. Anyone heard anything about the new AMD chips? I thought i heard they were going to be released in Sept? They are 8 core chips right?
Click to expand...

Glad to have you back.

It looks like the AMD chips are going to be released in October time frame but at this point I wouldn't put it past 2012.

http://blogs.amd.com/work/2011/09/07/the-start-of-a-new-era/
 
I've got my SR-2, 2600K, and 2500K working on some Einstein@home. I think I can gain us a spot or two there in the near future. I prefer running BOINC is the reason I'm there. If there's a BOINC project I can make a faster impact in please let me know.
 
In case anybody missed it yesterday, we got a seriously large Vault bump from PSP courtesy of SazanEyes and Razor_FX_II. Hope your roof leak has been remedied Sazan.

Whilst working away I’ve had too much time on my hands to read up on GIMPS. I didn’t realize that it could be crunched using GPUs and am now experimenting using 2 x 560Tis. This should allow things to move more swiftly but will have a knock on effect on GPUGrid. A bit of a dilemma but I guess we’ll have the opportunity to make up lost ground once the temps drop on the other side of the pond.

Lest I forget, a welcome back to our soggy Rocky Mountains explorer.

DooKey said:
I've got my SR-2, 2600K, and 2500K working on some Einstein@home. I think I can gain us a spot or two there in the near future. I prefer running BOINC is the reason I'm there. If there's a BOINC project I can make a faster impact in please let me know.
Click to expand...

Really nice boost there DooKey.

I’ve not actually run Einstein yet, so I’m not too sure about the points system and therefore how easy it is to move up, but it’s probably not the highest impact project. Due to the number of teams involved, for each place we move up there's only 1 point to be had. On the recently targeted project, RNA World, there are 27 points on offer for each place with others such as Poem and Docking falling somewhere in-between. Apologies if you already know this but the formula used by the Vault is:

Project Score = 10000 - ((Position - 1) * (10000 / Total teams).

I’m sure someone will come along and correct me if I’ve ballsed anything up.

I think the main thing is to crunch whatever interests you, regardless of points accrual. Having said that, if both happen to co-coincide then that’s a real bonus.
 
Last edited:
Thanks for the crunching power DooKey. Those are 3 really solid builds! I think the BOINC project that would give us the most DC Vault bang for our buck would be yoyo@home. Im not sure how many PPD a SR-2 could do there but another million points there would move us up about 10 spots or so. But like phoenicis said, we are just happy to have you aboard, so crunch whichever projects that interest you. :)

phoenicis - Great work on Muon! We are gaining pretty steadily now. I'll add another quad this coming week.
 
Sounds good on the yoyo@home. I'll jump on that tonight.
 
Last edited:
phoenicis, how did you setup your gpus to crunch on gimps, i've looked at the instructions and they are pretty sparse. does it grab the wu automatically or do you have to transfer them to the gpu? I have all my gpus currently on primegrid as they were not efficent on gpu grid while doing muon1.
 
dmaine said:
phoenicis, how did you setup your gpus to crunch on gimps, i've looked at the instructions and they are pretty sparse. does it grab the wu automatically or do you have to transfer them to the gpu? I have all my gpus currently on primegrid as they were not efficent on gpu grid while doing muon1.
Click to expand...

I couldn't find a Noddy guide anywhere so everything was a bit trial and error unfortunately. To ease the pain for others I'm willing to give a guide a bash sooooooo the steps for Nvidia Cuda graphics cards under Windows 64 bit are:

  • Download the mfaktc-0.17 Windows x64 client and cudart64_32_16.dll files from here (links are in the pdf guide) and create a folder into which you extract the files.
  • Create a empty text file within the folder named 'worktodo.txt'.
  • Sign into your GIMPS account and open up the 'manual test' sub link 'assignments'.
  • Select 'trial factoring' from the drop down menu, enter 1 into the number of cpus and then the number of assignments you want. I request no more than a 100 at a time, any more and you risk the server timing out. They're working on automating the assignment/results process but aren't there yet.
  • Copy and paste the provided assignments into the worktodo.txt file you created. The TF assignments typically only run to a bit level of 71 so if you have a fast Fermi GPU the given assignments will be swiftly completed. You can slow this down by manually increasing the bit level of the assignments by a little to say 73 or 74 which will increase the depth of the Trial Factoring. You do this by changing the last one or two digits of the assignment. From what I've read this is apparently normal practice although I'm not entirely sure how far you should take this.
  • If you're running more than one GPU on a single PC you'll have to run multiple instances following the above instructions for each. Create a shortcut to the mkfakt executable in each folder then right click and open properties. Add -d 0 to the target for the first gpu, -d 1 for the second etc. It's similar to running multiple F@H GPU clients in this respect.
  • There are settings you can play with within the config file but best leave as is until one of us gets brave.
  • Run using the executable or, as appropriate, the shortcuts.
  • When you are ready to send results, sign into your GIMPs account, select the 'manual testing' sub link called 'results' and use the upload button to upload the results.txt file within the folder you created earlier. Always wait until an exponent has completed it's full run to the set bit level before uploading it's results and remember to delete the results file after a successful upload, a new one will be created when you complete more work.
With a fast GPU, your CPU will be the bottleneck so it is recommended that you use a full core with HT switched off. There is no facility to run on multiple threads so, although I've not tried this yet, you can run more than one instance per GPU thereby effectively doubling the amount of CPU cores supporting each GPU. This is said to increase output by over 50% and push you closer to 100% GPU usage. I'm really going to watch out for the heat output when I give this a go at the weekend.

I'm again working away so please forgive any mistakes/omissions as all the above is from memory. Don't hesitate to shout if you hit a problem although it may be a case of the blind leading the blind I'm afraid.

Best of luck.

Bob

PS Although GIMPS is sucking up some Muon resources the serious drop in Muon output today is from my better half inadvertently cutting off the ventilation to my fastest rig which, rather unsurprisingly, caused it to power off for most of today. Normal service has been resumed.

Edit: Removing apostrophe corruption
 
Last edited:
thanks that sounds a bit complicated. if i can get my job off my butt i might fool around with that. since i switched all my gpus to primegrid since the wierd gpu utilization when running muon and gpugrid at the same time, i havent tried to go back. there arent many gpu only dc projects that we can make big gains, gimps looks like the best if we can figure out the process.
 
It's really much less complicated than it looks. This is probably why I don't write text books for a living:D With the latest drivers already in place, setting up a single GPU should take no more than 10 minutes.

I'd agree GIMPS is looking like it could be a good target. It's in the early days of the experiment but, if all goes well with the 4 GPUs crunching, I'm estimating that over 500 Vault points could be possible within a month or so.
 
I downloaded and read up on the ATI version today and will have a go with 6x HD48xx's tomorrow if it doesn't use to much CPU time.
 
how many of the TF work units are you able to do in a day? I wonder how the 460 and 560 would do on a double check unit that was taking a few days using 4 threads of my 2600.
 
Last edited:
dmaine said:
how many of the TF work units are you able to do in a day? I wonder how the 460 and 560 would do on a double check unit that was taking a few days using 4 threads of my 2600.
Click to expand...

Running a single instance and assuming the current typical assignment to a bit level of 71, one of my 560Tis (with a factory overclock to 880MHz) can get through almost 30 TF units per day for a total output of approx 130GHz days per day.

The parallel processing nature of a GPU is apparently ideal for trial factoring but less so for the Lucas Lehmer test which is quoted at 15% of the speed on the same card. I've not tried running LL units on a GPU but they are the most needed work type and the upside is said to be that, unlike the TF units, they use very little CPU resources.

Razor_FX_II said:
I downloaded and read up on the ATI version today and will have a go with 6x HD48xx's tomorrow if it doesn't use to much CPU time.
Click to expand...

Thanks for the prompting Razor. I've amended the post to clarify that it is aimed at Nvidia Cuda cards.
 
Last edited:
You must log in or register to reply here.
Back
Top