DC Vault 2

They usually are not notified. However, it was mentioned in the PSP forums that the idea was brought up do to the complications. It doesn't explicitly point out the violation of DC-Vault inclusion rules, but the message was pretty clear in my book.

http://www.mersenneforum.org/showthread.php?t=19293

Project eligibility is governed by the following guidelines:

The project must:

be active
accept new members and teams immediately upon registration
have parsable team stats
have team stats that are updated regularly
provide a client program which runs on a local PC

The project must not:

be a keylogger or mouseclick counter
have a maximum number of teams or members
 
Now is the time to discuss next months push. Any other suggestions?

I was going to suggest Malaria Control but their feeder appears to not be running and so not sure not sure what the issue is. I'm thinking their server may need more attention and can't always keep up. However, we should still consider a push there soon.

Moo! might be a good one to push since it feeds RC5-72. For those that don't want to run their client. However, if you are running AMD, take note the issues with the 7970's not liking the app that BOINC downloads. There is a thread in their forums that addresses it and explains how to get it working. Make sure to follow these instructions if you do run Moo! http://hardforum.com/showpost.php?p=1040794811&postcount=1947
 
Last edited:
At our current pace, we will be over taken in ~ 2 months. ~320 days to be over taken again. ~663 days for another over take. We wont take another position for ~1,008 days.

If anyone has unlimited data, I encourage you to run the Majestic-12 client. Even if you do it minimally, it could give us more time. We need some big hitters to help push. :D
 
I have 1gb up 1gb down connection with unlimited data cap. Do we have a Majestic-12 setup write up?
 
If anyone is interested, we still need setup guides for OGR-28 and RC5-72 clients. RC5-72 can be supported in BOINC through Moo! but you would have to follow my notes on it. Otherwise, we should have a guide for using their client.
 
I've got Majestic-12 going on my laptop and joined the [H]. We'll see how it goes.
 
If anyone has unlimited data, I encourage you to run the Majestic-12 client. Even if you do it minimally, it could give us more time. We need some big hitters to help push. :D

I have a standard Uverse 12M down but crap for up (about 640K?) and 350GB/month before AT&T starts to cap - how upload dependent is it?
 
I have a standard Uverse 12M down but crap for up (about 640K?) and 350GB/month before AT&T starts to cap - how upload dependent is it?

On my Atom server, which is too weak to saturate its 100/100 connection, it averages just under 4mbps down and less than 1mbps up - it's currently at 7mbps down and only 200kbps up as it's not currently uploading results. The upload is used partially for crawling as obviously it has to send the requests for pages etc, and then for uploading completed results after they're compressed. I would say you're looking around a 10:1 ratio of down:up.
 
And you can limit the number of crawlers too if you need to reduce the amount of data usage. Also, make sure to use the compression. This may make it more CPU intensive, but will cut on data. When I'm on unlimited connection, I don't bother with the compression because I don't care about the data...lol.
 
Okay, it's up and (I think) running... I decided to play it safe for now and just chose a default 2M up/512K down DSL profile, then can move up from there once I see what the numbers look like. Of course, right now, all I'm getting is a response code of NoMoreURLsAvailable, so I assume it'll find something whenever someone kicks the server again.

For now, I'm off to take advantage of TigerDirect's $15 off sale and snag a 750ti for $125...:D
 
Okay, it's up and (I think) running... I decided to play it safe for now and just chose a default 2M up/512K down DSL profile, then can move up from there once I see what the numbers look like. Of course, right now, all I'm getting is a response code of NoMoreURLsAvailable, so I assume it'll find something whenever someone kicks the server again.

For now, I'm off to take advantage of TigerDirect's $15 off sale and snag a 750ti for $125...:D

I see you on the stats now, nice :)
 
Yeah...when you first start the client, sometimes it can take up to a couple hours to get work. But once you have work, you can pretty well keep it fed just fine.

750ti's are good if you have a need for very low power draw cards. However, I would almost recommend holding off for some of the newer cards to come out. A lot of projects still don't use CUDA 6 which is where the real benefits of Maxwell come in. Just some food for though.
 
Now is the time to discuss next months push. Any other suggestions?

I was going to suggest Malaria Control but their feeder appears to not be running and so not sure not sure what the issue is. I'm thinking their server may need more attention and can't always keep up. However, we should still consider a push there soon.

Moo! might be a good one to push since it feeds RC5-72. For those that don't want to run their client. However, if you are running AMD, take note the issues with the 7970's not liking the app that BOINC downloads. There is a thread in their forums that addresses it and explains how to get it working. Make sure to follow these instructions if you do run Moo! http://hardforum.com/showpost.php?p=1040794811&postcount=1947

Does Moo! have work for CPUs, or a more relevant question might be is it worth running it on a CPU rather than GPU?

Looks like we could gain a few spots in SETI quite quickly, although the points per position aren't that much.
 
I haven't ran it on CPU since it came out and didn't run it long since GPU's crunch through them so much quicker. But if you are wanting to maximize your output, run it on both at the same time. I have typically just ran it on older GPU's because they supported even the lowliest of GPU's.
 
Yeah...when you first start the client, sometimes it can take up to a couple hours to get work. But once you have work, you can pretty well keep it fed just fine.

750ti's are good if you have a need for very low power draw cards. However, I would almost recommend holding off for some of the newer cards to come out. A lot of projects still don't use CUDA 6 which is where the real benefits of Maxwell come in. Just some food for though.

Well, actually, it IS going to replace a low power draw card - one of a pair of your old 210s that has been slowly crunching away. But more importantly, I hit a few games in my Steam backlog that my 9800GTX+ can't quite run as well as I'd like at 1920x1080, so this allows me to keep the 9800GTX+ as a secondary DC card and upgrade gaming ability on the same PSU for now... :cool:
 
That should work pretty well for you then. I have two of them at the moment. My next large purchase will probably be a 780, but the price will still have to drop a lot before it is even close to my price range.
 
That's what I figured - plus, it will be a good extra card to have around if/when I get to build a new HTPC for the family room... They seem great at decoding video without putting stress on the CPU, and I'm figuring when I build box #4, I can easily do it <350 W with a good i7 and this card in a small case hidden out of sight. Power bill is already high enough for SWMBO, so until I get solar done, no more 200W video cards for me!

Frankly, I'm hoping to see what comes from the Maxwell architecture later this year - 780 performance for 120-140W would be pretty damn sweet! Or there's always the upcoming AMD Tonga architecture, promising the same lower power to performance ratio as Maxwell... Be nice to retire my 4850 (and then upgrade the XP box it's stuck running on) - but as long as it keeps putting up that 60K ppd on Moo!, I'm not in a BIG hurry...
 
I haven't ran it on CPU since it came out and didn't run it long since GPU's crunch through them so much quicker. But if you are wanting to maximize your output, run it on both at the same time. I have typically just ran it on older GPU's because they supported even the lowliest of GPU's.

I have no GPUs crunching any more to lower heat output and power draw (and also to raise the cash from selling them). If it's not really worth running it on CPUs due to GPUs being much faster, I'll focus my CPUs elsewhere, but depending on how warm it gets I'll run my 5850 on it. It was incredibly warm here recently to the point I shut down all my rigs temporarily, then the last week it's been pouring down with rain - perfect DC weather :)
 
For those running Collatz in June, you may want to either raise your cache levels or perhaps make sure you have a backup project lined out.

Potential Outage After June 5 thru June 15
From June 5-15, the temporary Collatz system admin will be the cat sitter. That doesn't sound very reassuring does it? Sorry, but if the server goes down and requires anything other than a simple reboot, there may be an extended outage.

Unfortunately, the chances of that happening are higher than normal since Commonwealth Edison (a.k.a. ComEd, the electrical company in the area) failed to notify me on their original attempt to replace my power meter last week but were unable to do so since it evidently takes at least two people to do it. Since they didn't call before the last attempt, Murphy's Law suggests that they will not call before the next attempt and that it will take place while I am on vacation. The cat sitter doesn't know how to switch over to the generator if the outage exceeds the UPS battery so if the power is out for an extended period, Collatz may remain down until I get back. 27 May 2014, 2:54:44 UTC · Comment
 
I stimulated the stats export at CPDN with a not so appealing point of view for most people. It seems to have warranted the attention of one of the two scientists which happened to possibly fix their exports issue. I don't normally approve of going about things that way, but it seems to have worked. Maybe now the project wont get pulled from the Vault.
 
I was all set for setting up an MJ-12 box at home but my internet has been consistently dire for the last week or two. Waiting for them to get off their collective butts and take a look, will throw a box at it once the connection's back to working as it should.
 
I've got a bunch of cores crunching PSP right now in the hopes they'll add me to our team and our rank will increase. Considering the 300+ point jump that would create, I'm hoping they'll come around, allow more automation for team joining and stay in the Vault.

I assume each project will be contacted before being removed from the Vault with explicit reasons they are being removed. Maybe this would light a fire under certain admins...

I got a pm from ltd that he will update the team and stats next week...
 
Well, when he does, let me know exactly how long it took to do it.

Also, we haven't had much chatter on next months push. It is only a few days away. Ideas?
 
Also, if there are any other projects suggested for addition or removal in the Vault, please speak up or at least give an opinion. The more ideas and info we get the better.
 
If you have some goals at SIMAP, you may want to get them in while you can.

http://boincsimap.org/boincsimap/forum/viewtopic.php?f=45&t=1591

I can't read that, so the front page has the news
SIMAP will leave BOINC by end of 2014 SIMAP is currently moving towards new algorithms for similarity calculations and scoring. After extensive tests and evaluations during the last months it turned out that the best way to continue SIMAP will be to use dedicated local hardware in our institute. SIMAP will therefore use BOINC only to maintain the current SIMAP database until the end of 2014. BOINC clients for the new SIMAP database will not be released and there will be no workunits for SIMAP after 2014. After 10 very successful years of SIMAP within BOINC this decision was not an easy one for us. However, we thank all users for their support in the past and in the upcoming last period until the end of the year.

With this news, I would suggest possibly a push there while it is around. They are preparing work now and estimate it to be ready June 7th.

Here is the English thread... http://boincsimap.org/boincsimap/forum_thread.php?id=88#276
 
Speaking of SOB, are these WUs supposed to take a few days to complete? I'm running my 980x with 3 workers at 4.0 Ghz.
 
Most of mine take ~ 2 weeks depending on the units. But the system I'm running them on is not OC'd and is 3.2GHz.
 
If you have some goals at SIMAP, you may want to get them in while you can.

With this news, I would suggest possibly a push there while it is around. They are preparing work now and estimate it to be ready June 7th.

Here is the English thread... http://boincsimap.org/boincsimap/forum_thread.php?id=88#276

I might throw more logs on the fire for SIMAP until the work runs out at the end of the year. I'm still finishing all the PrimeGrid units my clients downloaded during the challenge and I'd rather not abort them, but I set "nonewwork" flag.
 
I might throw more logs on the fire for SIMAP until the work runs out at the end of the year. I'm still finishing all the PrimeGrid units my clients downloaded during the challenge and I'd rather not abort them, but I set "nonewwork" flag.

My goal is 1 million points, so if I hit that, I'm happy. :D
 
Speaking of SOB, are these WUs supposed to take a few days to complete? I'm running my 980x with 3 workers at 4.0 Ghz.

Yep, My i7-620m has 15 days estimated to completion. But Prime95 will update SOB once per day with current status and estimated completion and points will be assigned. You can also choose manual communication from the menu and check send completion dates to update more frequently...

Test/Status will also show the estimated completion date...
 
I stimulated the stats export at CPDN with a not so appealing point of view for most people. It seems to have warranted the attention of one of the two scientists which happened to possibly fix their exports issue. I don't normally approve of going about things that way, but it seems to have worked. Maybe now the project wont get pulled from the Vault.

Also, we haven't had much chatter on next months push. It is only a few days away. Ideas?

Looking at the Vault, it unfortunately looks like non-BOINC projects like Majestic-12, MUON1, and SoB all could give us big gains with a hard enough prolonged push. I have M-12 up and running, and right now I appear to be on pace to use about 100GB of my 350GB cap on the project. I know my wife/daughter use over 100GB/month on Netflix monthly, so I might be able to double my production there, but I'm not sure we have enough team bandwidth to actually make much forward progress...

Climate Prediction is a BOINC project that looks like we could just dominate if we hit it hard - with just Jeffy & Gilthanis hitting it, our overtake list shows 32 teams we'll overtake in the next month. Problem is, each of those spaces is only worth about 1.3 points, or just over 40 Vault points. So I'm not sure how many inactive teams we'll be able to quickly jump before things get difficult (and the current Stats mess doesn't help!), and if there's talk about removing it from the Vault, why waste a push?

Yoyo is another BOINC project where we're a few hundred points behind the top team, and each spot is worth about 9 points. So if we can jump the closest 6-7 teams, we can gain 50-60 points there...

And the last one I see is Asteroids - each spot is worth a little over 11 points, and enough RAC there would give us 5-6 spots, or almost 70 points...

Biggest problem with Asteroids and Yoyo is that we're ranked in the top 30 for both, so it's hard to pass a lot of teams!
 
Back
Top