Questions and stuff
 

Finally got some more cores cooking on the llrnet "nplb.rieselprime.org" port 300 server. Got some old and some new systems on it. Have a couple dual core 64-bit ones that still don't want to play nice with the llrnet though.

Being new to this effort, and not making any manual reservations, I am assuming that llrnet reports the prime results to the server and that the results do not need to be posted anywhere else? My intent is to just keep running through the llrnet server and not do any manual reservations.

@Anonymous - Thank you for the url for the sr5 llrnet client. That one pretty much solved the majority of my issues. Works just fine on the OpenSuse 10.1 thru 10.3 64-bit systems. I can't say the same for the mepis-64, as something is wierd with three systems running that yet. But I guess over the next few days I may just install OpenSuse on them and that will allow me to bring some more cores on board. :) Thank you again. :)

Actually, since the primes are in top-5000 range, you [i]will[/i] need to report primes to the top-5000 website (and the "report primes here" thread in this forum, though that's pretty easy, especially if you're already reporting them to the top-5000)--if you want to be automatically notified by email of all new primes you find, just send me or Adam (username AES) your email address and we'll get it plugged into the system. :smile:

[QUOTE=Anonymous;132748]if you want to be automatically notified by email of all new primes you find, just send me or Adam (username AES) your email address and we'll get it plugged into the system. :smile:[/QUOTE]

you have a PM.

[quote=Brucifer;132750]you have a PM.[/quote]
Okay, PM received and email entered into database. :smile:

How come there is a set of stats at:
[url]http://www.rieselprime.org/nplb/NPLB_Scoring.htm[/url]

and another set of stats at:
[url]http://nplb.rieselprime.org/stats/stats.php?a=reset[/url]

I assume that the first one by kar_bon must reflect the manual reservation stuff??? And the second is the automated llrnet stuff????

Is there any centralized composite scoring for the nplb effort? Just wondering as this sorta looks a little scattered out? :)

[quote=Brucifer;133195]How come there is a set of stats at:
[URL]http://www.rieselprime.org/nplb/NPLB_Scoring.htm[/URL]

and another set of stats at:
[URL]http://nplb.rieselprime.org/stats/stats.php?a=reset[/URL]

I assume that the first one by kar_bon must reflect the manual reservation stuff??? And the second is the automated llrnet stuff????

Is there any centralized composite scoring for the nplb effort?[/quote]
The stats at [URL="http://www.rieselprime.org"]www.rieselprime.org[/URL] are both for manual work, and LLRnet work (there are separate scoring tables for each). It's essentially the "centralized scoring" that you're looking for. However, it is only updated manually, and not very often at that because it's a lot of work for Karsten to keep all his pages updated.

The one at [URL="http://nplb.rieselprime.org"]nplb.rieselprime.org[/URL] is the live, automated stats, for Adam's port 300 server only. (The [URL="http://www.rieselprime.org"]www.rieselprime.org[/URL] LLRNet stats, by contrast, include all servers, past and present.) So, the [URL="http://nplb.rieselprime.org"]nplb.rieselprime.org[/URL] stats probably the ones you'll want to watch if what you're interested in is watching rankings and all that.

Sometime in the future we may have similar stats available for the port 5000 server ([URL="http://llrnet.ironbits.net"]llrnet.ironbits.net[/URL]), but until then, port 5000 is only included in the manually-updated [URL="http://www.rieselprime.org"]www.rieselprime.org[/URL] LLRnet stats.

Thankyou for the reply. The question came up on the AMD_Users forum, so I copied and pasted your reply following the question at the AMD_Users forum. Attributing the answer to you of course! :)

heh... rather quiet around here lately now that the rally is over with and the scores have all been combined. Just sitting here bored, late at night, doing one of those late night things. :)

[quote=Brucifer;134887]heh... rather quiet around here lately now that the rally is over with and the scores have all been combined. Just sitting here bored, late at night, doing one of those late night things. :)[/quote]

Quiet is always good when you have to deal with guys like Gary and Anon, they :blahblah: too much!

Well the wife decided that she really didn't want to retire yet cause there are some other goodies that she still wants to get. So that "crises" is over with, at least for this month... She's talking about doing another year and a half or so now. So anyway I'll be crunching some again. Just aren't going to be investing in a bunch of new stuff other than maybe another one or two quads at some point in the future. As they used to tell us in the service, gotta be flexible... :)

[quote=Brucifer;136194]Well the wife decided that she really didn't want to retire yet cause there are some other goodies that she still wants to get. So that "crises" is over with, at least for this month... She's talking about doing another year and a half or so now. So anyway I'll be crunching some again. Just aren't going to be investing in a bunch of new stuff other than maybe another one or two quads at some point in the future. As they used to tell us in the service, gotta be flexible... :)[/quote]
Okay, great! For this upcoming rally, could you please put everything on server IB8500? (In fact, that's sort of what it was originally designed for. :smile:) That will take a lot of load off the main servers.

Edit: Ah, just read the rally thread, see you beat me to the punch. :smile:

[QUOTE=Anonymous;136196]Okay, great! For this upcoming rally, could you please put everything on server IB8500? (In fact, that's sort of what it was originally designed for. :smile:) That will take a lot of load off the main servers.

Edit: Ah, just read the rally thread, see you beat me to the punch. :smile:[/QUOTE]

Better late than never............... LOL :geek:

[quote=em99010pepe;134903]Quiet is always good when you have to deal with guys like Gary and Anon, they blah too much![/quote]


:blahblah::blahblah::blahblah::blahblah::blahblah::blahblah::blahblah::blahblah:

Oh, I forgot...

:blahblah::blahblah::blahblah::blahblah::blahblah::blahblah::blahblah::blahblah:


Darn...couldn't blah more. It only limits me to 10 blahs!


Gary

Twas good to get back home. Vacations are good and provide the break from everything, but it's always nice to get home and sleep in your own bed, your own bathroom, and not to forget the good old refrigerator!!! But this heat wave sux. Setting temperature records now for this time of the month/year. Only running about half the farm now cause the poor old A/C is having to do triple duty.

It is a hot year every where.
Just read that some scientists are saying there will be no floating ice left at the north pole by September of this year :yikes:

[QUOTE=IronBits;136942]It is a hot year every where.
Just read that some scientists are saying there will be no floating ice left at the north pole by September of this year :yikes:[/QUOTE]

yeah, I was just reading that too. Really some seriously hosed up weather all over. Look at all the heavy snow fall, the heavy rains and resulting flooding around the planet. And not to forget the really active tornadoes in the US this year too.

Of course we in the US are watching our severely impacted economic situation closely, but ours isn't the only one taking some serious hits either. While we have the added catastrophe of the flooding in the mid-west, China has taken some serious financial hits too from the heavy flooding damage there. Lots of other weather related issues around the planet too.

For those that follow the Mayan(?) calendar thing about the end coming in 2012??? What is supposed to cause it?

You can find lots of hits with Google to read more. ;)
[SIZE=-1]Every so often a force from space causes global cataclysms and mass mutations.

[/SIZE]Authors disagree about what humankind should expect on Dec. 21, 2012, when the Maya's "Long Count" calendar marks the end of a 5,126-year era.

Part of the 2012 mystique stems from the stars.
On the winter solstice in 2012, the sun will be aligned with the center of the Milky Way for the first time in about 26,000 years.
This means that "whatever energy typically streams to Earth from the center of the Milky Way will indeed be disrupted on 12/21/12 at 11:11 p.m. Universal Time"

The date December 21st, 2012 A.D. (13.0.0.0.0 in the Long Count), represents an extremely close conjunction of the Winter Solstice Sun with the crossing point of the Galactic Equator (Equator of the Milky Way) and the Ecliptic (path of the Sun), what that ancient Maya recognized as the Sacred Tree. This is an event that has been coming to resonance very slowly over thousands and thousands of years. It will come to resolution at exactly 11:11 am GMT.
The astronomer Philip Plait has stated very clearly that the Mayan calendar does not end in 2012 at all, that it is like the odometer on your car, as each section of the odometer reaches 9 and then clicks over to 0, the next number to it starts a new cycle, so that when all the numbers again reach 0 all the way across the odometer - the last number will change from 1 to 2 and the new cycle starts all over again.

and I'll leave you with this link: [url]http://www.greatdreams.com/2012.htm[/url]

How's your prime hunting doing these days? [B]:smile:
[/B]

[QUOTE=IronBits;136944]How's your prime hunting doing these days? [B]:smile:
[/B][/QUOTE]

rather poorly...

It's intersting that you guys are talking about global warming and the such here.

I have an actual $50 bet with a business associate that we made 5 months ago that for 3 out of the next 5 years, the avg. global tempurature will be > 0.5 degrees Celsius above the long-term average since 1870. Of course I am wagering that it is above that average.

Last year, there was one of the largest drops in history, which is what inspired our somewhat heated debate. It dropped from 0.8 degrees above the avg. to 0.2 degrees above the avg. He's convinced that's indicative that the 'temporary upward blip' in global tempuratures is over. I told him he's nuts because 5 of the last 6 years before 2007 had been > 0.5 degrees C above the long-term average and that the trend has been strongly upward for nearly 25 years.

He's an ultra-conservative and is completely convinced that global-warming is a myth. I finally told him to put his money where his mouth is and make the bet. He didn't hesitate. Based on what I've heard from several people and in the news, I think I'm already in good shape to take a 1-0 lead this year. :smile:

We're using a neutral and official whether-related website to get the avg. global temp. so everything is on the up-and-up. I'd have to dig it up from an old Email if anyone wants to see it. It's quite interesting.


Gary

[quote=Brucifer;136933]Twas good to get back home. Vacations are good and provide the break from everything, but it's always nice to get home and sleep in your own bed, your own bathroom, and not to forget the good old refrigerator!!! But this heat wave sux. Setting temperature records now for this time of the month/year. Only running about half the farm now cause the poor old A/C is having to do triple duty.[/quote]

Bruce, if I may ask, where abouts do you live? One of our regulars at NPLB lives in TX but I can't remember who it is. I thought it might be you but maybe it's Beyond. You must be in a southern state. It's been somewhat mild here in the midwest...perhaps 2-3 days above 90 so far in the Kansas City area.


Gary

[QUOTE=IronBits;136944]Y
and I'll leave you with this link: [url]http://www.greatdreams.com/2012.htm[/url]
[/QUOTE]

hmmm...... where are all the 60's flower children at now???? :) In the grand scheme of things, we really don't know squat. Haven't figured out the weather on this planet. Sorta makes one wonder what is really at the core of the weather changes....:ermm:

Well, one thing that I have thought of for quite some time... if all that snow/ice is melting, it's putting a lot more moisture in the air, for now...
I'm more worried about when all the fresh water reserves are melted away.

[quote=gd_barnes;136952]Bruce, if I may ask, where abouts do you live? One of our regulars at NPLB lives in TX but I can't remember who it is. I thought it might be you but maybe it's Beyond. You must be in a southern state. It's been somewhat mild here in the midwest...perhaps 2-3 days above 90 so far in the Kansas City area.


Gary[/quote]
I live in Texas. I don't know of any other NPLB members that do. Probably me you're remembering (or trying to). It's been hot and very dry lately (we're in a drought), but not record-setting AFAIK.

I was just refering to record setting for the area I live in, Eastern Washington, and the month/day records is all. Nothing on a national/world record scale if thats what anyone was thinking. :) But we've popped over 100F which while not the hottest I've been in by any means, still makes life difficult for the A/C units. :) And also we are getting higher humidity than we used to get when the temperature was in the 100's. It used to be that this was a pretty dry area.

What are the urls to the stats?

EDIT: Never mind, I found it. :blush:

[quote=Brucifer;139405]What are the urls to the stats?

EDIT: Never mind, I found it. :blush:[/quote]
Even though you found it, I'll include the URL here for others' reference:

[URL]http://nplb.rieselprime.org[/URL]

That's not to be confused with our "old" stats at [URL]http://www.rieselprime.org/nplb/NPLB_Scoring.htm[/URL]; those were replaced with the more-frequently-updated (and, in the case of the AES300 server, automatically updated) stats at the nplb.rieselprime.org web site. :smile:

So someone tell me a little about the future plans for cuda. I went out and bought a play toy, a gpu as in nvida 9800GT. So now I'm wondering if there's gonna be any application for these little hummers, whether llr or sieving?

[quote=Brucifer;152134]So someone tell me a little about the future plans for cuda. I went out and bought a play toy, a gpu as in nvida 9800GT. So now I'm wondering if there's gonna be any application for these little hummers, whether llr or sieving?[/quote]
First of all, I would recommend that you take a look at [URL="http://www.mersenneforum.org/showthread.php?t=10275"]this thread[/URL] in the Hardware forum which contains a whole lot of discussion regarding running prime search apps on GPU's.

As for how any of that would apply to NPLB: As detailed in the above-linked thread, there are currently no prime-searching apps capable of running on GPU's. It is possible that at some point in the future some trial factoring/sieving applications may be ported to GPU's, though there are no ongoing projects of that sort that I know of. :sad:

Thank you.

I need a recommendation for a good cooling fan for a Q9400 installed on an Intel DG35EC motherboard. Something that uses the stock fan mounting holes would be the preferred item. I'm not a long time Intel user so am not familiar with what's what with intel. :smile:

Zalman HSFs are pretty good. I have several Intel quads and they are using the Zalman CNPS9700LED. My Intel Q8400 has a CNPS9900LED. They are fairly large so need a wide case. eg Antec 902 or Coolermaster CM690. The fan is bolted onto a backplate on the otherside of the motherbaord so you have to dismantle the whole machine when you install it.

I tried one Xigmatek HSF and it was useless as it used push pins instead of bolts. Well if a pin breaks the HSF falls off the CPU so cooling is shite!

yeah, I've been out looking at fans. The intel method is sure an extremely weak sister compared to the amd method, not to mention a major PITA to install them. :-( Sigh.....

@project_admins: I have two ISP's here, Verizon, and Charter. Charter has issues in that I cannot access Prof Caldwell's top-5000 site to report primes. My Verizon account is not working and the earliest that a Verizon Tech will be here is five days away. The problem lies in the wiring from my house box to their big dsl cloud in the sky.

The outcome of this is that I need a version of the old port 9000 with the clean-up work on it that I can just run like before.

[quote=Brucifer;180221]@project_admins: I have two ISP's here, Verizon, and Charter. Charter has issues in that I cannot access Prof Caldwell's top-5000 site to report primes. My Verizon account is not working and the earliest that a Verizon Tech will be here is five days away. The problem lies in the wiring from my house box to their big dsl cloud in the sky.

The outcome of this is that I need a version of the old port 9000 with the clean-up work on it that I can just run like before.[/quote]
Well, right now we have PRPnet G2000, which is loaded with non-top-5000 doublecheck work right now (k=300-400, n=300K-350K). Since (if memory serves) you're already set up for PRPnet, that might be what you're looking for.

BTW: do you know what kind of issues are preventing you from accessing the top-5000 site through Charter? Is it having trouble resolving the domain in DNS? If so, that can probably be fixed by plugging in an alternate DNS server.

In case you want to try accessing directly by IP: primes.utm.edu is 208.87.72.63

it's a route issue. Can't even get it doing the ip address thing. Impossible to talk to the help desk folks as they just work from their canned worksheet, then get pissed, etc. One of the benefits of all this offshore/outsourcing crap in this country. :smile:

@max, I'll see if I still have a prpnet thing on one of these systems to give that a whirl and see what happens. Thnx :smile:

I just build me another quad and I thouht I should try a little bit noprimesleftbehind...
...but I ran into some minor problems here...
I have tried several servers - nplb-gb.1.no-ip.org, noprimeleftbehind.net, llrnet.noprimeleftbehind.net and nplb.ironbits.net - but I only get 'unknown host' every time I tries...
Clearly I'm doing something wrong here... but what ??

[quote=Zachariassen;181950]I just build me another quad and I thouht I should try a little bit noprimesleftbehind...
...but I ran into some minor problems here...
I have tried several servers - nplb-gb.1.no-ip.org, noprimeleftbehind.net, llrnet.noprimeleftbehind.net and nplb.ironbits.net - but I only get 'unknown host' every time I tries...
Clearly I'm doing something wrong here... but what ??[/quote]
First of all, nplb-gb1.no-ip.org (not nplb-gb.1.no-ip.org) is down ATM, so you won't get anything there for now. nplb.ironbits.net (a.k.a. llrnet.noprimeleftbehind.net) is up; I can access it without a problem at [URL]http://nplb.ironbits.net/[/URL]. Are you trying to connect to it through the LLRnet client, or with a web browser?

One thing to note: nplb.ironbits.net, noprimeleftbehind.net, and llrnet.noprimeleftbehind.net all point to the same IP address. While they will lead to different web pages if visited with a browser, the LLRnet client can use any of them interchangeably. (In fact, by extension, any of the other domains hosted on David's server will work in the client, too: ironbits.net, free-dc.org, stats.free-dc.org, etc.)

[quote=mdettweiler;181965]First of all, nplb-gb1.no-ip.org (not nplb-gb.1.no-ip.org) is down ATM, so you won't get anything there for now. nplb.ironbits.net (a.k.a. llrnet.noprimeleftbehind.net) is up; I can access it without a problem at [URL]http://nplb.ironbits.net/[/URL]. Are you trying to connect to it through the LLRnet client, or with a web browser?[/quote]
Thanks for the answer mdettweiler:smile:
I tried to connect though the LLRnet client.
I have changed the 'llr-clientconfig.txt-file like this:
-- server location
server = "nplb.ironbits.net"
-- server = "noprimeleftbehind.net"
-- server = "llrnet.noprimeleftbehind.net"
-- server = "nplb-gb1.no-ip.org"
-- server = "llrnet.ironbits.net"
port = 5000
I've tried all of these servers and the last one was 'nplb.ironbits.net' - still 'unknown host'...
I have to say that I run nplb on two other computers, and on them I use the 'nplb.ironbits.net'-server... works like a charm there...

[quote=Zachariassen;181970]Thanks for the answer mdettweiler:smile:
I tried to connect though the LLRnet client.
I have changed the 'llr-clientconfig.txt-file like this:
-- server location
server = "nplb.ironbits.net"
-- server = "noprimeleftbehind.net"
-- server = "llrnet.noprimeleftbehind.net"
-- server = "nplb-gb1.no-ip.org"
-- server = "llrnet.ironbits.net"
port = 5000
I've tried all of these servers and the last one was 'nplb.ironbits.net' - still 'unknown host'...
I have to say that I run nplb on two other computers, and on them I use the 'nplb.ironbits.net'-server... works like a charm there...[/quote]
Hmm...weird. Just on a hunch, is the computer in question using Linux? Specifically, a Debian- or Red Hat-based distro of Linux?

[quote=mdettweiler;181979]Hmm...weird. Just on a hunch, is the computer in question using Linux? Specifically, a Debian- or Red Hat-based distro of Linux?[/quote]
Yes - Linux Ubuntu 9.04 (64-bit)

[quote=Zachariassen;182025]Yes - Linux Ubuntu 9.04 (64-bit)[/quote]
Okay, try using 98.174.238.119 as the server address and see if that works.

[quote=mdettweiler;182050]Okay, try using 98.174.238.119 as the server address and see if that works.[/quote]
YESSS !!!!
It works :smile:
Thanks mdettweiler - much appreciated - again... thanks :tu:

I run several Linux servers and workstations and I do not have that issue.
I think that your server entry is incomplete as you do not have the "www." precedent in front.
It should be[QUOTE]server = "www.noprimeleftbehind.net"[/QUOTE]

The IP address is not likely to change soon but using the correct server name is preferred in case that should happen.
If you can spare the time to do a test with this server name it would be much appreciated.

[quote=AMDave;182056]I run several Linux servers and workstations and I do not have that issue.
I think that your server entry is incomplete as you do not have the "www." precedent in front.
It should be

The IP address is not likely to change soon but using the correct server name is preferred in case that should happen.
If you can spare the time to do a test with this server name it would be much appreciated.[/quote]
It works for me with or without the "www.", so I don't think that's the issue. Rather, I think Zachariassen was hit by a rather annoying bug that occurs on Red Hat- or Debian-based distros when the LLRnet binary was not originally compiled on one of those distros, where the client isn't able to resolve DNS addresses. The solution is to either compile it yourself, or use someone else's binary that was compiled on one of those distros. An alternative workaround is to use the IP address instead, which worked for Zachariassen in this case.

I know that I compiled the client myself a while back (on Ubuntu, which is Debian-based so it works for those systems) and posted it for download. The only disadvantage of that build is that it doesn't have the GUI enabled (since I don't have the GUI development libraries installed on my machine). Zachariassen, if you don't mind not having a GUI, I can see if I can dig up a link to download my Debian build, so you don't have to use the IP address.

Regarding the IP address workaround: there shouldn't be a problem with that changing, since David has a static IP. Of course, this workaround wouldn't work too well for the nplb-gb1.no-ip.org servers, which are on a dynamic IP that can change at any time without warning.

[quote=AMDave;182056]I run several Linux servers and workstations and I do not have that issue.
I think that your server entry is incomplete as you do not have the "www." precedent in front.
It should be

The IP address is not likely to change soon but using the correct server name is preferred in case that should happen.
If you can spare the time to do a test with this server name it would be much appreciated.[/quote]
I have tried with <www.> in front - and it didn't work... sorry.

[quote] Zachariassen, if you don't mind not having a GUI...[/quote]I don't mind mdettweiler - I always run it in terminal like this : sudo ./llrnet -no-gui

Edit /etc/hosts
Put the IP and hostname in there
IP addy hostname
IP addy hostname
Save it and your done ;)

Help request:

I have an Intel Core 2 Duo E6300 running at stock 1.86GHz. It has 2GB DDR2 RAM. OS is Windows XP Pro SP3. Its been running stably for ages, couple of years at 24/7. It is running a network version of AVG plus the netDIMES project and BOINC FreeHAL@Home.

When I run NPLB from two separate folders (sub-directories) off the C:\ drive the two instances run OK. However, the GUI won't open on the second instance, only on the first (I think that's the right order). I have added the extra port and now see both instances crunching on the GUIport 6999 and GUIport 6998.

How do I get the GUI to display regardless of which Systemtray icon I right click on and select 'New GUI window'?

[quote=vaughan;184396]Help request:

I have an Intel Core 2 Duo E6300 running at stock 1.86GHz. It has 2GB DDR2 RAM. OS is Windows XP Pro SP3. Its been running stably for ages, couple of years at 24/7. It is running a network version of AVG plus the netDIMES project and BOINC FreeHAL@Home.

When I run NPLB from two separate folders (sub-directories) off the C:\ drive the two instances run OK. However, the GUI won't open on the second instance, only on the first (I think that's the right order). I have added the extra port and now see both instances crunching on the GUIport 6999 and GUIport 6998.

How do I get the GUI to display regardless of which Systemtray icon I right click on and select 'New GUI window'?[/quote]
I'm not sure if it's possible to make one LLRnet client point to another's GUI. Maybe modifying the second client's gui.bat to cd to the first client's directory before launching the GUI would work? (I don't know if the client actually utilizes the gui.bat for the "New GUI Window" button, or has the link hardcoded, so this may or may not work.)

Ah. That's tricky.

I could always get one GUI to monitor all of the clients (including other machines and OSs across the network).

I always got the start up sequence to kick that one off first so I always knew it was 'the one on the left'.

I don't recall if I ever pushed to get both GUIs on the same Win-box to look at each other, though.

I have some instructions for the GUI connections that I wrote up a couple of years ago when Brucifer and I were thrashing out primes for Lars Dauche's project. I'll dig them up and re-work them with vaughan, see if we can get it to work and if so we'll get them posted up.

/me kicks the tyres on his new VM server ... oh yeah! lets roll.

vaughan

I recall that I have tested this before.

can you double check that both clients on that machine were started without the no-gui flag and both client config files have "GUIremote = 1" uncommented?

There's no other reason for it not to work.

Be careful not to 'control' them both from the GUIs at the same time.
You can cause the client to crash if it gets conflicting instructions from 2 GUIs at the same time.

[QUOTE=AMDave;184427]...
can you double check that both clients on that machine were started without the no-gui flag and both client config files have "GUIremote = 1" uncommented?
[/QUOTE]
Yes, it is uncommented in both C:\NPLB and C:\NPLB1 llr-clientconfig.txt files.

Also I'm using the left systray icon but that;s the one that fails to display. If I then try the right hand systray icon the GUI window opens. Then I add the 6998 client and its "normal" ie working as it should.

Same thing happens on Lisa's XPS laptop so I just run one instance on that machine. Anyways, LLRnet makes it run very hot; at least 6deg C hotter than the most aggressive BOINC applications. I have to start two instances, run the GUI from the RHS one then let the C:\NPLB1 end after one test. This lets the laptop cool down from near 90C to the mid 70s.

pls zip and email me the client config and the bat files.

Dave - files sent by e-mail. Thanks.

Gotcha. Replied. Work in progress ... :geek:

@Gary/Max -- I fired off a couple of systems on the small primes last night. It looked to run normal on one machine on 3 of 4 cores. On the 4th core the count-up display was looking all screwed up, and running like a rocket ship. The time per itteration was like .02ms on the other 3 cores, so those things were running fast. So I tried it on a second system to see what it was like, and that one acted like the 4th core on the first system. So I shut the stuff off. I don't have a slow system to run the stuff on to see if it's going normal or not. But anyway you need to check your last night's results from "bruce".

[quote=Brucifer;188917]@Gary/Max -- I fired off a couple of systems on the small primes last night. It looked to run normal on one machine on 3 of 4 cores. On the 4th core the count-up display was looking all screwed up, and running like a rocket ship. The time per itteration was like .02ms on the other 3 cores, so those things were running fast. So I tried it on a second system to see what it was like, and that one acted like the 4th core on the first system. So I shut the stuff off. I don't have a slow system to run the stuff on to see if it's going normal or not. But anyway you need to check your last night's results from "bruce".[/quote]
What exactly do you mean by the count-up display "running like a rocket ship" on the 4th core? If the machines are Core 2 Quads, then the 4th core can be expected to have a little bit slower iteration time than the other three due to memory bottlenecking inherent in the design of the C2Q. Of course, if you meant something different, then that doesn't apply.

you could watch the first three cores counting up through the .00 section of the count, but the 4th core didn't show the .00 count, it just counted up li 10.00%, 20.00%, 30.00% etc. Just accelerating through the tests way too fast compared to the first three cores.

[quote=Brucifer;188948]you could watch the first three cores counting up through the .00 section of the count, but the 4th core didn't show the .00 count, it just counted up li 10.00%, 20.00%, 30.00% etc. Just accelerating through the tests way too fast compared to the first three cores.[/quote]
Hmm...strange, indeed. However, I just thought of something: it might simply be that the first three cores were still doing work from the latter end of one k (near n=425K, where tests would take a while), whereas the last one was doing stuff near the beginning of the next k, near n=50K, which would be very small tests. For tests that small, I can easily see how things would be going so fast that the client doesn't even try to keep up with each and every 1% increment but rather updates in 10% increments.

LLR(net) doing a test in 10000 iterations per screen output by default.

so for port GB8000 (where n~1M) every 10000 it. will display in about 1% steps.

at n~100k this would be 10% per step.

check the n-range for every core (see lresults.txt).

[quote=mdettweiler;188949]Hmm...strange, indeed. However, I just thought of something: it might simply be that the first three cores were still doing work from the latter end of one k (near n=425K, where tests would take a while), whereas the last one was doing stuff near the beginning of the next k, near n=50K, which would be very small tests. For tests that small, I can easily see how things would be going so fast that the client doesn't even try to keep up with each and every 1% increment but rather updates in 10% increments.[/quote]

It would be near n=250K not n=425K.

[quote=Brucifer;188948]you could watch the first three cores counting up through the .00 section of the count, but the 4th core didn't show the .00 count, it just counted up li 10.00%, 20.00%, 30.00% etc. Just accelerating through the tests way too fast compared to the first three cores.[/quote]

As Max implied, I'm sure this is OK. The timing difference between the smallest and largest exponent for the 12th drive (port G7000) is fairly extreme. A test at the high end of the range should take about 25 times as long as a test at the low end of the range. I'm fairly certain that you just happened to encounter a change in k-value amongst the cores.

Give it another shot. If the same thing happens, wait about half an hour and check it again. All cores should have about the same timings at that point.


Gary

Can anyone explain what happened to the stats yesterday?

Over the last few days bok's stats have shown my results:
Oct 11. . . .Oct 10. . . .Oct 9. . . . .Oct 8. . . . .Oct 7. . . . .Oct 6. . . .Oct 5
21,470. . . .12,276. . . .16,889. . . .16,001. . . .16,464. . . .16,809. . . .17,454

Why was Oct 10 so low and Oct 11 hasn't even finished yet?

those stats from FreeDC project NPLB will not updated constantly every full hour and the timezone is different to the NPLB-stats page.

so when that page updates at say 03:00 AM but not for the last 5 hours, the points from 2 hours will be shown at the new day!

Because sometimes another errant non-BOINC project causes the non-BOINC updater to get stuck and Bok has to kick it hard in butt with a steel capped boot.

So, you are seeing several updates during the day on Oct 10, then the errant project hangs the updater ... until it feels the pain of the size 12 1/2 steel capped boot in the butt. In the mean time, the stats roll-over occurred so any new updates are recorded against the next day, 11 Oct. So the next time the stats updater picks up you NPLB stats, it doesn't play a game of catch-up but goes straight for the win and grabs your latest total (as it does for every observation). The increase gets recorded against the date of 11 Oct and so do the rest of your updates for 11 Oct which results in a higher than average score, but is the exact offset of the increases that were missed on the previous day.

so...it is a time based observation with a frequency greater than once per 24 hours. Hence the drop and the spike when the updater gets stuck and 'punted' either side of the rollover. :smile:

Hope that helps.

Anyone happen to know where Geoff Reynolds and SR2Sieve are hanging out now?

This link is no longer valid.

[url]http://www.geocities.com/g_w_reynolds/sr2sieve/[/url]

he is changing to google-pages [url]http://sites.google.com/site/sr2sieve/[/url]
but not all available yet there!

see his post here: [url]http://www.mersenneforum.org/showpost.php?p=194042&postcount=523[/url]

[QUOTE]he is changing to google-pages [URL]http://sites.google.com/site/sr2sieve/[/URL]
but not all available yet there!

see his post here: [URL="http://www.mersenneforum.org/showpost.php?p=194042&postcount=523"]http://www.mersenneforum.org/showpos...&postcount=523[/URL][/QUOTE]

Thanks Karsten

So what is the difference in the points formula between the regular database score computation and the prpnet score computation? I assume that the prpnet stuff is going to get added to the regular database point system?

[quote=Brucifer;197224]So what is the difference in the points formula between the regular database score computation and the prpnet score computation? I assume that the prpnet stuff is going to get added to the regular database point system?[/quote]
Yes, that's the plan. The only reason why it's not in there yet is because there's one last tweak Dave needs to do first, but once that's all set everything prior to that point will be added in retroactively.

The points will be calculated by the same formula; actually, at this point, the DB doesn't even know the difference between LLRnet and PRPnet results. :smile: We may at some point in the future have it do more advanced stuff with PRPnet results (which include a bit more information--such as the client name as well as username--than do LLRnet results), though none of that's even on the drawing board yet.

So why are they different to begin with???

[quote=Brucifer;197235]So why are they different to begin with???[/quote]
Er...I'm not sure what you mean. There's not really a "difference" per se; it's only that PRPnet records more information with each results than LLRnet does. However, the DB isn't set up to utilize that extra information at this time, since it was designed initially for LLRnet results. That's OK, as the extra information recorded by PRPnet is just extra icing on the cake, so to speak; what we end up doing is converting the results to LLRnet format with a script and then importing them to the DB.

[quote=mdettweiler;197265]Er...I'm not sure what you mean. There's not really a "difference" per se; it's only that PRPnet records more information with each results than LLRnet does. However, the DB isn't set up to utilize that extra information at this time, since it was designed initially for LLRnet results. That's OK, as the extra information recorded by PRPnet is just extra icing on the cake, so to speak; what we end up doing is converting the results to LLRnet format with a script and then importing them to the DB.[/quote]
i think he wants to know why we are changing to prpnet as llrnet has been working well for us:)

Actually :-) what I'm curious to is the number of points granted in the prpnet G3000 stats per pair returned, as it looks like 460+ per.... versus what is granted for a returned pair for the regular database points per returned pair.

As for prpnet, I like it personally. I ran it earlier when the initial beta round was going and it was stable for me. I've got it running on 12 cores now, granted they are mainly slow ones, but none the less they crunch out pairs. :-) I intend to add some fast cores to the prpnet and shift some off the port 7000 stuff and see if I can manage to find some big primes.

[quote=Brucifer;197277]Actually :-) what I'm curious to is the number of points granted in the prpnet G3000 stats per pair returned, as it looks like 460+ per.... versus what is granted for a returned pair for the regular database points per returned pair.

As for prpnet, I like it personally. I ran it earlier when the initial beta round was going and it was stable for me. I've got it running on 12 cores now, granted they are mainly slow ones, but none the less they crunch out pairs. :-) I intend to add some fast cores to the prpnet and shift some off the port 7000 stuff and see if I can manage to find some big primes.[/quote]
PRPnet:[code]How are user stats computed?
The score is computed as such:
(decimal length of the candidate / 10000) ^ 2
[/code]NPLB:[quote]Participants initially ordered by Pair Score (n^2/160e9) descending.[/quote]I'm sure an approximate (should be quite close, but not precise, as NPLB only considers the n, allowing a small variation in decimal length resulting in the same score) conversion formula could be worked up. I'll see about trying to do that...

I think the conversion factor is about 144.9905. That is, PRPnetScore ~= 144.9905 * NPLBScore, or PRPnetScore / 144.9905 ~= NPLBScore. Here's how I got that: (where c is the conversion factor)[code](digits / 10000)^2=c*(n^2/160e9)
digits^2 / 1e8 = c*n^2/1600e8
digits^2 = c*n^2/1600
*
(n*log_10(2))^2 = c*n^2/1600
log_10(2)^2*n^2 = c*n^2/1600
log_10(2)^2*1600 = c
144.9905... ~= c[/code]* Here I make the assumption that digits=n*log_10(2), which is decently close, but not perfect. Because of the differing algorithms, an exact comparison can not be made.

Edit: I noticed that I used log_2(10) where I should have used log_10(2), resulting in 17656.33 instead of 144.9905. It's fixed now, and is producing a number that matches Karsten's numbers below and makes more sense: 144.9905. This paragraph was from before I fixed it:
Hmm, this seems wrong...it would mean that your 1080 tests [URL="http://nplb-gb1.no-ip.org:3000/user_stats.html"]from the 5th drive[/URL], with a (PRPnet) score of 536349, are only worth about 30.377 (NPLB) points, but from looking at [URL]http://www.noprimeleftbehind.net/index.php?content=user_pairs&page=2[/URL] that seems more like 6 decently sized candidates. Either I went very wrong somewhere (I don't see where), or one or both of the scoring formulas I assumed is not accurate. (I have verified that in the newest released PRPnet source, that's what it really does, so that really just leaves an error on my part or NPLB's n^2/160e9 formula being wrong) Can anyone see where I went wrong, and can one of the NPLB admins verify what the formula is?

the scores from G3000 are not correct as shown [url=http://nplb-gb1.no-ip.org:3000/user_stats.html]here[/url]:

G3000 for n~740k:

496 points / pair with PRPnet
3.42 points / pair with LLRnet

Ah, I see where the confusion's coming from. That's PRPnet's built-in stats, which are NOT official NPLB stats. When we get DB import turned on for PRPnet servers, the results will be imported and the stats will be calculated according to the same formula we've always used for LLRnet.

Note the disclaimer I put in our PRPnet servers thread (which has actually been there from the beginning):
[quote]*User stats are calculated directly by the server, and may not match up with our official stats at [URL]http://www.noprimeleftbehind.net/[/URL]. These stats are primarily useful for comparing different users on that specific server, and use a completely different formula for calculating score. They are provided for informational purposes only. :smile:[/quote]

People will be hounding us pretty fast on this Max: Can you give us an approximate time frame on when the results and primes scores will be interfaced with the main NPLB scores? People aren't going to want to go very long waiting for that. Thanks.

Thanks for the calculations Tim. For everyone's reference, Tim's multiplier for the difference between PRPnet scores and official NPLB scores is for "entertainment" only. In no way will the score calculated by the server be used for the official NPLB score. At this moment, nothing found on a PRPnet server is reflected on the pages.


Gary

[quote=gd_barnes;197301]People will be hounding us pretty fast on this Max: Can you give us an approximate time frame on when the results and primes scores will be interfaced with the main NPLB scores? People aren't going to want to go very long waiting for that. Thanks.[/quote]
I'm not sure. Currently I'm waiting to hear back from Dave; I'm wondering if he's away for Thanksgiving or something. (Do they do Thanksgiving in Australia? :smile:)

Well at any rate, thanks to all for your explanations!!! :smile:

Problems cropping up. On port 7000, my clients are intermittantly handing up when trying to send results to the server, and they just hang forever. This started showing up yesterday, and has been pretty heavy today so far. So I have cut back on the number of clients hitting 7000 and have put a bunch onto the prpnet server. So will see how that goes. ???????

[quote=Brucifer;197387]Problems cropping up. On port 7000, my clients are intermittantly handing up when trying to send results to the server, and they just hang forever. This started showing up yesterday, and has been pretty heavy today so far. So I have cut back on the number of clients hitting 7000 and have put a bunch onto the prpnet server. So will see how that goes. ???????[/quote]
Hmm...strange. I don't see anything in the log files indicating that the server crashed and had to be restarted. However, I do see that there are a bunch of [url=http://nplb-gb1.no-ip.org/llrnet/rejected_7000.txt]rejected results[/url] that showed up today, from about 1:45 to 2:00 AM server time (CST).

Hey, wait a minute...I just had a thought. Gary, when did you do the UPS run-down test? Was that by chance around 1:45 AM?

That's exactly when I did it. I restarted it with that same command that you recommended when I changed the servers over to another machine. Why would it cause Bruce to have problems?

I don't know what has happened with the port 7000 stuff. I have quit running it. I'd go checking my systems and on a quad, 3 out 4 cores would be locked up at the point of trying to send to the server OR sucking down new work.... ????? Really has me buffaloed as I have done nothing to these systems and they were all working just fine for days before. And they work just fine on the prpnet server. Ya got me! ????

[quote=gd_barnes;197422]That's exactly when I did it. I restarted it with that same command that you recommended when I changed the servers over to another machine. Why would it cause Bruce to have problems?[/quote]The rejected results would be explained by the fact that the server "forgot" everything that happened since it last saved jobList.txt (which is saved every 5 minutes), due to the abrupt shutdown. I still don't see how that could cause Bruce's problems, though. Most likely they're unrelated.
[quote=Brucifer;197439]I don't know what has happened with the port 7000 stuff. I have quit running it. I'd go checking my systems and on a quad, 3 out 4 cores would be locked up at the point of trying to send to the server OR sucking down new work.... ????? Really has me buffaloed as I have done nothing to these systems and they were all working just fine for days before. And they work just fine on the prpnet server. Ya got me! ????[/quote]
I'm not sure what's causing these problems. Do you by chance know when this first started happening? If it started around 1:45-2 AM CST on 11/30, then there might be some connection to the restart. Otherwise, though, I have absolutely no idea.

If it did start around the restart, then it could just be that the clients goofed because they couldn't connect to the server (or, in some cases, may have been cut off in the middle of talking to the server, which can cause all sorts of crazy problems). My internet connection cuts out on a regular basis and as such I've seen similar issues with my own clients. They'd hang if they were interrupted in the middle of a communication.

This stuff all started after the last batch of pairs was loaded in. Prior to that point in time things were running along just fine. Then I notice hangups starting, few at first, but growing progressively worse. I also noticed that there was a rejected pairs list growing on the server that hadn't happened before. Then we had the UPS event, and then things just reached the point that I'd have to spend all my time going back and re-starting systems. And that's when I just said heck with it and put it all on prpnet and some other stuff and quite the 7000 altogether.

Another thought Max, would be for you to set up a linux prpnet server for 7000, load it up with a huge load of stuff, tar it up so I could suck it down and run it here at home on my net, and then when it goes dry, tar it up and give it back to you.

[quote=Brucifer;197485]This stuff all started after the last batch of pairs was loaded in. Prior to that point in time things were running along just fine. Then I notice hangups starting, few at first, but growing progressively worse. I also noticed that there was a rejected pairs list growing on the server that hadn't happened before. Then we had the UPS event, and then things just reached the point that I'd have to spend all my time going back and re-starting systems. And that's when I just said heck with it and put it all on prpnet and some other stuff and quite the 7000 altogether.

Another thought Max, would be for you to set up a linux prpnet server for 7000, load it up with a huge load of stuff, tar it up so I could suck it down and run it here at home on my net, and then when it goes dry, tar it up and give it back to you.[/quote]
If we were going to do that, we might as well just make it into a public PRPnet server to replace the current G7000. Since this one's doing non-top-5000 work, not having the email notification working isn't particularly crucial, and as such, it would work out OK to switch it to PRPnet now.

So has AMDave ever surfaced?

[quote=Brucifer;197914]So has AMDave ever surfaced?[/quote]
Yes, he's back now. Hopefully we'll have the DB import stuff online within a few days.

So how goes the war on getting the manual reservation stuff added to the scoring system???

[quote=Brucifer;200461]So how goes the war on getting the manual reservation stuff added to the scoring system???[/quote]
No news yet. Dave, how are things coming on that front? Also, what about the doublecheck database--any progress with that?

[quote=Brucifer;200461]So how goes the war on getting the manual reservation stuff added to the scoring system???[/quote]

I am waiting for it since Jan 08.

Question:
Where did [url]http://www.noprimeleftbehind.net/[/url] go?

When I clicked on my bookmark today it took me to a page that says:

Error connecting to database

[quote=vaughan;203815]Question:
Where did [URL]http://www.noprimeleftbehind.net/[/URL] go?

When I clicked on my bookmark today it took me to a page that says:

Error connecting to database[/quote]

I posed that question in the LLRnet servers thread. There was an Email exchange between Max and David (Ironbits). The stats database may be in the process of being moved. Hopefully Max can answer.

[quote=gd_barnes;203816]I posed that question in the LLRnet servers thread. There was an Email exchange between Max and David (Ironbits). The stats database may be in the process of being moved. Hopefully Max can answer.[/quote]
Hmm...I don't think Dave actually took the DB offline yet. I'll ask him.

Migration not in progress yet.
There should be no database outage during the migration anyway.
That was an unplanned outage.
I investigated and found the database service was stopped.
I restarted it and it appears to be ok.
Thanks for the heads-up Max.