Sierp base 6 - team drive #3
 

We are starting Conjectures 'R Us (CRUS) team drive #3 for Sierp base 6 starting at n=30K and continuing until all k's have a prime or we just get tired of searching! :smile: Initially included in the drive were 50 of the remaining 56 k's that needed a prime. We have now found 45 primes so there are 11 k's remaining, all of which are in the drive. See the Sierp base 6 web page for the k's remaining.

The file has been sieved for n=400K-1M. The sieve depth is P=130T. No more sieving will be needed for this range. Many thanks to Lennart for sieving P=25T-130T.

We will be running the remainder of the drive on CRUS PRPnet server port 1400. The server will hand out the final range of n=700K-1M.

Instructions for running a PRPnet server and download links can be found [URL="http://www.mersenneforum.org/showthread.php?t=12223"][COLOR=#000080]here[/COLOR][/URL]. The info. specific to this server that needs to be entered into your prpclient.ini file is:

server=G1400:100:1:noprimeleftbehind.net:1400

Server info.:

[B]CRUS PRPnet server (updated 2014-06-11 09:00 GMT):[/B]
maintained by mdettweiler on gd_barnes machine
Short identification: G1400
server: noprimeleftbehind.net
port: 1400
Sierp base 6: 16 k's remaining to n=1M
n-range: 700K-1M
currently processing at n= [COLOR=#0000ff](complete)[/COLOR]
Server summary: [URL="http://noprimeleftbehind.net:1400/all.html"][COLOR=#000080]http://noprimeleftbehind.net:1400/all.html[/COLOR][/URL]

Primes found from the team drive plus all individual efforts for Sierp base 6 for n > 30K (red indicates top-5000 primes):
[code]
prime found by
[COLOR=red]139413*6^1279992+1 BOINC
[/COLOR][COLOR=red]33706*6^910462+1 mdettweiler[/COLOR]
[COLOR=red]125098*6^896696+1 mdettweiler[/COLOR]
[COLOR=red]31340*6^833096+1 mdettweiler[/COLOR]
[COLOR=red]59506*6^780877+1 Lennart[/COLOR]
[COLOR=red]10107*6^559967+1 Lennart[/COLOR]
[COLOR=red]113966*6^511831+1 MyDogBuster[/COLOR]
[COLOR=red]172257*6^349166+1 mdettweiler[/COLOR]
[COLOR=red]121736*6^298935+1 mdettweiler[/COLOR]
[COLOR=red]124221*6^278517+1 mdettweiler[/COLOR]
[COLOR=red]51255*6^264988+1 mdettweiler[/COLOR]
[COLOR=red]61340*6^262453+1 mdettweiler[/COLOR]
[COLOR=red]26375*6^230349+1 Lennart[/COLOR]
[COLOR=red]68195*6^215736+1 Flatlander[/COLOR]
[COLOR=red]108527*6^185267+1 Lennart[/COLOR]
[COLOR=red]74612*6^173463+1 Lennart[/COLOR]
71301*6^172271+1 Lennart
145076*6^149946+1 Lennart
123285*6^147200+1 Lennart
157473*6^131124+1 MyDogBuster
36772*6^126672+1 MyDogBuster
118147*6^122688+1 MyDogBuster
98860*6^119849+1 MyDogBuster
72785*6^118347+1 Beyond
66520*6^96685+1 Beyond
135717*6^85064+1 Beyond
33676*6^82971+1 IronBits
125803*6^80211+1 IronBits
154797*6^77401+1 gd_barnes
90546*6^68012+1 mdettweiler
60017*6^64043+1 gd_barnes
129233*6^64027+1 gd_barnes
20721*6^63016+1 gd_barnes
98636*6^58201+1 gd_barnes
33298*6^57853+1 gd_barnes
151003*6^56371+1 gd_barnes
135500*6^53522+1 gd_barnes
27868*6^52504+1 gd_barnes
93241*6^50452+1 gd_barnes
86253*6^47459+1 gd_barnes
37815*6^39336+1 mdettweiler
18115*6^39155+1 gd_barnes
36028*6^37879+1 gd_barnes
137047*6^31754+1 gd_barnes
59606*6^30490+1 gd_barnes
[/code]Status:
[code]
n-range tested by Status # of primes
1.5M-2M BOINC in progress
1M-1.5M BOINC complete 1
700K-1M PRPnet (G1400) complete 4
600K-700K rogue complete 0
540K-600K Lennart complete 1
480K-540K MyDogBuster complete 1
460K-480K Puzzle-Peter complete 0
400K-460K MyDogBuster complete 0
390K-400K Mini-Geek complete 0
325K-390K mdettweiler complete 1
324K-325K henryzz complete 0
322K-324K mdettweiler complete 0
316K-322K Batalov complete 0
314K-316K henryzz complete 0
240K-314K mdettweiler complete 4
228K-240K Lennart complete 1
214K-228K Flatlander complete 1
205K-214K Lennart complete 0
200K-205K Flatlander complete 0
166K-200K Lennart complete 3
154K-166K Flatlander complete 0
150K-154K Lennart complete 0
100K-150K PRPnet (G3000) complete 7
80K-100K LLRnet (IB6) complete 4
70K- 80K gd_barnes complete 1
67K- 70K mdettweiler complete 1
45K- 67K gd_barnes complete 10
38K- 45K mdettweiler complete 1 (plus 1 other)
30K- 38K gd_barnes complete 3
[/code][B]All sieved files have been tested. Eventually we will begin sieving a higher n-range.[/B]

Let's blow away those stubborn k's! :grin:


Gary

I'll reserve the first range of n=30K-38K to get us started.

Right out of the gate...

59606*6^30490+1 is prime

Now down to 55 k's

Reserving 38K-45K. :smile:

137047*6^31754+1 is prime

Found PRP by LLR, proven with PFGW:

37815*6^39336+1 is prime! :banana:

BTW: At what n-value does this drive enter top-5000 territory?

[quote=Anonymous;124587]Found PRP by LLR, proven with PFGW:

37815*6^39336+1 is prime! :banana:

BTW: At what n-value does this drive enter top-5000 territory?[/quote]


Assuming a now-median k-value of k=90546 and the current 5000th place prime of 26267432109*2^333333-1 as of 3:10 AM EST US on 2/3/08, it would be at n=128958.

Of course it might be a smidge higher when we get there. It might actually be n=128960 or something.


:smile: Gary

[quote=gd_barnes;124642]Assuming a now-median k-value of k=90546 and the current 5000th place prime of 26267432109*2^333333-1 as of 3:10 AM EST US on 2/3/08, it would be at n=128958.

Of course it might be a smidge higher when we get there. It might actually be n=128960 or something.


:smile: Gary[/quote]
Well, it looks like we have our work cut out for us, then! :smile:

36028*6^37879+1 is prime

30K-38K complete; 3 primes already reported

reserving 45K-50K

86253*6^47459+1 is prime

45K-50K complete; 1 prime already reported

reserving 50K-58K (2 files)

93241*6^50452+1 is prime

38K-45K complete, one prime already reported. lresults attached. :smile:

27868*6^52504+1 is prime

135500*6^53522+1 is prime

Edit: 50K-54K complete; 3 primes

151003*6^56371+1 is prime

33298*6^57853+1 is prime

The very nice range of n=50K-58K is now complete with 5 primes found!

Reserving n=58K-64K (2 files).


Gary

98636*6^58201+1 is prime

reserving n=64K-67K for some filler work

The "filler work" started out quite prolific:

129233*6^64027+1 is prime

60017*6^64043+1 is prime

Both in the first 27 tests of the 64K-67K range and within 10 tests of each other! :smile:

n=58K-61K complete; 1 prime

n=64K-67K filler work complete; 2 primes

still working n=61K-64K

Taking 67K-70K. :smile:

20721*6^63016+1 is prime

Anon, you can remove k=20721 from your file if you want to.

[quote=gd_barnes;127864]20721*6^63016+1 is prime

Anon, you can remove k=20721 from your file if you want to.[/quote]
Okay, done. :smile:

90546*6^68012+1 is prime!

(Found PRP with PRP, proven with PFGW.) :smile:

reserving 70k-72k

[quote=jaat;128013]reserving 70k-72k[/quote]
Welcome to Conjectures 'R Us, jaat! :smile: Hope you find lots of primes!

61K-64K complete; 1 prime

Reserving 72K-76K (2 files)

Edit: more files will be posted later today.

Files have been posted up to n=100K. That's all that I have sieved. It looks like more sieving will be needed in the near future. We'll probably want to sieve n=100K-400K this time around and break off n=100K-200K when at an appropriate depth.

This is a very good base with plenty of primes but also with plenty of stubborn k's to keep it interesting.


Gary

67K-70K done, results attached. One prime previously reported. :smile:

Reserving 76K-80K (2 files) to keep my machine entertained while I'm on vacation. :smile:

Taking 80K-86K for IronBits' new CRUS LLRnet server. :smile:

First prime from the new LLRnet server!
 

user=Free-DC_IronBits
[03/20/08 06:27:42]
125803*2^80211-1 is prime! Time : 437.0 sec.

Apparently IronBits' LLRnet server has been configured for base 2 Riesel display format (oops!) so the primes show up funny in the results file and on his website. Oh well, I can easily fix the results files with a search/replace though. :smile: (That's probably better than changing the setting now, after we've already got it going.)

Thus, the real prime is [b]125803*6^80211+1[/b]. I verified it with PFGW. :smile:

Either Gary or I will remove the prime from all the unreserved files shortly...

Excellent on the new server! Thanks for arranging that Anon. :smile:

Meanwhile:

n=72K-76K complete; no primes

I'm still working on n=76K-80K

k=125803 has now been removed from all posted files.

I'm leaving k=125803 in my range in case there is a lower prime.

Anon, I also have files for n=80K-82K, 82K-84K, and 84K-86K that now have the k removed. If you want to mess with that for the server, let me know. If not, no big deal.


Gary

[quote=gd_barnes;129335]k=125803 has now been removed from all posted files.

I'm leaving k=125803 in my range in case there is a lower prime.

Anon, I also have files for n=80K-82K, 82K-84K, and 84K-86K that now have the k removed. If you want to mess with that for the server, let me know. If not, no big deal.


Gary[/quote]
No, I think we'll leave the k in for the server files. Removing k's in the middle of a range creates too much of a hassle when processing the results, as I found out the hard way with some of the earlier Riesel Base 16 results. :smile:

[quote=Anonymous;129318]
Apparently IronBits' LLRnet server has been configured for base 2 Riesel display format (oops!) so the primes show up funny in the results file and on his website.[/quote] Picky picky ;p fixed. :smile:

154797*6^77401+1 is prime

n=76K-78K is now complete. Still working on n=78K-80K. After completing that, I'll move the core over to the Sierp base 6 LLRnet server.

Better hurry then, I have 8 cores on it full time :)

[quote=IronBits;129542]Better hurry then, I have 8 cores on it full time :)[/quote]

Holy smokes...excellent! No problem. If it starts to run low, just load n=86K-90K into it. If you do that now, hold off a few mins. and I'll remove k/n pairs for the k-value that I just found a prime for.

Edit: k=154797 has now been removed from all posted files.


Gary

Port [B]6[/B] has 2006 lines in knpairs.txt
File size of 25,050
first k/n pair 26375 80001
last k/n pair 125803 85997
You can keep an eye on it here: [URL]http://crus.ironbits.net[/URL] :smile:

When you want to put more on it, a simple message like this works for me ;)

Hey IronBits, Port 6 CRUS add the following (enclosed .txt) attachment to the end of knpairs.txt
and I'll get-er-done :wink: or you can send it to me in an email, what ever works best for you guys.

[quote=IronBits;129547]Port [B]6[/B] has 2006 lines in knpairs.txt
File size of 25,050
first k/n pair 26375 80001
last k/n pair 125803 85997
You can keep an eye on it here: [URL]http://crus.ironbits.net[/URL] :smile:

When you want to put more on it, a simple message like this works for me ;)

Hey IronBits, Port 6 CRUS add the following (enclosed .txt) attachment to the end of knpairs.txt
and I'll get-er-done :wink: or you can send it to me in an email, what ever works best for you guys.[/quote]
Here's 86K-90K, you can go ahead put it in the server. :smile:

Thanks, Done! I renamed it to knpairs_6_86k-90k.txt :wink:

Went from
Port [B]6[/B] has 1756 lines in knpairs.txt
File size of 21,935
first k/n pair 31340 80612
last k/n pair 125803 85997

to
Port [B]6[/B] has 3696 lines in knpairs.txt
File size of 46,098
first k/n pair 31340 80612
last k/n pair 135717 89996

n=78K-80K complete; no primes

Now running 2 cores on the server. Will add a 3rd on Friday.

Taking 90K-100K for IronBits' port 6 LLRnet server. :smile:

[quote=jaat;128013]reserving 70k-72k[/quote]

Jaat,

Can you give us a status update on your range?


Thanks,
Gary

[QUOTE=gd_barnes;130098]Jaat,

Can you give us a status update on your range?


Thanks,
Gary[/QUOTE]

Sorry guys, I have been very slow with the progress as I did not estimate the time this range will take. I have started the range but I will unreserve it since it is of low n and hence any primes here will save a lot of time. I have completed till 36772*6^70074+1 and no primes till then. Also, I can email the residues if need be.

jaat

[quote=jaat;130150]Sorry guys, I have been very slow with the progress as I did not estimate the time this range will take. I have started the range but I will unreserve it since it is of low n and hence any primes here will save a lot of time. I have completed till 36772*6^70074+1 and no primes till then. Also, I can email the residues if need be.

jaat[/quote]


OK, thanks for the update. Yeah, the non-powers-of-2 bases take much longer for the same size. Also, the fact that it is a larger base, n=70K is the same as n=~181K base 2, can make it even more misleading because it LLR's much slower than n=181K base 2.

No big deal about the residues since it apparently is < 70 tests.

Anon, I just had a machine finish another manual range at NPLB, I'll reserve n=70K-72K here and put it on that machine.

IronBits, I'll just leave the 2 CPU's on your server until I finish up this range.


Gary

2 new primes just found by the port 6 LLRnet server! :w00t:

33676*6^82971+1 is probably prime! (IronBits)
135717*6^85064+1 is probably prime! (Beyond)

I will verify these with PFGW shortly...

Edit: First one proven. Still working on the second one. :smile:

Excellent! And I just noticed that we're at n=~92K already. Smokin! :smile:

Meanwhile...Sierp base 2 odd-n just picked off a nice big one...see another thread.

[quote=Anonymous;130321]2 new primes just found by the port 6 LLRnet server! :w00t:

33676*6^82971+1 is probably prime! (IronBits)
135717*6^85064+1 is probably prime! (Beyond)

I will verify these with PFGW shortly...

Edit: First one proven. Still working on the second one. :smile:[/quote]
Both primes now proven with PFGW. Gary, I'll leave it up to you to remove the k's from files. :smile:

[quote=Anonymous;130325]Both primes now proven with PFGW. Gary, I'll leave it up to you to remove the k's from files. :smile:[/quote]

And how might I do that? There are no files! :smile:

I'll remove them from my sieve files for n=100K-400K, though. It'll be up to you to remove them from the 'master' sieved file since I only have one with factors removed to P=500G.

[quote=gd_barnes;130326]And how might I do that? There are no files! :smile:

I'll remove them from my sieve files for n=100K-400K, though.[/quote]
Oh, duh duh duh! *bangs head* I don't know how I forgot that! (Okay, maybe I do: It's 2:25 AM my time right now. :smile:)

I'll remove the k's from the team-sieve file, and post the updated file shortly. :smile:

n=70K-72K complete; no primes

user=Free-DC_Beyond
[04/02/08 20:11:04]
66520*6^96685+1 is prime!

I'll be verifying this with PFGW shortly...

[quote=Anonymous;130760]user=Free-DC_Beyond
[04/02/08 20:11:04]
66520*6^96685+1 is prime!

I'll be verifying this with PFGW shortly...[/quote]
Verified with PFGW. Stay tuned for an updated team-sieve file with k=66520 removed. :smile:

[quote=Anonymous;130760]user=Free-DC_Beyond
[04/02/08 20:11:04]
66520*6^96685+1 is prime!

I'll be verifying this with PFGW shortly...[/quote]


Dang, you beat me to the punch! I was just thinking that it had been a long time since we had a prime so went through the last 10 days worth and found this. So I'm glad to see you're on it since this was just found yesterday.

Beyond has pulled his machines off of this effort to check them. As far as I can tell, no one is running the server right now. I pulled my 2 off to put on sieving because otherwise we were going to run the server dry before breaking off a sieving range. So I'll go ahead and throw a machine on it for a while. The last k/n pair in yesterday's results was at n=97.6K so we're moving right along.


Gary

The server is now complete to n=100K for all k's that were originally included in the drive, although we will shortly be in the process of verifying that every k/n pair has been processed.

The server is now processing k=10107, 13215, and 14505 from n=60K-100K that were previously left for manual reservations. It is currently at n=~70K on those.

Thanks for pitching in one of your speedy cores Anon. In looking at the server results, my 2 cores about equal your 1 core, maybe even a little less. lol


Gary

[QUOTE=gd_barnes;131218]The server is now complete to n=100K for all k's that were originally included in the drive, although we will shortly be in the process of verifying that every k/n pair has been processed.

The server is now processing k=10107, 13215, and 14505 from n=60K-100K that were previously left for manual reservations. It is currently at n=~70K on those.

Thanks for pitching in one of your speedy cores Anon. In looking at the server results, my 2 cores about equal your 1 core, maybe even a little less. lol


Gary[/QUOTE]

don't forget to compare residues from 14505 with those i send you in PM ...
just to be sure phrot is ok !

[quote=tnerual;131219]don't forget to compare residues from 14505 with those i send you in PM ...
just to be sure phrot is ok ![/quote]

Yep, I have that on my 'agenda'. lol

Early this morning, the server started on the 3 previously manual k-values including k=14505. It is currently at n=~70K on those 3.

Thanks for the reminder.


Gary

:exclaim::exclaim::exclaim::exclaim::exclaim::exclaim::exclaim::exclaim::exclaim::exclaim:

I just noticed something really odd when I was processing the results for the port 6 LLRnet server.

I found a bunch of k/n pairs for k=157473 in both the results file and the original sieved file, yet k=157473 doesn't show up either in the list of primes found >30K at the beginning of this thread, nor on the Sierp. base 6 reservations page. Was this k erroneously entered as part of the sieve that Gary did for n=30K-100K? It would appear so. Fortunately, n>100K shouldn't be affected, since I put together my equations file from scratch when starting that sieve.

The question now is: was this k accidentally put in the 30K-100K sieve, or was it just not put on the Sierp. base 6 page? Gary, do you by any chance have Geoff's file with all the primes for Sierp. base 6 n<30K? If so, is k=157473 listed in there as having a prime?

Anon

[quote=Anonymous;131266]:exclaim::exclaim::exclaim::exclaim::exclaim:

I just noticed something really odd when I was processing the results for the port 6 LLRnet server.

I found a bunch of k/n pairs for k=157473 in both the results file and the original sieved file, yet k=157473 doesn't show up either in the list of primes found >30K at the beginning of this thread, nor on the Sierp. base 6 reservations page. Was this k erroneously entered as part of the sieve that Gary did for n=30K-100K? It would appear so. Fortunately, n>100K shouldn't be affected, since I put together my equations file from scratch when starting that sieve.

The question now is: was this k accidentally put in the 30K-100K sieve, or was it just not put on the Sierp. base 6 page? Gary, do you by any chance have Geoff's file with all the primes for Sierp. base 6 n<30K? If so, is k=157473 listed in there as having a prime?

Anon[/quote]

n>100K shouldn't be affected?? That's what is incorrect! The sieving and testing for n<=100K is correct. Even more interesting is that the number of k's remaining is correct! :surprised

I keep just about everything. There was no prime from Geoff on this k.

Now, how can all of this be? Of course it's my fault here. I blundered when we found a prime for k=154797. When I went to remove it from the reservation page, I accidentally removed the similar looking k=157473 instead. In a page full of 40+ 5 and 6-digit numbers, they all run together sometimes. Sorry! :sad:

The sieving and testing for n<=100K is correct because I removed the correct k from files with a larger n-value. I just removed the incorrect k on the reservations page. That will be corrected shortly.

We now have a bit of a mess on our hands on the n>100K sieving. Here is what is needed:

1. Remove k=154797 from the current n=100K-150K and n=150K-400K sieves.

2. Sieve k=157473 for n=100K-400K to P=6T, break off n=100K-150K, and combine it with the n=100K-150K file from #1. Fortunately with a single k, speedy sr1sieve can be used.

3. Continue sieving k=157473 for n=150K-400K to whatever our current sieving limit is and combine it with the n=150K-400K file from #1.


Obviously the first priority is to start quickly sieving k=157473. I had dropped back to 2 cores on the sieve for my P=400G range. I'll immediately suspend that and start sieving k=157473. I will also suspend some Riesel base 19 n<2K testing on 2 cores (for Simelink) that I'm running on my work laptop and bring the sieving here back up to 4 cores.

Can you let KrZip know what is up and post a sieved file for him with k=154797 removed? He can just continue along without k=157473 on the other 34 k's and when we get it up to P=8T or whatever, we can then merge it in. There's no reason to hold him up or involve him in correcting the situation.

Edit: NICE CATCH!! This is only a 'small' mess now. Had it gotten into the server like this and had a lot of testing done, THEN we would have had a much bigger mess!


Gary

[quote=gd_barnes;131274]n>100K shouldn't be affected?? That's what is incorrect! The sieving and testing for n<=100K is correct. Even more interesting is that the number of k's remaining is correct! :surprised

I keep just about everything. There was no prime from Geoff on this k.

Now, how can all of this be? Of course it's my fault here. I blundered when we found a prime for k=154797. When I went to remove it from the reservation page, I accidentally removed the similar looking k=157473 instead. In a page full of 40+ 5 and 6-digit numbers, they all run together sometimes. Sorry! :sad:

The sieving and testing for n<=100K is correct because I removed the correct k from files with a larger n-value. I just removed the incorrect k on the reservations page. That will be corrected shortly.

We now have a bit of a mess on our hands on the n>100K sieving. Here is what is needed:

1. Remove k=154797 from the current n=100K-150K and n=150K-400K sieves.[/quote]
Okay, I'll do that ASAP.

[quote]2. Sieve k=157473 for n=100K-400K to P=6T, break off n=100K-150K, and combine it with the n=100K-150K file from #1. Fortunately with a single k, speedy sr1sieve can be used.

3. Continue sieving k=157473 for n=150K-400K to whatever our current sieving limit is and combine it with the n=150K-400K file from #1.


Obviously the first priority is to start quickly sieving k=157473. I had dropped back to 2 cores on the sieve for my P=400G range. I'll immediately suspend that and start sieving k=157473. I will also suspend some Riesel base 19 n<2K testing on 2 cores (for Simelink) that I'm running on my work laptop and bring the sieving here back up to 4 cores.[/quote]
Okay, sounds good. We do have one other option that we could do, though: simply leave k=157473 as a manual-reservation k for n>100K. :smile: That way, we don't have to worry about messing with merging it with the big sieve file and all that--we can go ahead release 100K-150K for PRPing, and simply leave k=157473 out of the team drive.

What do you think?

[quote]Can you let KrZip know what is up and post a sieved file for him with k=154797 removed? He can just continue along without k=157473 on the other 34 k's and when we get it up to P=8T or whatever, we can then merge it in. There's no reason to hold him up or involve him in correcting the situation.[/quote]
Yep, I'll post an updated n=150K-400K sieve file soon. :smile:

Anon :smile:

[quote=Anonymous;131295]Okay, I'll do that ASAP.


Okay, sounds good. We do have one other option that we could do, though: simply leave k=157473 as a manual-reservation k for n>100K. :smile: That way, we don't have to worry about messing with merging it with the big sieve file and all that--we can go ahead release 100K-150K for PRPing, and simply leave k=157473 out of the team drive.

What do you think?


Yep, I'll post an updated n=150K-400K sieve file soon. :smile:

Anon :smile:[/quote]

I starting cranking k=157473 on speedy sr1sieve on 4 cores late last night, 2 on my fastest siever. It will be at P=6T by Saturday and at P=8T by Monday.

When I get to P=6T, I'll send you 2 separate files:
1. One with k/n pairs for n=100K-150K that you can merge in with the range that has been broken off.
2. One with k/n pairs for n=150K-400K that you can merge in with the big drive for future reservations.

When I get to P=8T, I'll send you factors from P=6T-8T to apply against the big n=150K-400K file.

I just need to make sure I'm up to P=6T before KrZip has finished P=7T-8T for the rest of the k's. That way, for future reservations, he'll be using a file that contains the missed k-value.

You might mention in the sieving drive to hold off on any reservations above P=8T until late Saturday.

I like this better than having a 'random' k-value like this be a manual reservation.


Gary

[quote=gd_barnes;131300]I starting cranking k=157473 on speedy sr1sieve on 4 cores late last night, 2 on my fastest siever. It will be at P=6T by Saturday and at P=8T by Monday.

When I get to P=6T, I'll send you 2 separate files:
1. One with k/n pairs for n=100K-150K that you can merge in with the range that has been broken off.
2. One with k/n pairs for n=150K-400K that you can merge in with the big drive for future reservations.

When I get to P=8T, I'll send you factors from P=6T-8T to apply against the big n=150K-400K file.

I just need to make sure I'm up to P=6T before KrZip has finished P=7T-8T for the rest of the k's. That way, for future reservations, he'll be using a file that contains the missed k-value.

You might mention in the sieving drive to hold off on any reservations above P=8T until late Saturday.

I like this better than having a 'random' k-value like this be a manual reservation.


Gary[/quote]
Okay, sounds good. :smile:

New files for n>100K are all ready and will be posted soon. :smile:

In the meantime, reserving 100K-120K for LLRnet port 6.

80K-100K completed by LLRnet IB6, lresults attached. :smile:

New files posted for 120K-125K. Come and get 'em! :grin:

Note: due to the increased testing times at these higher n-levels, we have dropped the file sizes to n=500 (.5K) apiece. :smile:

Reserving 120K-130K for LLRnet port 6. :smile:

Some way or another this got missed in the base 6 server from June 1st:

user=Beyond_Free-DC
[06/01/08 05:18:48]
72785*6^118347+1 is prime! Time : 1050.0 sec.


At 92097 digits, it's just a little short of top-5000.


Gary

Anon and Ironbits,

Can you pull n=120K-130K out of port 6 and remove k=72785 and then feed it back in? At this high of an n-value, it will save significant CPU time for an n=10K range. With no work in 3 weeks on the server and us currently at n=118.7K, it shouldn't create a problem.

I figure that leaving k=72785 in the server file up to n=120K is a small price to pay to make the n=100K-120K results file easy to match up with the original sieved file.


Gary

[quote=gd_barnes;136750]Anon and Ironbits,

Can you pull n=120K-130K out of port 6 and remove k=72785 and then feed it back in? At this high of an n-value, it will save significant CPU time for an n=10K range. With no work in 3 weeks on the server and us currently at n=118.7K, it shouldn't create a problem.

I figure that leaving k=72785 in the server file up to n=120K is a small price to pay to make the n=100K-120K results file easy to match up with the original sieved file.


Gary[/quote]
Okay, I guess that would work, since testing hasn't reached that level yet. IronBits, I'll send you knpairs for 120K-130K with the k removed shortly; then you can remove 120K-130K from the server, and add in the new 120K-130K. :smile:

Done, however, I noticed it's missing that 1st line with all them 000000000000000 in it.
Not sure if hurts anything, but the old one is out, the new one is in.

[quote=IronBits;136799]Done, however, I noticed it's missing that 1st line with all them 000000000000000 in it.
Not sure if hurts anything, but the old one is out, the new one is in.[/quote]
Uh...I hope you didn't remove the [i]entire[/i] old file, just the stuff for n>120K. The new file has only 120K-130K (when the old file had some stuff left below 120K, which should be left intact). That's why I left out the header line, just like I do for "refill" files--I figured since you wouldn't be taking out the [i]whole[/i] file, just the stuff from a certain point on, the original header line wouldn't be removed at all.

The header line is a necessary component--without it, the server will have no idea what base, Sierpinski/Riesel, or whatever, those specific k and n values are supposed to fit into. Whether or not you'll have to re-add it will depend entirely on whether you removed the entire file or just the >120K part.

bangs head. I'll go see what I can figure out. Hope I did it correctly.
stopped server
I looked at the 1st line in the new file you sent, searched for that line in the knpairs.txt
deleted that line and everything below to EOF.
cat newfile.txt >> knpairs.txt
restarted server.

[quote=IronBits;136805]bangs head. I'll go see what I can figure out. Hope I did it correctly.
stopped server
I looked at the 1st line in the new file you sent, searched for that line in the knpairs.txt
deleted that line and everything below to EOF.
cat newfile.txt >> knpairs.txt
restarted server.[/quote]
Okay, you're good then. :tu:

98860*6^119849+1 is prime! (908.3771s+0.0042s)

Found on IB6. Can someone please update the candidates on that server to reflect this.

[quote=MyDogBuster;154344]98860*6^119849+1 is prime! (908.3771s+0.0042s)

Found on IB6. Can someone please update the candidates on that server to reflect this.[/quote]
Well, we're actually not removing the primed k's from the LLRnet server reservations in the middle of the range; instead, we just remove them from all further files that are sent. Otherwise, it messes up the server and the results processing immensely. :smile:

It shouldn't be too big a deal, though--it won't add too much more work, and it will only be up to the end of the currently loaded range (n=130K).

[QUOTE]Well, we're actually not removing the primed k's from the LLRnet server reservations in the middle of the range; instead, we just remove them from all further files that are sent. Otherwise, it messes up the server and the results processing immensely. :smile:

[/QUOTE]

Okay no problem.

Our golden boy strikes again! Ian has been on an amazing prime streak lately. To find a prime on base 6 after less than n=2K processing after we had waited more than n=20K on the last one...truly remarkable. Perhaps we should let him start testing all 7 of our Sierp conjectures for bases <= 32 that have only one k remaining. He'd probably prove half of them in no time! lol

On another note:

After LLRnet reaches n=130K on this server, I'm going to recommend that we dismantle it and not run LLRnet for any non-power-of-2 bases at CRUS.

With Phrot running 30-40% faster than LLR and "regular" LLR running 5-10% faster than "LLRnet" LLR, we're losing too much in the way of CPU resources. Couple those things with not easily being able to remove k's from files and manual Phrot becomes as much as 50% faster than LLRnet.

It's simply not worth it to run LLRnet for non-power-of-2 bases. I found that out when running an LLRnet rally on the base 5 project for about 36 hours.

Any other opinions on this?


Gary

[QUOTE]Our golden boy strikes again! Ian has been on an amazing prime streak lately. To find a prime on base 6 after less than n=2K processing after we had waited more than n=20K on the last one...truly remarkable. Perhaps we should let him start testing all 7 of our Sierp conjectures for bases <= 32 that have only one k remaining.
[/QUOTE]

Just point me to where you need me. I'll finish up port 6 then. Ya never know when another prime might pop up.

[QUOTE=gd_barnes;154677]After LLRnet reaches n=130K on this server, I'm going to recommend that we dismantle it and not run LLRnet for any non-power-of-2 bases at CRUS.

With Phrot running 30-40% faster than LLR and "regular" LLR running 5-10% faster than "LLRnet" LLR, we're losing too much in the way of CPU resources. Couple those things with not easily being able to remove k's from files and manual Phrot becomes as much as 50% faster than LLRnet.

It's simply not worth it to run LLRnet for non-power-of-2 bases. I found that out when running an LLRnet rally on the base 5 project for about 36 hours.[/QUOTE]

Of note, I am working on new software that I am calling PRPNet. It should be possible to replace LLRNet with it. The client can run either LLR or Phrot. I have a long way to go, but if I'm lucky I might have something ready for testing by the end of the year.

This software is neither designed to compete with BOINC (PrimeGrid) nor work like LLRNet. In fact, it will be very similar to ECMNet. Eventually I'll start a new thread containing the software and from there solicit users for enhancements.

[quote=rogue;154724]Of note, I am working on new software that I am calling PRPNet. It should be possible to replace LLRNet with it. The client can run either LLR or Phrot. I have a long way to go, but if I'm lucky I might have something ready for testing by the end of the year.

This software is neither designed to compete with BOINC (PrimeGrid) nor work like LLRNet. In fact, it will be very similar to ECMNet. Eventually I'll start a new thread containing the software and from there solicit users for enhancements.[/quote]
:w00t: :w00t: :w00t: :w00t: :w00t:

I can't wait! :w00t: :smile:

BTW: If you'd like a testbed server for getting the bugs worked out initially, I'd be glad to host one one at [url]http://nplb-gb1.no-ip.org/llrnet/[/url]. I've already got two LLRnet servers running there, and would easily be able to add more LLRnet (or PRPnet! :grin:) servers. :smile:

[quote=gd_barnes;154677]Our golden boy strikes again! Ian has been on an amazing prime streak lately. To find a prime on base 6 after less than n=2K processing after we had waited more than n=20K on the last one...truly remarkable. Perhaps we should let him start testing all 7 of our Sierp conjectures for bases <= 32 that have only one k remaining. He'd probably prove half of them in no time! lol

On another note:

After LLRnet reaches n=130K on this server, I'm going to recommend that we dismantle it and not run LLRnet for any non-power-of-2 bases at CRUS.

With Phrot running 30-40% faster than LLR and "regular" LLR running 5-10% faster than "LLRnet" LLR, we're losing too much in the way of CPU resources. Couple those things with not easily being able to remove k's from files and manual Phrot becomes as much as 50% faster than LLRnet.

It's simply not worth it to run LLRnet for non-power-of-2 bases. I found that out when running an LLRnet rally on the base 5 project for about 36 hours.

Any other opinions on this?


Gary[/quote]
Yes, I agree--we should dismantle CRUS port 6 after it dries, in favor of manual testing in the meantime, and then possibly in the future the PRPnet system that Rogue is developing. But in the meantime, I've got a little suggestion about how to dry the server without having to crunch the remaining numbers suboptimally. :grin:

We can use a variation on the system that I use to manually crunch numbers from NPLB LLRnet servers on LLR 3.7.1c. (I think Carlos uses a similar system sometimes.) The same system can also be used to crunch LLRnet numbers with Phrot. :smile:

I was going to post here how it works, but...on second thought, it's somewhat complex and may not work too well for anyone who isn't quite familiar with the intricacies of my process for doing this. So, I've got a better idea:

How about I use my little method to grab reasonably-sized manual files from the server, post them here, and then users can crunch them, and I'll submit the results back to the server in their name using my litlte method? This shouldn't be hard at all, especially if David increases the jobMaxTime on the server to, say, two weeks (instead of 7 days as it currently is), to avoid people accidentally stepping on in-progress manual reservations. :smile:

I know it sounds somewhat convoluted, but in actual practice, I'm thinking that it wouldn't be too bad. :smile:

Max :smile:

[quote=mdettweiler;154758]Yes, I agree--we should dismantle CRUS port 6 after it dries, in favor of manual testing in the meantime, and then possibly in the future the PRPnet system that Rogue is developing. But in the meantime, I've got a little suggestion about how to dry the server without having to crunch the remaining numbers suboptimally. :grin:

We can use a variation on the system that I use to manually crunch numbers from NPLB LLRnet servers on LLR 3.7.1c. (I think Carlos uses a similar system sometimes.) The same system can also be used to crunch LLRnet numbers with Phrot. :smile:

I was going to post here how it works, but...on second thought, it's somewhat complex and may not work too well for anyone who isn't quite familiar with the intricacies of my process for doing this. So, I've got a better idea:

How about I use my little method to grab reasonably-sized manual files from the server, post them here, and then users can crunch them, and I'll submit the results back to the server in their name using my litlte method? This shouldn't be hard at all, especially if David increases the jobMaxTime on the server to, say, two weeks (instead of 7 days as it currently is), to avoid people accidentally stepping on in-progress manual reservations. :smile:

I know it sounds somewhat convoluted, but in actual practice, I'm thinking that it wouldn't be too bad. :smile:

Max :smile:[/quote]


VERY good idea! If we can use Phrot to crunch the manual ranges, send the results to you, you convert them to LLRnet format, and then send them to the server, sign me up! I'll definitely crunch a few ranges to help Ian dry this puppy out!

BTW, I meant to ask, do we have sieve files for the 1st 3 k's that are not in the drive that I left for manual-k reservations? I'm sorry, I know I could figure it out by browsing through some posts or on one of my machines somewhere but if you know off the top of your head, that would help.

What I'm finding is that manual-k reservations aren't popular at CRUS and really aren't serving much useful purpose. What I'd like to do, perhaps with Ian's or other's help, is to get those 3 k's up to n=130K like the rest of the k's and then include them in future manual files for n>130K.

These two things would make this effort nice and clean and easy to adminster in the future. At this high n-range, it would be easy to remove k's for primes found from manual files because it would only be very sporadic.

Easy to administer efforts make for happy admins that in turn make for happy crunchers! :smile:


Gary

[quote=gd_barnes;154832]VERY good idea! If we can use Phrot to crunch the manual ranges, send the results to you, you convert them to LLRnet format, and then send them to the server, sign me up! I'll definitely crunch a few ranges to help Ian dry this puppy out![/quote]
Okay, cool. I'll send an email to David later tonight asking him to up the jobMaxTime to 20 days, to give everyone a more reasonable amount of time to complete such manual reservations. However, I would have to produce the manual files on request--i.e., I can't post a bunch of them ahead of time and let people pick them up from there. If I did that, then most of them would probably expire on the server before people even get the chance to pick them up.

However, having them generated on request can also be somewhat beneficial: the files can be custom-tailored to whatever size the requestee wishes. :smile:

[quote]BTW, I meant to ask, do we have sieve files for the 1st 3 k's that are not in the drive that I left for manual-k reservations? I'm sorry, I know I could figure it out by browsing through some posts or on one of my machines somewhere but if you know off the top of your head, that would help.[/quote]
Yes, we do have sieve files for them, up to at least n=150K if memory serves. (I think we might in fact have them available up to n=200K, but I don't think I ever got around to breaking them off from the now-dormant team sieve effort's big file. However, the sieve is completed more than adequately for up to 200K.)

I'll email them to you later tonight. (Bug me if you haven't got them by tomorrow. :smile:)

[quote]What I'm finding is that manual-k reservations aren't popular at CRUS and really aren't serving much useful purpose. What I'd like to do, perhaps with Ian's or other's help, is to get those 3 k's up to n=130K like the rest of the k's and then include them in future manual files for n>130K.[/quote]
Okay, that sounds good. :smile:

[quote]These two things would make this effort nice and clean and easy to adminster in the future. At this high n-range, it would be easy to remove k's for primes found from manual files because it would only be very sporadic.[/quote]
One thing to keep in mind, though: once we've loaded a given range into a server, we *cannot* remove any k's from the files, especially once the server has started giving out k/n pairs in that given range. We tried that when we were first experimenting with LLRnet on CRUS efforts, and it was a huge mess. Not so much of a mess on the server end, but on the processing end...well, I'm sure you can imagine all the troubles it would cause to try to verify things with the original sieve file in such a case. :smile:

Max :smile:

Okay, I've just gotten email confirmation from David that IB6 is now set to have a jobMaxTime of 14 days. That should give plenty of time for users to complete their pseudo-manual reservations from the server. :smile:

I'll add a note about this new little system to the first post of this thread. In the meantime, anyone who wants to grab a chunk, just post here in this thread saying how many k/n pairs you want. :smile:

I know there is only about 4000 left so send me about 1500 of those puppies. I'll continue my LLRNET till I receive them.

PS: Please be sure to include a few probable primes so I can keep Gary guessing as to how I find them. LOL

[quote=MyDogBuster;154850]I know there is only about 4000 left so send me about 1500 of those puppies. I'll continue my LLRNET till I receive them.

PS: Please be sure to include a few probable primes so I can keep Gary guessing as to how I find them. LOL[/quote]


HA HA HA!!

I know what you're going to do. If Max sends you a range that your ESP tells you won't have primes, you'll state that "somehow" all of your machines have gone down and will need to unreserve the range. You'll then wait for others to reserve it and shortly thereafter your machines will have a miraculous "recovery". Then when he sends one that you "know" WILL have primes, you'll search it and a day later...bang; yet another top-5000 prime!

Oh, why don't you just save Max some time. Wait until others reserve the primeless ranges and when the "known" prime range gets close, reserve it! Or maybe you already KNOW that another prime is coming in the next range!

Maybe I need to reserve YOUR range!

:missingteeth::missingteeth:



Gary

[quote=mdettweiler;154837]Okay, I've just gotten email confirmation from David that IB6 is now set to have a jobMaxTime of 14 days. That should give plenty of time for users to complete their pseudo-manual reservations from the server. :smile:

I'll add a note about this new little system to the first post of this thread. In the meantime, anyone who wants to grab a chunk, just post here in this thread saying how many k/n pairs you want. :smile:[/quote]


Nice work on that and very good idea!

One thing I'll suggest: I see that Ian is requesting a specific # of pairs. To avoid any confusion, I would suggest sending him a specific n-range that is close to his # of pairs. It appears that something like a range of n=130K-133.5K would be close to the # of pairs he wants.


Gary

[QUOTE]I know what you're going to do. If Max sends you a range that your ESP tells you won't have primes, you'll state that "somehow" all of your machines have gone down and will need to unreserve the range. You'll then wait for others to reserve it and shortly thereafter your machines will have a miraculous "recovery". Then when he sends one that you "know" WILL have primes, you'll search it and a day later...bang; yet another top-5000 prime!
[/QUOTE]

Geez, can't fool you for very long.

[quote=gd_barnes;154870]Nice work on that and very good idea!

One thing I'll suggest: I see that Ian is requesting a specific # of pairs. To avoid any confusion, I would suggest sending him a specific n-range that is close to his # of pairs. It appears that something like a range of n=130K-133.5K would be close to the # of pairs he wants.


Gary[/quote]
Gary, that's now how it works. Essentially, what I'm dong is asking the server to give me the next x # of k/n pairs that it would hand out to a client. Then, after the user crunches them, I simply return them as a client would--the server is none the wiser, and in the end all the results will be merged together just as well as they would have been if we'd crunched them using the normal server process. :smile:

Ian, I've attached a file containing 1500 k/n pairs to this message. Let me know when you've got it so that I can take the attachment offline. :smile:

BTW: if you find a prime in this range, be sure to report it as you would for any other Phrot-found prime (assuming, that is, that you're using Phrot). No special designation is necessary because of the special method by which we're doing these k/n pairs. :smile:

[quote]Ian, I've attached a file containing 1500 k/n pairs to this message. Let me know when you've got it so that I can take the attachment offline. :smile:
[/quote]Got it, Thanks

Edit (by Max): Okay, thanks, I've taken the attachment offline.

[quote=mdettweiler;154948]Gary, that's now how it works. Essentially, what I'm dong is asking the server to give me the next x # of k/n pairs that it would hand out to a client. Then, after the user crunches them, I simply return them as a client would--the server is none the wiser, and in the end all the results will be merged together just as well as they would have been if we'd crunched them using the normal server process. :smile:[/quote]

Gotcha...makes sense. Thanks for the explanation.

118147*6^122688+1 is prime! (916.3466s+0.0043s)

95475 digits - getting closer

32k's left

[quote=MyDogBuster;155235]118147*6^122688+1 is prime! (916.3466s+0.0043s)

95475 digits - getting closer

32k's left[/quote]


I knew it. I should have stolen your range before you could get to it. lol