20081127, 13:47  #1 
I quite division it
"Chris"
Feb 2005
England
31·67 Posts 
1000 < k < 2000
As NPLB will be testing this whole range in the new year I have decided to unreserve my 22 ks between 1400 and 1500. This will mean a deeper sieve for these ks and more fresh primes to find in the drive.
I have tested: 350,000 to 366,314 370,000 to 377,129 400,000 to 402,970 lresults available by request. Chris 
20081127, 16:10  #2 
Just call me Henry
"David"
Sep 2007
Cambridge (GMT/BST)
3^{3}·7·31 Posts 
Thomas is doing a lot of work in this range currently
see the RPS primes thread 
20081127, 17:39  #3  
May 2007
Kansas; USA
10100001101010_{2} Posts 
Quote:
Yes, I've noticed that also. That is OK. We'll inform him of our effort but if he chooses to continue testing as is, he'll easily stay WAY ahead of our drives. It'll likely be at least February before we would do any top5000 searches. Plenty of sieving to do yet. I just got a note from Ian that he has completed sieving to P=300G yesterday. To start top5000 work at n=~350K, we'd likely need it sieved to P=4T6T. He thought that he could reach P=1T by year end himself. After the 1st drive is complete, we'll start LLRing the lower ranges and I will contribute on the sieving. Gary Last fiddled with by gd_barnes on 20081127 at 17:45 

20081127, 17:43  #4  
May 2007
Kansas; USA
10100001101010_{2} Posts 
Quote:
Chris, Thanks for your consideration there! :) To make the ranges easier for Karsten to report, you might continue your effort and fill in the gap for at least n=350378K and 400K403K. Then he can just show each k with "350k378k, 400k403k". It'll make it nice and clean. What is your sieve depth? I don't want you to assume that we will be sieved far further than you are. I'm just guessing at an optimum depth of P=4T6T for the n=350K400K range for all ~500 k's but it could be higher. Even if optimum is higher, I may have us stop short of the optimum so that we can get more top5000 primes reported sooner before the 5000th place prime climbs too much. Gary Last fiddled with by gd_barnes on 20081127 at 17:46 

20081127, 19:15  #5  
I quite division it
"Chris"
Feb 2005
England
31·67 Posts 
Quote:
Okay, I'll fill in the gaps as suggested. Would you like me to stick a core on sieving for a while? 

20081128, 10:16  #6  
Mar 2006
Germany
2·11·131 Posts 
Quote:
i will mark them unreserved when you ready. thanks 

20081202, 15:38  #7 
Feb 2003
2^{2}·3^{2}·53 Posts 
I wasn't aware of this! As you already noticed, I'm currently testing quite a few of the k's in the range 1400<k<1600 for n=350500k, but I didn't want to interfere with your plans. If necessary I can provide the lresults files.
Thomas 
20081202, 19:19  #8 
A Sunny Moo
Aug 2007
USA (GMT5)
3·2,083 Posts 
Yes, we'll be searching this entire range next year, but since we're starting all the way back at n=50K, the option is still open for searchers with existing individualk reservations to continue ahead of us if they wish. Yes, we'll still be searching all the k's, but with doing 500 k's at a time and all, an individual can quite easily stay ahead of us for firstpass testing, and NPLB will essentially just doublecheck them.

20081204, 09:56  #9  
May 2007
Kansas; USA
2×7×739 Posts 
Quote:
Please know that we in no way want to step on yours or anyone else's reservations with our upcoming k=10052000 drives. If you like the effort that you are doing with those k's, please go right ahead with what you are doing. Of course we'd like to have you participate in our effort when we get to that point if you like. We should have well sieved files on all of the k's to be searched at once. But you or some others may not like searching the range in that manner so we aren't claiming "ownership" of the k's. Even with what we are doing, people are still free to reserve their own k in that range (even if it isn't currently reserved) at RPS or NPLB or through no project at all and search until their heart's content and report any top5000 primes however they like. What we'll be doing with existing reservations is informing people of when we will start and approximately how fast we think we will progress (or provide a link to our thread so they can see for themselves). For ~500 k's, it will be a slow methodical upward search by nvalue. It is intended that way so that we don't find primes first on k's that were reserved by others. But we want to leave all of the k's in there so that we can double check other's work and not leave any primes behind. I/we would be grateful if you could forward us the results files when you are done with your effort or as you progress. My Email is: gbarnes017 at gmail dot com . Thanks for offering. Thanks, Gary Last fiddled with by gd_barnes on 20081204 at 09:57 

20081206, 00:27  #10 
Apprentice Crank
Mar 2006
2·227 Posts 
1999*2^1159651 is prime, but it isn't listed on www.rieselprime.org.
I found it during a routine hardware test, so I never bothered to reserve it. There are no other primes for that k between n=50K and n=120K. I haven't attached results files since the tests are quite short, but you can PM or email me if you want them. edit: I'm not interested in testing this k any further. Last fiddled with by MooooMoo on 20081206 at 00:30 
20081206, 01:29  #11  
I quite division it
"Chris"
Feb 2005
England
31×67 Posts 
Quote:
I'm unreserving them all. I'm also unreserving the following ks (with apologies for the mess ): k=1785 tested to [50k and] 200k361k (I gave this k to Kosmaj some time ago, it was then tested to 400k.) http://mersenneforum.org/showpost.ph...0&postcount=53 k=1875 tested [to 100k and] 334k387k k=1995 tested [to 100k and] 334k422k k=2205 tested 210k350k k=2625 tested 189k310k k=4389 tested 187,600 to at least 197,763 k=4515 tested 191,000 to at least 284,191 (The lresults files are corrupted for the last two k, so I've shown the last prime found by me as the upper limit. I am sure of the lower limits.) Thanks Chris 

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
1000 TFLOPs  ramgeis  PrimeNet  2  20140408 10:27 
1000Core Computer  Rodrigo  Hardware  15  20110115 05:07 
1000<k<2000  VBCurtis  Riesel Prime Search  33  20080928 07:15 
k=3001000 area  nuggetprime  Riesel Prime Search  2  20070409 11:44 
The 1000day club  GP2  Data  15  20040611 01:57 