Riesel base 3 reservations/statuses/primes - Page 64

gd_barnes · 2015-08-14, 04:45

Quote:

Originally Posted by rebirther

Reserving R3 from n=100-150k (4-5G) for BOINC

Question: Why are you doing k=4G-5G to n=150K? k=1G-4G is not reserved and is still at n=100K. Just wondering.

rebirther · 2015-08-14, 05:09

Quote:

Originally Posted by gd_barnes

Question: Why are you doing k=4G-5G to n=150K? k=1G-4G is not reserved and is still at n=100K. Just wondering.

VBCurtis gave me the sievefiles. He is sieving the range deeper soon as I have finished the current range.

VBCurtis · 2015-08-14, 05:18

Quote:

Originally Posted by gd_barnes

Question: Why are you doing k=4G-5G to n=150K? k=1G-4G is not reserved and is still at n=100K. Just wondering.

When I started sieving for R3, I was told (earlier in this thread) that Lennart was sieving 1G-4G from 100k to 500k, and that I should begin my work at 4G. So, I have two sieves- 4G-5G and 5G-6G. I'm sieving 5-6G now while Reb gets to 150k; I'll send him 100-150k from the 5-6G sieve then continue sieving 4G-5G from 150k up.

I expect 100-150 to have so many primes that the sieve will speed up substantially; I expect my next piece for boinc to be 150-225k, followed by 225-500.
If Lennart hasn't reappeared by the time all that is done, I'll sieve 1G-4G with sr2sieve next.

Edit- Specifically posts 573-575 and 662.

gd_barnes · 2015-08-14, 05:27

OK thanks for the info. guys.

R3 is finally up to date in the first post here and on the pages.

KEP · 2015-08-14, 11:01

Quote:

Originally Posted by gd_barnes

OK thanks for the info. guys.

R3 is finally up to date in the first post here and on the pages.

Not anymore

R3 20.1G-23G is complete and R3 23G-27G is in progress. I hope that the way I optimized the sievedepth and added 12 more phases (according to FFT jumps) that I'll see a speed increase in the time needed to start a 1G range. Resultfile is currently uploading to GoogleDrive (2.61GB) and as soon as that completes, I'll e-mail Gary the download link.

KEP · 2015-08-20, 15:42

I'm reserving for my Sandy Bridge, 27G-29G.

I need to run this range, to compare the efficiency of srbsieve, using my optimized srbsieve.ini file. If I loose efficiency on the Sandy Bridge, then it can deffinently not be generalized that my srbsieve.ini file can be used be anyone. The range will start tomorrow, since the Sandy Bridge is not located the same place as my Haswell, wich host the optimized srbsieve.ini file. However NewPGen work will be prepared today.

VBCurtis · 2015-08-20, 20:02

I have begun sieving 1G-4G from 100k.

rebirther · 2015-08-21, 15:02

R3 tested to n=150k (100-150k) (4-5G)

513 primes found

Results emailed, Base released

KEP · 2015-08-29, 09:30

I'm gonna have to un-reserve 27G-29G.

I'm unfortunantly not the only one using the Sandy Bridge and almost 16 hours a day, one or more Facebook games is played on the Sandy Bridge. This gaming, using google Chrome, requires a whopping 90% of the CPU ressources. Hence I'm only having an average of 10-20% of the CPU time available for srbsieve. Therefor I can no longer make an accurate efficiency test on my Sandy Bridge and have now decided to stop 27G-29G and resume working on 8G-9G on my Sandy Bridge and once 23G-27G completes during next week on my Haswell, I'll resume 9G-13G to n=100K.

I expect R3 8G-13G to complete within the next 4 weeks and after the completion of 8G-13G I'll resume starting new ranges.

Sorry for the inconvenience that this is causing, but I must stress that I'm still enroute to have R3 completely started to n=25K.

Finally, if someone has any suggestions on how to reduce the amount of CPU ressources spent on playing games on Facebook, please let me know in a PM (Not here, since it is off topic)

Take care and have a nice one.

rebirther · 2015-08-29, 20:22

Reserving R3 from n=100-150k (5-6G) for BOINC

KEP · 2015-09-05, 13:41

R3 23G-27G is complete to n=25K. When using the optimized .ini file, you can expect to complete 1G in ~20 days on an I5 4670 @ 3.8 GHz running with 1600 MHz DDR3 RAM.

For R3 23G-27G following findings were made:

583,129,717 MOB's
1,416,845,873 primes
24,410 k's remain

Resultfile (3.6GB) will be sent, as soon as upload completes in about 10 hours time :)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Bases 6-32 reservations/statuses/primes	gd_barnes	Conjectures 'R Us	1398	2021-08-06 12:49
Bases 251-500 reservations/statuses/primes	gd_barnes	Conjectures 'R Us	2305	2021-08-04 15:09
Bases 33-100 reservations/statuses/primes	Siemelink	Conjectures 'R Us	1693	2021-08-01 08:40
Bases 101-250 reservations/statuses/primes	gd_barnes	Conjectures 'R Us	908	2021-08-01 07:48
Sierp base 3 reservations/statuses/primes	gd_barnes	Conjectures 'R Us	388	2020-10-21 19:42

2015-08-14, 05:27	#697
gd_barnes May 2007 Kansas; USA 3²·13·89 Posts	OK thanks for the info. guys. R3 is finally up to date in the first post here and on the pages.

2015-08-20, 15:42	#699
KEP Quasi Admin Thing May 2005 3C6₁₆ Posts	I'm reserving for my Sandy Bridge, 27G-29G. I need to run this range, to compare the efficiency of srbsieve, using my optimized srbsieve.ini file. If I loose efficiency on the Sandy Bridge, then it can deffinently not be generalized that my srbsieve.ini file can be used be anyone. The range will start tomorrow, since the Sandy Bridge is not located the same place as my Haswell, wich host the optimized srbsieve.ini file. However NewPGen work will be prepared today.

2015-08-20, 20:02	#700
VBCurtis "Curtis" Feb 2005 Riverside, CA 2×2,437 Posts	I have begun sieving 1G-4G from 100k.

2015-08-21, 15:02	#701
rebirther Sep 2011 Germany B40₁₆ Posts	R3 tested to n=150k (100-150k) (4-5G) 513 primes found Results emailed, Base released

2015-08-29, 09:30	#702
KEP Quasi Admin Thing May 2005 2·3·7·23 Posts	I'm gonna have to un-reserve 27G-29G. I'm unfortunantly not the only one using the Sandy Bridge and almost 16 hours a day, one or more Facebook games is played on the Sandy Bridge. This gaming, using google Chrome, requires a whopping 90% of the CPU ressources. Hence I'm only having an average of 10-20% of the CPU time available for srbsieve. Therefor I can no longer make an accurate efficiency test on my Sandy Bridge and have now decided to stop 27G-29G and resume working on 8G-9G on my Sandy Bridge and once 23G-27G completes during next week on my Haswell, I'll resume 9G-13G to n=100K. I expect R3 8G-13G to complete within the next 4 weeks and after the completion of 8G-13G I'll resume starting new ranges. Sorry for the inconvenience that this is causing, but I must stress that I'm still enroute to have R3 completely started to n=25K. Finally, if someone has any suggestions on how to reduce the amount of CPU ressources spent on playing games on Facebook, please let me know in a PM (Not here, since it is off topic) Take care and have a nice one.

2015-08-29, 20:22	#703
rebirther Sep 2011 Germany 2⁶·3²·5 Posts	Reserving R3 from n=100-150k (5-6G) for BOINC

2015-09-05, 13:41	#704
KEP Quasi Admin Thing May 2005 2·3·7·23 Posts	R3 23G-27G is complete to n=25K. When using the optimized .ini file, you can expect to complete 1G in ~20 days on an I5 4670 @ 3.8 GHz running with 1600 MHz DDR3 RAM. For R3 23G-27G following findings were made: 583,129,717 MOB's 1,416,845,873 primes 24,410 k's remain Resultfile (3.6GB) will be sent, as soon as upload completes in about 10 hours time :)