Can someone run a couple of poly searches for me?

schickel · 2012-04-10, 08:51

I was hoping someone would have the time and resources to run a couple of poly searches for me. (Believe it or not, I decided to buy a GPU to run some projects on, but didn't realize msieve uses Nvidia rather than the ATI that I bought....it pays to check first!)

The first is a c151 from 572000:i3094:

Code:

2119140345775000108180962800888015075769919837681286337336416586905122205436924899536701602539379676667489862940462606633001801302615803564132806646527

The second is a c153 from 363270:i1759

Code:

639818713524495629538702644398102175871519663684229775565480903998487599472886004787307388009209457487386090128633004519078074161082111880734247598519847

(The second one is different from the one at the end of 363270 in the DB.....)

[EDIT: I'm hoping I can do them a little faster than the last couple of c15x jobs I did if I have a (more) decent poly.....]

poily · 2012-04-10, 15:37

Try this one for the c153.

Code:

n: 639818713524495629538702644398102175871519663684229775565480903998487599472886004787307388009209457487386090128633
deg: 5                                                                                                               
skew: 523854                                                                                                         
c5: 34756680                                                                                                         
c4: 5266016511076                                                                                                    
c3: -29995600624358361218                                                                                            
c2: -360079237313642767836789                                                                                        
c1: 2964946819160389790182814124390                                                                                  
c0: -63006984115844908195013033867842800                                                                             
Y1: 428681183752426013                                                                                               
Y0: -112980548970114676128587090489                                                                                  
# m: 5131382905518585018495790672707506529455054822008532962260546524623400134994083111014129201125322601830748977034
# norm: 1.558301e+21                                                                                                 
# alpha: -6.725791e+00                                                                                               
# murphy_e: 3.524674e-12

schickel · 2012-04-10, 17:46

Quote:

Originally Posted by poily

Try this one for the c153.

Code:

n: 639818713524495629538702644398102175871519663684229775565480903998487599472886004787307388009209457487386090128633
deg: 5                                                                                                               
skew: 523854                                                                                                         
c5: 34756680                                                                                                         
c4: 5266016511076                                                                                                    
c3: -29995600624358361218                                                                                            
c2: -360079237313642767836789                                                                                        
c1: 2964946819160389790182814124390                                                                                  
c0: -63006984115844908195013033867842800                                                                             
Y1: 428681183752426013                                                                                               
Y0: -112980548970114676128587090489                                                                                  
# m: 5131382905518585018495790672707506529455054822008532962260546524623400134994083111014129201125322601830748977034
# norm: 1.558301e+21                                                                                                 
# alpha: -6.725791e+00                                                                                               
# murphy_e: 3.524674e-12

Thanks! I'll give that a shot after I clear my current job out.

Gimarel · 2012-04-11, 08:24

Here are some Polys for the c151 with Murphy above 5e-12. I didn't do any testsieving so I don't know which one is the best.

Code:

# norm 1.436519e-14 alpha -6.373245 e 5.333e-12 rroots 5
skew: 826298.12
c0: 20845566678164297365120804561756705
c1: 183294175802435976662843101867
c2: -2477653833691984307550241
c3: -1500710162407695595
c4: 3387135843768
c5: 303240
Y0: -93083802748178365489228305348
Y1: 50774633142117043
# norm 1.418318e-14 alpha -6.930302 e 5.289e-12 rroots 5
skew: 1835730.88
c0: 82815140119816432465056299833490720
c1: 5681578590449662888712750738744
c2: 8412717872557527052674616
c3: -3334831675722595883
c4: -2076031905768
c5: 402696
Y0: -87950721685558702461613852347
Y1: 574803280795475831
# norm 1.411744e-14 alpha -5.548414 e 5.283e-12 rroots 5
skew: 886423.30
c0: -26555535096681986150415217156944000
c1: 102045801315230312254936098516
c2: 1173381052126279136450844
c3: 830408978575928693
c4: -1463848987804
c5: 204516
Y0: -100713801366198969390446374939
Y1: 405376078854585133
# norm 1.353711e-14 alpha -6.128619 e 5.194e-12 rroots 3
skew: 1385667.62
c0: -75486620634676097091282234634648275
c1: 1274714303067357415392086545455
c2: -51674158425751281959109
c3: -1946356708187504659
c4: -850365410344
c5: 204516
Y0: -100713558166090349607205438318
Y1: 405376078854585133
# norm 1.373296e-14 alpha -6.193639 e 5.070e-12 rroots 5
skew: 1474887.67
c0: 1700940406413034553879044098357857028
c1: 1959433932728088480876706655796
c2: -3287162790508621978788219
c3: -2454677018300970782
c4: 1993355023160
c5: 302640
Y0: -93120714196329552804201303461
Y1: 61879896886832137
# norm 1.285081e-14 alpha -5.869421 e 5.031e-12 rroots 3
skew: 1336775.79
c0: 12555211792437348741530435237557365
c1: 1127439682400476824013798698447
c2: 816538447743226651810659
c3: -1273883666325951763
c4: -1032932753224
c5: 204516
Y0: -100713630540313963989416743606
Y1: 405376078854585133
# norm 1.303593e-14 alpha -6.474448 e 5.021e-12 rroots 3
skew: 1761452.94
c0: -1378929369151656833474677995565906608
c1: 2383742248887901356908810518272
c2: 3092242137603604808187317
c3: -1955840148157053758
c4: -1767332316010
c5: 107100
Y0: -114624617579907696437145817979
Y1: 307875930233096759
# norm 1.318960e-14 alpha -7.214836 e 5.017e-12 rroots 3
skew: 1557188.76
c0: 40721056008484718287780464636378120
c1: 3716708849833597523245141215698
c2: 1149604206475581903526983
c3: -3386881447048054356
c4: -3606976735796
c5: 104880
Y0: -115105223808476848502333242393
Y1: 62628908437674203

schickel · 2012-04-11, 11:02

Quote:

Originally Posted by Gimarel

Here are some Polys for the c151 with Murphy above 5e-12. I didn't do any testsieving so I don't know which one is the best.

Thank you for that run. I'll have to decide which one I want to run first; I put them both on hold so long ago, I don't remember which has a higher priority....

schickel · 2012-04-12, 13:06

Quote:

Originally Posted by poily

Try this one for the c153.

Again, thanks for the search. I'm going to do the c153 first, since 363270 is larger, so if the next line needs ECM, I can do it while the c151 is running.....

schickel · 2012-04-26, 12:25

Quote:

Originally Posted by poily

Try this one for the c153.

Code:

n: 639818713524495629538702644398102175871519663684229775565480903998487599472886004787307388009209457487386090128633
deg: 5                                                                                                               
skew: 523854                                                                                                         
c5: 34756680                                                                                                         
c4: 5266016511076                                                                                                    
c3: -29995600624358361218                                                                                            
c2: -360079237313642767836789                                                                                        
c1: 2964946819160389790182814124390                                                                                  
c0: -63006984115844908195013033867842800                                                                             
Y1: 428681183752426013                                                                                               
Y0: -112980548970114676128587090489                                                                                  
# m: 5131382905518585018495790672707506529455054822008532962260546524623400134994083111014129201125322601830748977034
# norm: 1.558301e+21                                                                                                 
# alpha: -6.725791e+00                                                                                               
# murphy_e: 3.524674e-12

Code:

Thu Apr 26 05:20:33 2012  matrix is 3478570 x 3478796 (1017.9 MB) with weight 261412505 (75.14/col)
Thu Apr 26 05:20:33 2012  sparse part has weight 232039396 (66.70/col)
Thu Apr 26 05:20:33 2012  using block size 21845 for processor cache size 512 kB
Thu Apr 26 05:20:54 2012  commencing Lanczos iteration (2 threads)
Thu Apr 26 05:20:54 2012  memory use: 841.8 MB
Thu Apr 26 05:22:09 2012  linear algebra at 0.0%, ETA 45h52m
Thu Apr 26 05:22:33 2012  checkpointing every 80000 dimensions

jasonp · 2012-04-26, 16:49

What machine is this running on? Does it actually have a 512kB L2 or do I have to mess with the cache detection code again?

Batalov · 2012-04-26, 19:07

Xeons regularly report this size and have a tiny block as a result.

schickel · 2012-04-27, 06:39

Quote:

Originally Posted by jasonp

What machine is this running on? Does it actually have a 512kB L2 or do I have to mess with the cache detection code again?

Machine info from CPU-Z:

jasonp · 2012-04-27, 10:08

Okay, looks like the current code is correct then. AMD machines have a relatively simple way of reporting cache sizes, and Intel machines added a similar mechanism around 2010. The problem is that some Xeon machines still have trouble, as Serge has pointed out.

2012-04-10, 08:51	#1
schickel "Frank <^>" Dec 2004 CDP Janesville 2×1,061 Posts	Can someone run a couple of poly searches for me? I was hoping someone would have the time and resources to run a couple of poly searches for me. (Believe it or not, I decided to buy a GPU to run some projects on, but didn't realize msieve uses Nvidia rather than the ATI that I bought....it pays to check first!) The first is a c151 from 572000:i3094: Code: 2119140345775000108180962800888015075769919837681286337336416586905122205436924899536701602539379676667489862940462606633001801302615803564132806646527 The second is a c153 from 363270:i1759 Code: 639818713524495629538702644398102175871519663684229775565480903998487599472886004787307388009209457487386090128633004519078074161082111880734247598519847 (The second one is different from the one at the end of 363270 in the DB.....) [EDIT: I'm hoping I can do them a little faster than the last couple of c15x jobs I did if I have a (more) decent poly.....] Last fiddled with by schickel on 2012-04-10 at 08:53 Reason: adding PS

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Generalized Cullen and Woodall Searches	rogue	And now for something completely different	40	2020-12-08 13:21
Prime Gap Searches Crowdfunding	pinhodecarlos	Prime Gap Searches	12	2017-07-28 18:05
Searches and defaults	Nelson	Forum Feedback	18	2010-07-17 19:01
Thoughts on future NPLB searches	em99010pepe	No Prime Left Behind	38	2008-12-23 10:02
Conjectures 'R Us; searches needed	gd_barnes	Conjectures 'R Us	41	2008-01-18 21:47

2012-04-26, 16:49	#8
jasonp Tribal Bullet Oct 2004 3534₁₀ Posts	What machine is this running on? Does it actually have a 512kB L2 or do I have to mess with the cache detection code again?

2012-04-26, 19:07	#9
Batalov "Serge" Mar 2008 Phi(4,2^7658614+1)/2 2×4,663 Posts	Xeons regularly report this size and have a tiny block as a result.

2012-04-27, 10:08	#11
jasonp Tribal Bullet Oct 2004 DCE₁₆ Posts	Okay, looks like the current code is correct then. AMD machines have a relatively simple way of reporting cache sizes, and Intel machines added a similar mechanism around 2010. The problem is that some Xeon machines still have trouble, as Serge has pointed out.