20210909, 23:51  #1 
"Seth"
Apr 2019
3^{2}×43 Posts 
countsmooth and BrentSuyama
I'm working on improving gmpecm and would appreciate some help validating ecm's countsmooth BrentSuyama code. My trouble is that I can't seem to get ecm to use the same config as countsmooth.
Code:
$ factor 8333333333333333333432551 8333333333333333333432551: 8333333333333333333432551 $ factor 8333333333333333333432550 8333333333333333333432550: 2 3 3 5 5 7 23 59 223 239 379 1733 55691359 $ ./countsmooth N 8333333333333333333432550 B1 10000 B2 1000000 tests 1 D 6 S 12 v B1=10000, B2=1000000, X^12, D=6, 1666<=G<=166666 N=8333333333333333333432550 N+0: 55691359 (BrentSuyama, divides (D*38797)^61^6) B1smooth: 0, B2smooth: 0, found by BrentSuyama: 1.000000, Total: 1 But it's smooth with E=12 (countsmooth's S=12). To test this we can compute (D*38797)^61^6 = (6*38797)**61**6 = 159109595357329249021448903029823 and see that it is indeed a multiple of 55691359. Perfect! Now I'd like to verify this factor is found in ecm Code:
$ echo "8333333333333333333432551"  ecm v power 12 pm1 1e4 1e6 Using B1=10000, B2=1260528, polynomial x^1, x0=1824355551 (no factor found) Also `D` from countsmooth says it relates to the stride of G (" <D> Stride for roots of G in stage 2.") which I can't find any matching parameter to control in ecm. Any help would be appreciated. 
20210910, 00:44  #2 
"Seth"
Apr 2019
603_{8} Posts 
ecm no longer supports power / dickson in P1/P+1 after commit 36108424
By syncing back to 70d63375 I can test passing `power 12` but I still can't figure out how to set D  Reading https://www.rieselprime.de/ziki/BrentSuyama_extension I wonder if I've found my problem "Suppose instead that we compute T=S(6k)21 = 3E*(6k1)*(6k+1) whenever one of 6k+1 or 6k1 is prime. If both are prime, then we get to include two for the price of one. Even if only one is prime, the other may be a multiple of some other prime > B1, so with a bit of planning, we may be able to skip that prime on the way up, and thus again get two for the price of one. This is called "prime pairing"." The factor is found by K=38797, (6*K)^61 = (6*K+1) * (6*K1) * (higher order poly) but neither (6*K+1) or (6*K1) is prime. Does prime pairing happen for all K with higher powers? or does it still only happen when one / both of (6*K+1) and (6*K1) are prime? Last fiddled with by SethTro on 20210910 at 01:15 
20210910, 11:52  #3 
"Oliver"
Sep 2017
Porta Westfalica, DE
2^{8}·3 Posts 
When doing some testing, I also observed that the switches for this do not work anymore, but I did not know that it was voluntarily removed. Do you have any information why they did this and if it might come back on a later date?

20210910, 16:45  #4 
"Seth"
Apr 2019
3^{2}×43 Posts 
In 2008 ecm added fast stage 2.
In 2013 they cleaned up the old stage 2 which was needed for power/Dickson. I doubt it will be added back in the future 
Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Brent tables reservations  chris2be8  Factoring  446  20200429 17:08 
Curious about the Suyama test  siegert81  Math  2  20141123 21:12 
BrentSuyama extension of P1 factoring  S485122  Math  1  20090823 15:21 
brent suyama extension in P1  bsquared  Factoring  9  20070518 19:24 
Brent's p1  How to deal with memory problems?  jhillenb  Factoring  4  20050111 23:50 