mersenneforum.org  

Go Back   mersenneforum.org > Search Forums

Showing results 1 to 15 of 15
Search took 0.01 seconds.
Search: Posts Made By: bsquared
Forum: Factoring 2021-07-21, 18:30
Replies: 74
Views: 13,265
Posted By bsquared
I get 414353 relations: 93163 smooth + 321190...

I get 414353 relations: 93163 smooth + 321190 from 8007103 partial (1469149 1LP, 6537954 2LP). The factor base size was 413888. So that's 0.92% of survivors.

Pretty much 100% of yafu's TD...
Forum: Factoring 2021-07-21, 15:54
Replies: 74
Views: 13,265
Posted By bsquared
Just to touch on this a bit more, it is typical...

Just to touch on this a bit more, it is typical for yafu's small prime variation to reduce the full-TD load by about a factor of three with this second bound. E.g., :

trial division touched...
Forum: Factoring 2021-07-21, 13:02
Replies: 74
Views: 13,265
Posted By bsquared
Nice work! Great to see such a dramatic...

Nice work! Great to see such a dramatic improvement to already well-optimized code with relatively small changes.
Forum: Factoring 2021-05-27, 21:18
Replies: 74
Views: 13,265
Posted By bsquared
This is pretty cool. Just to be sure, those are...

This is pretty cool. Just to be sure, those are specific examples right? I don't believe you are trying to say that 151 is faster than 1 always.

I have runs some tests with an expanded list, up...
Forum: Factoring 2021-05-25, 21:02
Replies: 74
Views: 13,265
Posted By bsquared
Here is the list of multipliers yafu checks: ...

Here is the list of multipliers yafu checks:


static const uint8_t mult_list[] =
{ 1, 2, 3, 5, 6, 7, 10, 11, 13, 14, 15, 17, 19,
21, 22, 23, 26, 29, 30, 31, 33, 34, 35, 37, 38,
39, 41, 42,...
Forum: Factoring 2021-05-25, 18:34
Replies: 74
Views: 13,265
Posted By bsquared
Yup, I've found that it helps. Example: ...

Yup, I've found that it helps. Example:

multiplier 2:

starting SIQS on c77: 63779362122850320146727330876445758583990107403462620370720721521518029800433
...
using multiplier of 2
using SPV...
Forum: Factoring 2021-05-25, 16:09
Replies: 74
Views: 13,265
Posted By bsquared
Congrats! Did you do much experimenting with...

Congrats!

Did you do much experimenting with parameters (factor base size, large prime bounds, etc.)? Also, k=113, wow. yafu wants to pick k=2 for this input. Does having kN==1 mod 8...
Forum: Factoring 2018-03-05, 16:46
Replies: 74
Views: 13,265
Posted By bsquared
Wow, impressive speedup. How much of that came...

Wow, impressive speedup. How much of that came from the Q2(x) polynomials? I haven't had the chance to experiment with that idea since it first came up, but it is on my todo list.

Nice work!
Forum: Factoring 2017-02-06, 14:58
Replies: 74
Views: 13,265
Posted By bsquared
Well, scales better in the...

Well, scales better in the leading-constant-term-of-big-O-analysis sense, maybe. Thousands of lines of assembler code have been written to optimize the data flow for large numbers in yafu, so that...
Forum: Factoring 2016-01-18, 22:52
Replies: 74
Views: 13,265
Posted By bsquared
I've studied FactorizerTest and the squfof...

I've studied FactorizerTest and the squfof routine and as far as I can tell everything is being tested correctly.

The (1680, 5, 1.5) tests were all clearly the fastest. Again, nice work! If...
Forum: Factoring 2016-01-18, 20:28
Replies: 74
Views: 13,265
Posted By bsquared
That worked perfectly, many thanks!

That worked perfectly, many thanks!
Forum: Factoring 2016-01-18, 15:59
Replies: 74
Views: 13,265
Posted By bsquared
I was hoping it would be easier to build the...

I was hoping it would be easier to build the source :smile:. Haven't done it yet...


I just thought I'd note that for inputs less than 40 bits, Lehman's method is at least 20% faster than SQUFOF...
Forum: Factoring 2016-01-15, 18:26
Replies: 74
Views: 13,265
Posted By bsquared
Unfortunately I am largely ignorant of java. How...

Unfortunately I am largely ignorant of java. How to I run the FactorizerTest class?
Forum: Factoring 2016-01-11, 16:26
Replies: 74
Views: 13,265
Posted By bsquared
You've obviously put a lot of thought and effort...

You've obviously put a lot of thought and effort into this, good work. My first thought is wow, those multipliers are big! They are too big for me to effectively test them, in fact, except for on...
Forum: Factoring 2016-01-08, 17:48
Replies: 74
Views: 13,265
Posted By bsquared
Well done! New implementations of these...

Well done!

New implementations of these difficult algorithms are commendable. And as someone who is particularly guilty of unreadable code, it is nice to see the operations laid out "in plain...
Showing results 1 to 15 of 15

 
All times are UTC. The time now is 14:27.


Sat Nov 27 14:27:36 UTC 2021 up 127 days, 8:56, 0 users, load averages: 0.93, 1.12, 1.07

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.