Found a factor? Post it here. Or forever fold your crease. - Page 80

LaurV · 2014-11-18, 04:41

After months of dry ECM spell...

[edit: actually, I would be better with some TF here, hehe, the factor is only 66 bits!]

lorgix · 2014-11-18, 13:03

Quote:

Originally Posted by PageFault

Those are very smooth, nice finds. I assume they are P-1, Stage 1 factors? I have one from a few years back, somewhere in a backup.

I'm seeing my ECM finds to be all rough, although my data is lacking.

Crunch on, happy hunting!

The smooth ones have already been found with P-1.

TheJudger · 2014-11-18, 18:07

yep, they are found with P-1.

lycorn · 2014-11-20, 11:16

Quote:

Originally Posted by LaurV

Yaaarrrrr!

After months of dry ECM spell...

[edit: actually, I would be better with some TF here, hehe, the factor is only 66 bits!]

You´d probably be better off running ECM on exponents much lower than 9M, as they are much more suitable for ECM (and much harder to TF).

LaurV · 2014-11-20, 11:27

I digested what the server gave me.
I have stopped anyhow, after I found that factor, and concentrate now on DCTF, after the last arguments in the GPU72 thread.

lycorn · 2014-11-20, 12:40

That´s right, exponents handed by the server for ECM are in the 9M range. I wonder why so high. That´s why I get them manually.
Just out of curiosity, when you say you gave up ECM to concentrate on DCTF does that mean you are moving CPU(s) to DCTF, or you were using some program that would allow running ECM on a GPU? Or...?

VictordeHolland · 2014-11-20, 15:10

IIRC 2^1018-1 is still the maximum GPU-ECM can handle.

The difficulty with determining the desired TF level before running ECM lies with the differences in CPU and GPU capabilities and available resources. On ECM candidates with no (or very few) curves you MIGHT be better off running TF a bit deeper IF you have the GPU resources available (which should really be doing DCTF and LLTF). I personally wouldn't recommend running TF on anything <10M, since most of those have had a fair share of P-1 and ECM.

lycorn · 2014-12-02, 18:48

Look at this one:

ECM found a factor in curve #181, stage #2
Sigma=3947118002300977, B1=250000, B2=25000000.
UID: lycorn/asteroid, M507149 has a factor: 1834049842474634749280697499913 (ECM curve 181, B1=250000, B2=25000000)

Factor is prime, and its k=2² × 452049063114251124738661

Only ECM could have found it...

legendarymudkip · 2014-12-02, 22:25

M70637093 has a factor: 27886899376721683540674433 (84.5 bits)
k=2⁶ × 3³ × 7 × 11 × 13 × 23 × 7877 × 629899
k is pretty smooth and has a lot of very small factors.

VictordeHolland · 2014-12-03, 14:47

A nice P-1 stage 1 factor popped up yesterday 91.5 bits (28 digits):

P-1 found a factor in stage #1, B1=1500000.
UID: VictordeHollander/PCVICTOR, M68842181 has a factor: 3483933495938630920516781759 (P-1, B1=1500000)

k = 25303770488754786259 = 7 × 11 × 977 × 12641 × 26777 × 993703

Dubslow · 2014-12-03, 16:37

First a 2^2 * prime k, then an odd k...

2014-11-18, 04:41	#870
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 24063₈ Posts	Yaaarrrrr! After months of dry ECM spell... [edit: actually, I would be better with some TF here, hehe, the factor is only 66 bits!] Last fiddled with by LaurV on 2014-11-18 at 04:47

2014-12-03, 14:47	#879
VictordeHolland "Victor de Hollander" Aug 2011 the Netherlands 3²·131 Posts	P-1 stage 1 factor A nice P-1 stage 1 factor popped up yesterday 91.5 bits (28 digits): P-1 found a factor in stage #1, B1=1500000. UID: VictordeHollander/PCVICTOR, M68842181 has a factor: 3483933495938630920516781759 (P-1, B1=1500000) k = 25303770488754786259 = 7 × 11 × 977 × 12641 × 26777 × 993703

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Sorry about last post I hope u like this new factor method	ONeil	ONeil	4	2021-01-02 20:01
Factor found that should have been found by P-1	tha	Data	65	2020-08-05 21:11
10 fold increase in new CPU's	Uncwilly	PrimeNet	10	2019-11-13 18:18
POST PRIMES you've found here, and name or prover code	TTn	15k Search	415	2006-03-02 21:17
Post the numbers you have tried to factor here	hyh1048576	Factoring	11	2003-12-01 07:37

2014-11-18, 18:07	#872
TheJudger "Oliver" Mar 2005 Germany 5×223 Posts	yep, they are found with P-1.

2014-11-20, 11:27	#874
LaurV Romulan Interpreter "name field" Jun 2011 Thailand 41·251 Posts	I digested what the server gave me. I have stopped anyhow, after I found that factor, and concentrate now on DCTF, after the last arguments in the GPU72 thread.

2014-11-20, 12:40	#875
lycorn "GIMFS" Sep 2002 Oeiras, Portugal 2×7×113 Posts	That´s right, exponents handed by the server for ECM are in the 9M range. I wonder why so high. That´s why I get them manually. Just out of curiosity, when you say you gave up ECM to concentrate on DCTF does that mean you are moving CPU(s) to DCTF, or you were using some program that would allow running ECM on a GPU? Or...?

2014-11-20, 15:10	#876
VictordeHolland "Victor de Hollander" Aug 2011 the Netherlands 3²×131 Posts	IIRC 2^1018-1 is still the maximum GPU-ECM can handle. The difficulty with determining the desired TF level before running ECM lies with the differences in CPU and GPU capabilities and available resources. On ECM candidates with no (or very few) curves you MIGHT be better off running TF a bit deeper IF you have the GPU resources available (which should really be doing DCTF and LLTF). I personally wouldn't recommend running TF on anything <10M, since most of those have had a fair share of P-1 and ECM.

2014-12-02, 18:48	#877
lycorn "GIMFS" Sep 2002 Oeiras, Portugal 1582₁₀ Posts	Look at this one: ECM found a factor in curve #181, stage #2 Sigma=3947118002300977, B1=250000, B2=25000000. UID: lycorn/asteroid, M507149 has a factor: 1834049842474634749280697499913 (ECM curve 181, B1=250000, B2=25000000) Factor is prime, and its k=2² × 452049063114251124738661 Only ECM could have found it...

2014-12-02, 22:25	#878
legendarymudkip Jun 2014 2³·3·5 Posts	M70637093 has a factor: 27886899376721683540674433 (84.5 bits) k=2⁶ × 3³ × 7 × 11 × 13 × 23 × 7877 × 629899 k is pretty smooth and has a lot of very small factors.

2014-12-03, 16:37	#880
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3·29·83 Posts	First a 2^2 * prime k, then an odd k...