Missing: "no factor to 2^63"

Runtime Error · 2017-09-06, 22:15

Under history, there is usually a line that says "no factor to 2^x". However, I noticed that my LL candidate in the 81M range is missing this entry. My number's history starts with "no factor from 2^64 to 2^65" and works up to "no factor from 2^74 to 2^75". This exponent was assigned via primenet (I didn't pick it myself, so I assumed it was ready for LLing).

Is it possible that low bit level TFs weren't run on this number? Or is it just omitted from the history table?

[Note: I assume it's bad form to post the actual number I'm working on.]

Thanks!

Uncwilly · 2017-09-06, 23:58

The numbers in that range have been run to 2^63. Much of that was done some time ago. it is about 0.02 GHz-days to run the number from 2^62 to 2^63. That is less than 10 minutes on a single 4GHz core. If you are worried about your number don't.

Even if it was not done before, a user has been going back and redoing a huge amount of factoring and would have found a missed factor.
http://mersenneforum.org/showthread.php?t=19014

Runtime Error · 2017-09-07, 06:18

@Uncwilly Yes exactly. I saw the post for TJAOI and worried that this 81M exponent had been overlooked.

Perhaps PrimeNet can be fixed this so that others are not adverse to testing exponents with missing "reported" TF tests?

Runtime Error · 2017-09-07, 22:46

Here is the exponent in question. Thanks!

http://www.mersenne.ca/exponent/81675137

Madpoo · 2017-09-08, 03:47

Quote:

Originally Posted by Runtime Error

@Uncwilly Yes exactly. I saw the post for TJAOI and worried that this 81M exponent had been overlooked.

Perhaps PrimeNet can be fixed this so that others are not adverse to testing exponents with missing "reported" TF tests?

Although it's safe to say the "trivial" TF ranges have been done and re-done, there's no guarantee they've been done on any particular exponent.

If we got stats and some kind of log from one of the users doing TF re-checks (TJAOI for example) then that might be taken on their word that they did the trivial bit levels, but otherwise I wouldn't be in favor of just marking those all as done even though we're all pretty sure they were.

Prime95 · 2017-09-08, 05:06

Quote:

Originally Posted by Madpoo

If we got stats and some kind of log from one of the users doing TF re-checks (TJAOI for example) then that might be taken on their word that they did the trivial bit levels, but otherwise I wouldn't be in favor of just marking those all as done even though we're all pretty sure they were.

I'm sure they were done. Plus TJAOI has rechecked.

Runtime Error · 2017-09-09, 03:39

Thanks to all of you for the replies. Surely this thread can die now. Happy hunting!

2017-09-06, 22:15	#1
Runtime Error Sep 2017 USA 2×103 Posts	Missing: "no factor to 2^63" Under history, there is usually a line that says "no factor to 2^x". However, I noticed that my LL candidate in the 81M range is missing this entry. My number's history starts with "no factor from 2^64 to 2^65" and works up to "no factor from 2^74 to 2^75". This exponent was assigned via primenet (I didn't pick it myself, so I assumed it was ready for LLing). Is it possible that low bit level TFs weren't run on this number? Or is it just omitted from the history table? [Note: I assume it's bad form to post the actual number I'm working on.] Thanks!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Aouessare-El Haddouchi-Essaaidi "test": "if Mp has no factor, it is prime!"	wildrabbitt	Miscellaneous Math	11	2015-03-06 08:17
mfaktX result: "found 1 factor for M"	swl551	GPU Computing	2	2012-12-20 15:20
Specifing TF factor depth in "Manual Assignments"?	kracker	PrimeNet	2	2012-07-22 17:49
"Latest client software" missing v25.6	OneOfMany	Software	1	2008-09-18 16:48
P-1 factoring != "Mersenne numbers to factor"?	James Heinrich	Marin's Mersenne-aries	8	2004-05-17 11:09

2017-09-06, 23:58	#2
Uncwilly 6809 > 6502 """"""""""""""""""" Aug 2003 101×103 Posts 2×4,909 Posts	The numbers in that range have been run to 2^63. Much of that was done some time ago. it is about 0.02 GHz-days to run the number from 2^62 to 2^63. That is less than 10 minutes on a single 4GHz core. If you are worried about your number don't. Even if it was not done before, a user has been going back and redoing a huge amount of factoring and would have found a missed factor. http://mersenneforum.org/showthread.php?t=19014

2017-09-07, 06:18	#3
Runtime Error Sep 2017 USA 2·103 Posts	@Uncwilly Yes exactly. I saw the post for TJAOI and worried that this 81M exponent had been overlooked. Perhaps PrimeNet can be fixed this so that others are not adverse to testing exponents with missing "reported" TF tests?

2017-09-07, 22:46	#4
Runtime Error Sep 2017 USA 2×103 Posts	Here is the exponent in question. Thanks! http://www.mersenne.ca/exponent/81675137

2017-09-09, 03:39	#7
Runtime Error Sep 2017 USA 11001110₂ Posts	Thanks to all of you for the replies. Surely this thread can die now. Happy hunting!