skipped bit levels - Page 5

Mark Rose · 2015-12-17, 02:20

I found 810 GHzd of work in skipped bit levels, from 65 bits up, for exponents between 34.8M and 105M without any factors.

I'm going to run it all to see if anything comes up.

dbaugh · 2015-12-17, 02:30

I like spackling (clearing a bit range gap with no credit). If a 72.5 bit factor of an exponent is discovered that has TF no factor to 71, then spackling would be checking 71 to 72. I would argue that 72 to 73 is also spackling if the factor was found by P-1, ECM or TF with bail upon factor found.

How should a TF of 73 to 74, etc. be handled? I think it should be treated the same as 70 to 71 was. Unless we choose to not give credit for LL, DC, P-1 or ECM of exponents once a factor has been found. My understanding is that TF is currently discriminated against in this regard. If I am mistaken please clarify the existing rules for me. Finding a factor is better in some ways than LL because it does not need a DC. If one really wanted to save some database space, then erase all the other exponent history once a factor is found.

I actually like a full history and would like to be able to report bit ranges cleared by TF beyond factors found.

LaurV · 2015-12-17, 04:30

Quote:

Originally Posted by Madpoo

Allow results for lower than max TF bit levels, but give no credit

Ok with me. I will stick a tongue at you every time when I will report a result and don't get any credit, like that*:

but that is my problem, not yours...

-----------
* we tried to negotiate in the past with the gods to give us that emote, we don't associate it with yuck, but with something fierce and angry, which we are, but the gods applied the camel law**, i.e. punished us with the lighting rabbit...

---------
** once upon a time Camel was upset because he had no horns, other animals had a defense against predators, etc.. so he went to god to complain. Well, god was in a bad mood and said "Who is this ugly Camel bothering me requesting horns? Cut his ears!", so that is why the camels today have no horns, and no ears...

Mark Rose · 2015-12-17, 06:04

Quote:

Originally Posted by Mark Rose

I found 810 GHzd of work in skipped bit levels, from 65 bits up, for exponents between 34.8M and 105M without any factors.

I'm going to run it all to see if anything comes up.

I had a bug in my program, where I would miss the last missed bit. It's actually 3691 GHzd of work.

Mark Rose · 2015-12-17, 19:04

Quote:

Originally Posted by Mark Rose

I found a bunch of exponents with incongruent entries in their history versus their TF bit levels:

So I'll take each of these to one bit level beyond the last entry in the history.

This one will take a while though:

Quote:

96001991 is factored to 73 but: no factor from 2^66 to 2^79

I don't know why it's in the history.

0PolarBearsHere · 2015-12-18, 07:09

Quote:

Originally Posted by Mark Rose

So I'll take each of these to one bit level beyond the last entry in the history.

Are you starting at a bit level of 2?

Mark Rose · 2015-12-18, 07:35

Quote:

Originally Posted by 0PolarBearsHere

Are you starting at a bit level of 2?

I've been starting at 65 bits. TJAOI seems to have double checked everything below 61 bits. I'll add the 61 to 65 gaps to my testing tomorrow.

So far I've double checked over 1000 bit levels with no factors found.

Mark Rose · 2015-12-19, 18:28

So after testing 2000 skipped bit levels with no factors found, it's not worth my time to continue.

For what it's worth, exponents with no factors from M34800079 to M56500313 have been tested at skipped bit levels 65+.

For completeness, I'm going back and testing the skipped bit levels between 61 and 65.

Mark Rose · 2015-12-19, 23:47

Nothing between 64 and 65 bits, but I did just find this:

M35003827 has a factor: 14068547687256772529 [TF:61:64*:mfaktc 0.21 75bit_mul32_gs]

Which appears to be a skipped bit level.

alpertron · 2015-12-20, 13:36

Quote:

Originally Posted by Mark Rose

TJAOI seems to have double checked everything below 61 bits.

A small correction. At this rate, prime factors found by TJAOI will reach 2⁶⁰ at the end of December. Notice that on Fridays he submits prime factors of Mersenne number whose exponents are less than 1M found by ECM. These are obviously greater than that bound.

Madpoo · 2015-12-20, 18:37

Quote:

Originally Posted by Mark Rose

Nothing between 64 and 65 bits, but I did just find this:

M35003827 has a factor: 14068547687256772529 [TF:61:64*:mfaktc 0.21 75bit_mul32_gs]

Which appears to be a skipped bit level.

That's kind of cool... Kind of makes you think it could be worth some effort after all?

2015-12-17, 02:20	#45
Mark Rose "/X\(‘-‘)/X\" Jan 2013 C72₁₆ Posts	I found 810 GHzd of work in skipped bit levels, from 65 bits up, for exponents between 34.8M and 105M without any factors. I'm going to run it all to see if anything comes up. Last fiddled with by Mark Rose on 2015-12-17 at 02:40

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Normalising rent levels	Bundu	Math	4	2017-09-27 06:14
Racism or low light levels or...?	jasong	jasong	2	2016-09-25 05:07
Missing bit levels?	NBtarheel_33	Data	6	2016-05-31 15:27
Is the data missing or did we miss a couple TF bit levels	petrw1	PrimeNet	2	2015-05-07 05:09
Recommended TF bit levels for M(>10^8)	NBtarheel_33	Math	19	2008-11-03 17:19

2015-12-17, 02:30	#46
dbaugh Aug 2005 11² Posts	I like spackling (clearing a bit range gap with no credit). If a 72.5 bit factor of an exponent is discovered that has TF no factor to 71, then spackling would be checking 71 to 72. I would argue that 72 to 73 is also spackling if the factor was found by P-1, ECM or TF with bail upon factor found. How should a TF of 73 to 74, etc. be handled? I think it should be treated the same as 70 to 71 was. Unless we choose to not give credit for LL, DC, P-1 or ECM of exponents once a factor has been found. My understanding is that TF is currently discriminated against in this regard. If I am mistaken please clarify the existing rules for me. Finding a factor is better in some ways than LL because it does not need a DC. If one really wanted to save some database space, then erase all the other exponent history once a factor is found. I actually like a full history and would like to be able to report bit ranges cleared by TF beyond factors found.

2015-12-19, 18:28	#52
Mark Rose "/X\(‘-‘)/X\" Jan 2013 2·3³·59 Posts	So after testing 2000 skipped bit levels with no factors found, it's not worth my time to continue. For what it's worth, exponents with no factors from M34800079 to M56500313 have been tested at skipped bit levels 65+. For completeness, I'm going back and testing the skipped bit levels between 61 and 65.

2015-12-19, 23:47	#53
Mark Rose "/X\(‘-‘)/X\" Jan 2013 2×3³×59 Posts	Nothing between 64 and 65 bits, but I did just find this: M35003827 has a factor: 14068547687256772529 [TF:61:64*:mfaktc 0.21 75bit_mul32_gs] Which appears to be a skipped bit level.