Am I Wasting My Time?

pxp · 2014-05-26, 12:58

A couple of days ago I looked up several hundred values of 10^(2*n)+10^n+1 and was surprised by some of the small unfactored composites that showed up in that viewing. It was like I was the first person to ever look at these. Subsequently I noticed that some of the large composites that remained after a refresh were factorizable by looking at repunit factors for 3*n.

So I started fixing these. I'm up to n=263. Then I thought, maybe this is fixable without my adding factors one at a time. Am I wasting my time?

henryzz · 2014-05-26, 13:34

Quote:

Originally Posted by pxp

A couple of days ago I looked up several hundred values of 10^(2*n)+10^n+1 and was surprised by some of the small unfactored composites that showed up in that viewing. It was like I was the first person to ever look at these. Subsequently I noticed that some of the large composites that remained after a refresh were factorizable by looking at repunit factors for 3*n.

So I started fixing these. I'm up to n=263. Then I thought, maybe this is fixable without my adding factors one at a time. Am I wasting my time?

How is the data formatted that you are inputting? There are several formats that the database has been trained to read I think.
A useful one is basically submitting lines like below into report results:
6=2*3

pxp · 2014-05-26, 13:47

My data isn't formatted. I bring up each unfactored composite that I know has factors (by reference with a repunit-factors site) and copy/paste potential factors into the report-factors window and submit.

I have it from someone that 10^(2*n)+10^n+1 wasn't broken like this a year ago. I guess what I'm really asking is if the thing was there then, perhaps this is fixable on the database-maintenance end of things.

Batalov · 2014-05-26, 17:33

The repunit factorization database is the go-to database. All your numbers are factors of 10³ⁿ-1 and/or are a proper combination of Phi_3k(10). For prime n, they are simply Phi_3n(10) · 3 · 37. There's also a full set download on Kamada's pages (easy to find).

Factordb.com is a bit like wikipedia: not only you will not always find the answer (sometimes you will find, "The page "..." does not exist. You can ask for it to be created"); sometimes you will find a wrong answer. Factordb.com is an excellent swiss army knife. When you want to saw off a branch of a huge tree, you put away the knife and get the power saw.

You can, of course, feed the Kamada's DB into factordb.com, but as you can already see, there is no guarantee that the results will be there in a year. Sometimes, Syd restores his database from some backups and/or refactors some internal representations (and as a result gets new functionality/data or loses some old functionality/data).