There are Bad and Good news for Seventeen-Or-Bust... Which do you want to hear first? - Page 4

philmoore · 2016-11-23, 00:15

Shouldn't this be split off into a new thread, say, "Good news for Seventeen or Bust"? Nine years since the last discovery, let's hope we don't go that long until the next!

I guess we could now rename the project "Five or Bust". Oops, that name's already been taken, but if this search ended up as lucky as our dual Sierpinski problem search, we should find the last prime showing up with an exponent around 238 million, close to 72 million decimal digits. (I haven't actually recomputed the odds with the particular Proth weights, but I am just assuming that the sequences in both problems had similar weights.) What I find intriguing is that numbers this size can actually be tested on a time scale of a few months per test.

Great news, and good competition for GIMPS!

AG5BPilot · 2017-03-01, 19:02

Hey guys,

Just wanted to clear something up. I am the one who originally used the word "datacenter", but I was using that term generically, and unfortunately that led to some confusion. I don't know if the actual SoB server was in a real datacenter, in an office, or for all I know underneath Louie's bed. All I know is that the server was lost, there were backups, but the backups couldn't be recovered. Louie never shared the details about exactly what happened.

Also, someone asked about how much was recovered from log files sent to us from users. We actually got a lot more than I expected (I really wasn't expecting much, but every little bit helps.)

Of the recent tests -- which are, of course, the largest numbers and therefore the bulk of the processing -- we received residues for about 25% of all the candidates from log files sent by users. Isn't that great? People came through big-time.

One final note to George (or anyone else): The anti-spam system at PrimeGrid may prevent some newcomers from posting anywhere except our "Help" forum. If you ever want to post there and are blocked, just drop me a PM there or here and I'll set you up so you can post.

--Mike

Mark Rose · 2017-03-02, 00:39

That's good news! I wish I had been more diligent about keeping log files. I've sent you all I found across various systems.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Seventeen or Bust drive	philmoore	Lounge	6	2014-05-26 23:10
Good News	Flatlander	Lounge	5	2011-04-28 20:47
Seventeen or Bust finds a missed prime!	VJS	Lounge	3	2005-10-20 05:55
Seventeen or Bust found a prime!	Mystwalker	Math	5	2005-01-04 17:00
Seventeen or Bust finds Sixth Prime!	wblipp	Lounge	4	2003-12-16 21:08

2016-11-23, 00:15	#34
philmoore "Phil" Sep 2002 Tracktown, U.S.A. 10001011111₂ Posts	Shouldn't this be split off into a new thread, say, "Good news for Seventeen or Bust"? Nine years since the last discovery, let's hope we don't go that long until the next! I guess we could now rename the project "Five or Bust". Oops, that name's already been taken, but if this search ended up as lucky as our dual Sierpinski problem search, we should find the last prime showing up with an exponent around 238 million, close to 72 million decimal digits. (I haven't actually recomputed the odds with the particular Proth weights, but I am just assuming that the sequences in both problems had similar weights.) What I find intriguing is that numbers this size can actually be tested on a time scale of a few months per test. Great news, and good competition for GIMPS!

2017-03-01, 19:02	#35
AG5BPilot Dec 2011 New York, U.S.A. 97 Posts	Hey guys, Just wanted to clear something up. I am the one who originally used the word "datacenter", but I was using that term generically, and unfortunately that led to some confusion. I don't know if the actual SoB server was in a real datacenter, in an office, or for all I know underneath Louie's bed. All I know is that the server was lost, there were backups, but the backups couldn't be recovered. Louie never shared the details about exactly what happened. Also, someone asked about how much was recovered from log files sent to us from users. We actually got a lot more than I expected (I really wasn't expecting much, but every little bit helps.) Of the recent tests -- which are, of course, the largest numbers and therefore the bulk of the processing -- we received residues for about 25% of all the candidates from log files sent by users. Isn't that great? People came through big-time. One final note to George (or anyone else): The anti-spam system at PrimeGrid may prevent some newcomers from posting anywhere except our "Help" forum. If you ever want to post there and are blocked, just drop me a PM there or here and I'll set you up so you can post. --Mike

2017-03-02, 00:39	#36
Mark Rose "/X\(‘-‘)/X\" Jan 2013 2²·733 Posts	That's good news! I wish I had been more diligent about keeping log files. I've sent you all I found across various systems.