Seventeen or Bust drive

[pinhodecarlos]

Louie Helm has announced a two-month drive at Seventeen or Bust in an attempt to find another prime (or two or three) to help solve the Sierpinski problem:
http://www.seventeenorbust.com/
This project found their last prime just over six years ago at 3.9 million digits. Their status page shows that some candidates have been now searched up to 7.8 million digits. They certainly seem overdue for a prime.

For newcomers, this project uses the Prime95 software, but connecting to a different server. An alternative platform is also set up as a BOINC project under Primegrid:
http://www.primegrid.com/

[DrBackJack]

I'm running it now

[pinhodecarlos]

The client is based on Prime95 software meaning that it can detect all cores even on a 80-core machine. AVX Intel processors are a must with HT-off, no HT-on, ok?
I don't know if there's is a SETI.USA team created.
Also the project page is very good on stats where the output progress can be checked online.

[John P. Myers]

Note the seventeen-or-bust app on Primegrid has a 60% credit bonus, but will take around 8 days to complete.

[cineon_lut]

Note the seventeen-or-bust app on Primegrid has a 60% credit bonus, but will take around 8 days to complete.

Cool. I have one of my machines loaded up on PG and it has several of those. Deadline is mid-Jan, so hope my wingmates are on it before then!

[zombie67]

Note the seventeen-or-bust app on Primegrid has a 60% credit bonus, but will take around 8 days to complete.

In the preferences, it says:


This project has a 50% long job credit bonus and a 10% conjecture credit bonus.

Does anyone know what "10% conjecture credit bonus" means?

[John P. Myers]

Hmm actually it seems the way they calculate credits is base * conjecture * duration, which means the SoB bonus is 65%, not 60%.

The conjecture bonus is just a bonus they give for working on the conjecture-type projects (SR5, TRP (sieve & LLR), PSP, and SoB). I don't really know why they give this bonus, but whatever.

[DrBackJack]

Anyone know what the credit/wu is for these ?

EDIT: About 22,000 each with the current 60% bonus.

[cineon_lut]

Anyone know what the credit/wu is for these ?

EDIT: About 22,000 each with the current 60% bonus.

Cool. I just got a bronze, and my 12 in prog work units should shoot me way past silver for sob! I'm seeing the same numbers you are, so that's good!


Vic (mobile)

[STE\/E]

I'm running a couple SOB's to get my Turquoise Badge, http://www.primegrid.com/img/badges/sob_llr_turquoise.png didn't realize it but I was only 26k short of the Turquoise Badge so I need to run 2 anyway to get to 5 Million ...

[trigggl]

I just had one of these validate this week. It had been pending for 3 months. The wingman that finally finished it did it with a runtime of 2 million seconds.

[cineon_lut]

I just had one of these validate this week. It had been pending for 3 months. The wingman that finally finished it did it with a runtime of 2 million seconds.

So true, when mine validate all get a nice new badge and a gold status on wuprop! They're just sitting in the pending pile now. This one is going to be a while!


Vic (mobile)

[Slicker]

Their status page shows that some candidates have been now searched up to 7.8 million digits. They certainly seem overdue for a prime


Ummmmm... if primes followed a pattern and could be predicted, there wouldn't be the need for a search, would there? If you roll a die three times in a row and get a two each time, what are the odds of getting a two on the next roll? One in six. Previous results have no effect on the odds of the next roll. What we know from the patterns is that as numbers get larger, primes tend to be further and further apart. But, we can't extrapolate when the next one will be found as there is no real way of knowing that. Given that the number of digits doubled since the last prime was found, one begins to to question whether the 17 or bust concept holds true for really large numbers or whether it was just a coincidence that it worked for smaller numbers. Then again, a prime could be found at any time. Then again, the odds can't be any worse than stumbling upon a number which may prove the collatz conjecture wrong.