WE MADE IT!!!!!!!!!!!!!!!!!!!!!!

[paulunderwood]

Don't forget to update the wiki

[MooMoo2]

WOW. I was away last night on a ski trip, and I definitely didn't expect this when I came back . Congrats, eric_v.

Anyway, here's my info:

Full name: Michael Kwok (my bio page is http://primes.utm.edu/bios/page.php?id=505 if anyone's curious)
Age: 18
Country: United States

TPS, Primegrid, eric_v, Skligmund, and gribozavr will get the credit. Also, Rytis, pacionet, me, LLR, and NewPGen will all be in the project's bio page.

[biwema]

Congratulations! This twin is really great!

We were really lucky, that it came so early. To be honest, I feel now a little bit guilty because I suggested sieving 25G instead 5G... I did not know that.

Now we have the little problem, that our new exponent is not sieved that far. Therefore a distributed sieving is a good idea for those who have plenty of memory.

My opinion is, that everyone can do whatever one wants. Sieving, Testing 333333 or even testing 195000 (here you can still find a record twin ).
I suggest move to 333333 even it is not sieved enough.

Unfortunately I will be on holidays for until end of january. So I do do not have enough time to help coordinate now. that should not be a problem as there are volunteers around who are interested. You have already the most up to intermediate file.
Of course I will continue sieving during my vacation that we can remove many candidates as early as possible (The last weeks I stopped sieving, because I worked on a vertical proth search).

When I am back I will do my best to help coordinating (if necessary) and maybe writing some software for merging, and optimized compressing the dat. Maybe I could even try to improve the sieving process...

One more note: Please do not truncate the dat file too early. We can cut off 1 ore more G, after it is complete (sieving a larger range does not take more time, but maybe some people are interested in updating an file of candidtes to the most up to date sieving level after LLRing half way)

[pacionet]

Today, January 15, 2007, the Twin Prime Search (TPS) project in collaboration with PrimeGrid (BOINC platform) found the largest known twin primes, 2003663613*2^195000-1 and 2003663613*2^195000+1.
The two primes are 58711 digits long.
The discoverer is Eric Vautier, from France.
The credits of the discovery go to two projects : TPS and PrimeGrid and to the following people : Eric Vautier (France), Dmitri Gribenko (Ukraine), Patrick W. McKibbon (USA) .
The project has been coordinated by the following people: Michael Kwok (USA), Andrea Pacini (Italy) and Rytis Slatkevicius (Lithuania).
The discovery has been made using the LLR primality testing program and NewPgen sieving program.
Special thank to all the contributors.

[Flatlander]

I'm stunned! Congratulations to everyone!

[pacionet]

I upload the official announcement here:

http://www.twinprimesearch.org/official.pdf

[MooMoo2]

Originally Posted by pacionet

Today, January 15, 2007, the Twin Prime Search (TPS) project in collaboration with PrimeGrid (BOINC platform) found the largest known twin primes, 2003663613*2^195000-1 and 2003663613*2^195000+1.
The two primes are 58711 digits long.
The discoverer is Eric Vautier, from France.
The credits of the discovery go to two projects : TPS and PrimeGrid and to the following people : Eric Vautier (France), Dmitri Gribenko (Ukraine), Patrick W. McKibbon (USA) .
The project has been coordinated by the following people: Michael Kwok (USA), Andrea Pacini (Italy) and Rytis Slatkevicius (Lithuania).
The discovery has been made using the LLR primality testing program and NewPgen sieving program.
Special thank to all the contributors.

Beat me to it

Anyway, my "official press release" was a little different, but also similar:

-------------------------------------------------------

January 14, 2007:

Today, the Twin Prime Search (TPS) project, in colloboration with PrimeGrid, has found a twin prime. The primes this project found are 2003663613*2^195000-1 and 2003663613*2^195000+1, which are both 58711 digits long. They were found by Eric Vautier of France, and it surpasses the previous record world record of 51780 digits.

So, what exactly are twin primes, and what's so great about finding them?

Well, twin primes are just 2 primes separated by two. 3 and 5 are twin primes, and so are 29 and 31. However, they are extremely rare, and their density dramatically decreases the higher you go on the number line. In fact, it is currently unknown if there even are an infinite number of twin primes.

To search for large primes of this type, Michael Kwok created the Twin Prime Search (TPS) project, and soon after, Andrea Pacini created and maintained a website for the project. However, it was never certain that TPS would find a twin in a reasonable amount of time. For many weeks, it suffered from relatively low participation rates until Rytis, the founder of PrimeGrid, offered to help out TPS by allowing PrimeGrid users to participate in the project. PrimeGrid greatly increased the visibility of TPS, which brought in a greater number of participants. Although a PrimeGrid user did not find a twin, PrimeGrid tested more candidates than TPS.

Before finding the twin, PrimeGrid and TPS tested about a million candidates. Testing these candidates on a single PC would take years, so this timely discovery would not have been possible without the dozens of volunteers who contributed their spare CPU cycles to this project. Out of all of them, Skligmund and Gribozavr deserve special recognition and will share the credit for the twin's discovery with Eric Vautier. Skligmund tested the greatest number of candidates, while Gribozavr provided the project with pre-sieved candidates.

Hopefully, this will not be the last twin that TPS and PrimeGrid discovers. After this success, the projects are looking to find a twin of even greater size, which is one that is over a hundred thousand digits long. PrimeGrid will be used for doing the actual testing of candidates, while TPS will be used for providing PrimeGrid with pre-sieved ranges.

[Jens K Andersen]

I have updated the largest known simultaneous primes with credit to all the mentioned. Congratulations to all.

[jinydu]

Here is the full decimal expansion of for anyone who's interested. Obviously, the decimal expansion of would be the same, except the final 7 would be changed to a 9.

It took me quite a long time to make this file due to formatting problems. While copying from Mathematica to Notepad and then to Word, a lot of ugly spaces appeared that I had to manually delete, one by one. This decreased the size of the Word file from 33 pages to 18 pages.

For those who don't want to download the entire file, the first few and last few digits are:

141572625976528653369415...367988209288925295935487

where the ... indicates that 58,663 digits have been left out.

[Skligmund]

That a big freaking number!

[Flatlander]

The other one's bigger!

Maybe Perfectly Scientific Inc. will make a pair of posters.