Team StarQwest's hourly information report.
 

Hey everyone out there in primeland.:wink: I began team StarQwest about one year ago, and have tested 13 exponents for primality. I now want to grow my team (which only has two computers now) into a mass effort in the search for the ten million digit prime. Anyone is welcome. Any test (factoring, double checking, LL) is welcome. My goal is to put team StarQwest in the top 100 within a year. The team currently has 77 CPU years. #100 currently has 1,620 years. The more members, the better. Team StarQwest already has gained a piece of fame by testing the largest mersenne number EVER: 2^100,000,007 - 1. That test will finish in mid December of this year. Yxine is currently double checking that number, and his double check will finish in late December - early January. Whether you're a new member just joining GIMPS and looking for other members just like you, or whether you've been testing for many years and are interested in becoming part of a team, team StarQwest shoots for success.

Join now by responding in this thread or sending me a private messege. I will then give you the instructions and password to put into Prime95 to become an official member of team StarQwest!

M100,000,007
 

Just for the record, team StarQwest has now reached iteration 75 million on M100,000,007! I am three-fourths of the way done!!! 4.3 million more iterations to go before I surpass the world record on the most iterations performed on a single exponent (current record is 79,299,959)! The entire test will finish on December 14.

Team StarQwest has now surpassed the world record of 79,299,959 iterations on a single exponent!!! 20.7 million more to go. I will finish December 11 - 14!

congratulations, well deserved
[QUOTE=StarQwest;88933]Team StarQwest has now surpassed the world record of 79,299,959 iterations on a single exponent!!! 20.7 million more to go. I will finish December 11 - 14![/QUOTE]

M100,000,007
 

Team StarQwest has now reached 90 million iterations on M100,000,007. The Largest LL test EVER performed to date will finish in 30 days!

7 Easy Steps to Become a Member of Team StarQwest
 

Team StarQwest has had incredible success with M100,000,007, and as an effort to grow and expand the team, I am going to list the steps required to join and make them open to anyone who wants them. This also means (as the team name and password are given) that anyone will be able to view the team's account report and see any exponents being tested by the team. If you have any questions, feel free to post here in this thread or send me an e-mail (my e-mail address is listed under my personal account report on mersenne.org), and if you decide to join the team, please post here or send me a private message letting me know of your membership. Team StarQwest is open to anyone who owns a computer and loves primes, so here it is.

Team StarQwest Goes Public!
 

I now have my own website, with a page on Team StarQwest and GIMPS. Here is the URL address:

[url]http://www.starqwest.org/page4.html[/url]

M100,000,007 Final Results
 

The test is complete! Unfortunately, it is not prime. However, I will be posting more information on my website, [url]www.starqwest.org[/url], as well as screenshots of Prime95 finishing the test, all known interim residues, and some other neat things discovered along the way.

[QUOTE="StarQwest"]The test is complete! Unfortunately, it is not prime. However, I will be posting more information on my website, [url]www.starqwest.org[/url], as well as screenshots of Prime95 finishing the test, all known interim residues, and some other neat things discovered along the way.[/QUOTE]Cool, so now with your residue posted I can spoof a message to the server and instantly get 79 years credit on my account :)

PS: Just kidding guys, but someone might do it. Maybe you should obscure the last two hex digits of your residue.

It won't work. The test didn't credit it to my account automatically because it is over 79.3M. I had to e-mail George Woltman and have him manually credit my account.

Check out my website, [url]www.starqwest.org[/url]. It has some great new additions on the Team StarQwest page as well as a bonus pi page.

New milestone for Team StarQwest!
 

Team StarQwest will reach 200 CPU years by the end of February! My goal is to reach 1000 and to have dozens of members within a year. Instructions for joining are located at [url]http://www.starqwest.org/page4.html[/url]. Its free, and Team StarQwest is quickly moving into higher ranks.

Congratulations on being the first to test a 100M+ exponent, though. :)

Thanks.

Am I right in assuming that the double check verified
your test of 100,000,007 and that the extra row on the
Status Page shows this one doublecheck >79,300,000 in
all its splendid isolation?

David

[QUOTE=davieddy;95602]Am I right in assuming that the double check verified
your test of 100,000,007 and that the extra row on the
Status Page shows this one doublecheck >79,300,000 in
all its splendid isolation?[/QUOTE]

No. My bad. I'll fix the status page next update.

[QUOTE=StarQwest;95402]Team StarQwest will reach 200 CPU years by the end of February! My goal is to reach 1000 and to have dozens of members within a year. Instructions for joining are located at [url]http://www.starqwest.org/page4.html[/url]. Its free, and Team StarQwest is quickly moving into higher ranks.[/QUOTE]

I can't get to your site :(

Luigi

Edit: my problem. Now I got conneted

I see StarQuest and Yxine are both testing numbers over 100,000,000 (all asterisks on the report). Any hints which ones?

[QUOTE=petrw1;96925]I see StarQuest and Yxine are both testing numbers over 100,000,000 (all asterisks on the report). Any hints which ones?[/QUOTE]

Yxine is double checking M100,000,007. I have reserved M332,192,831 (the first 100 million digit mersenne number) and will begin trial factoring it tomorrow night (notice the number of days remaining for this test: it is >5,200!)

[QUOTE=StarQwest;96938]I have reserved M332,192,831 (the first 100 million digit mersenne number) and will begin trial factoring it tomorrow night (notice the number of days remaining for this test: it is >5,200!)[/QUOTE]

My, aren't we ambitious..and trusting that the software will work for a number that big. 14+ years - Assuming the project continues, I suspect every number below 79,300,000 will be complete by then and I suspect that will be worth a goodly number of points.

[QUOTE=petrw1;97006]My, aren't we ambitious..and trusting that the software will work for a number that big. 14+ years - Assuming the project continues, I suspect every number below 79,300,000 will be complete by then and I suspect that will be worth a goodly number of points.[/QUOTE]

It won't really take 14 years. Even if I kept using my current computer for the entirety of the test, it would finish in just under 11 years. However I fully expect to upgrade my computer in about a year, and by then, computers will be much faster and will be able to complete the test in much less time. I would say around 3-4 years max for this test to complete. When it finishes, it will be worth ~900 CPU years!

M332,192,831
 

I just finished trial factoring M332,192,831 to 2^75. I am now trial factoring to 2^76. So far, no factor has been found. Check out the "Team StarQwest" page of [url]www.StarQwest.org[/url] to see my current status.

Trial factoring to 2^76 on M332,192,831 is now 50% complete. Still no factor found. I will post updates on my website by this evening.

Trial factoring to 2^76 on M332,192,831 is now complete. Still no factor found. I am now trial factoring to 2^77, which will take approximately 7 weeks. If no factor is found, I will begin P-1 testing, which will take approximately 2 months. If still no factor is found, I will begin the LL test!!!

Team StarQwest Updates
 

Team StarQwest has now surpassed 200 CPU years of LL testing!!! The team has successfully tested 20 mersenne numbers, all of which have turned up composite. Trial factoring on M332,192,831 is now 20% complete to 2^77 (54.3% complete overall). This will finish by the end of April, and if no factor is found, I will begin P-1 factoring. I have posted all the updated information on [url]www.StarQwest.org[/url], as well as some other cool stuff, so be sure to check out the entire site. My StarQwest site contains all the info for the team and prime number testing, as well as other neat stuff, such as spacecraft designs and a musical score I composed. You can also download one BILLION digits of pi (3.1415926535...) if you so please, which took me over 17 hours of computer time to calculate!:geek: I have also fixed all mistakes and dead links that I found, so there shouldn't be any errors.

M332,192,831 update
 

Trial Factoring on M332,192,831 is now 50% complete to 2^77 (71.4% complete overall). It will be finished in approximately 25 days and will be worth approximately 20 CPU years for factoring! Wow, 3 months of trial factoring for a single exponent! I could have LL tested 3 ten million digit numbers in this time. P-1 factoring will likely take about 2 months. Even if no factor is found, this will put Team StarQwest around rank 350 for factoring!!! I would like Team StarQwest to be in the top 100 for factoring eventually and find lots of factors, so anyone interested in factoring work is more than welcome to join the team. Finally, if no factor of M332,192,831 is found and I begin the LL test, it will still be open for double check. Anyone (member or not) is more than welcome to double check this number.

No updates since April 3??

LL testing was expected to start a week ago??

Any problems??

100,000,007 doublechecked!
 

Comparing the residue posted at your website with the one listed in cleared.txt report, there is a LARGE chance that your exponent has been sucessfully DCed.
Congrats to you and yxine!

[QUOTE=petrw1;105248]No updates since April 3??

LL testing was expected to start a week ago??

Any problems??[/QUOTE]

I have had some technical difficulties with my website, so I couldn't post anything. However, I did finish trial factoring M332,192,831 to 2^77 and no factor was found. It is now P-1 factoring, which will take a lot longer than I thought, which is unfortunate. It will take another two months to finish stage 1 and I don't know how long to finish stage 2. I will post updates on the website when I get it working again.

I barely have the patience to stay on a project for a month, if you manage to be the first one to complete a 100-million digit Mersenne number test, I will be EXTREMELY impressed with your patience.

[QUOTE=jasong;105284]I barely have the patience to stay on a project for a month, if you manage to be the first one to complete a 100-million digit Mersenne number test, I will be EXTREMELY impressed with your patience.[/QUOTE]

Here is the StarQwest Mission Statement, from [url]www.StarQwest.org:[/url]

"The name of StarQwest is a symbol for pride and achievement, for glory and desire, for learning and truth. StarQwest and anything bearing its name will strive for greatness, test the boundaries, and go where no one has gone before. It is an honor and a responsibility, a vision and a dream. To set sail for the stars, and begin the quest of a lifetime, searching for greater understanding and majestic scientific endeavors, and for an amazing journey unlike anything ever seen. StarQwest reaches for the stars and beyond, to learn, to inspire, and to explore."

I am currently laying the groundwork for this test, which will be the largest ever performed, by doing early factoring work. Considering into my calculations that over time I will be upgrading to faster computers, I can see this test finishing in around 4 or 5 years time. This is one of my pet projects. I was the first to test a number larger than 100M. That number was 2^100,000,007-1, which contains over 30.1 million digits. Yxine became the first to successfully doublecheck that number just a couple of days ago. Both of our tests (mine and Yxine's) have now proven conclusively that M100,000,007 is not a prime number. This is the [I]first[/I] time that anything definate can be said about the primality of a number of this size. I don't know who will be willing to doublecheck M332,192,831, although I'm sure that someone will be brave enough to take up the task for themselves. It will be worth about 900 CPU years- the equivalent of ~120 ten million digit tests!

M332,192,831
 

I have resolved all the technical difficulties with my website and uploaded all the data about M332,192,831! P-1 stage 1 factoring is now 6.57% complete and will finish in about 10 weeks. Stage 2 will likely take 3-4 months. Ugh. There will likely be a cold day in hell before I know the primality status of this number.:wink:

[QUOTE="StarQwest"]This is the first time that anything definate can be said about the primality of a number of this size.[/QUOTE]Hehe, just for fun I make a definite statement about the primality of a really really big number.

MMMMMMMM4 is not prime! And indeed no matter how many more M's you might like to add you will never ever get a prime number.

[size=0]okay, I know, you probably meant a number with a prime exponent, right?

I'll go back to my corner now. Thanks for reading.[/size]

[QUOTE=retina;105300]Hehe, just for fun I make a definite statement about the primality of a really really big number.

MMMMMMMM4 is not prime! And indeed no matter how many more M's you might like to add you will never ever get a prime number.
[/size][/QUOTE]

And just for fun I make a definite statement, that googolplexplexplexplexplexplexplexplexplexplexplex is also not prime - no matter how many plex's you may add... :lol:

[QUOTE=Andi47;105302]And just for fun I make a definite statement, that googolplexplexplexplexplexplexplexplexplexplexplex is also not prime - no matter how many plex's you may add... :lol:[/QUOTE]

Yes, and all its factors are a googolplexplexplexplexplexplexplexplexplexplexplex twos and a googolplexplexplexplexplexplexplexplexplexplexplex fives.


[QUOTE=retina;105300]Okay, I know, you probably meant a number with a prime exponent, right?

I'll go back to my corner now. Thanks for reading.[/QUOTE]

Yes, that is correct. Although that is a good one. I hadn't thought of that. However, here is another one for you:

Is INFINITY prime or not prime? After all, if you divide it by any number other than one, you still get infinity. But, if you divide it by itself, you get an indeterminate form...

[QUOTE=StarQwest;105328]Is INFINITY prime or not prime? After all, if you divide it by any number other than one, you still get infinity. But, if you divide it by itself, you get an indeterminate form...[/QUOTE]

Hmmmm... A prime is a positive [b]integer[/b] p with p > 1 and p is only divisible by one and itself.

Is infinity an integer?
Infinity is not a real number, so it is not an integer (the set of integers is a subset of real numbers), and so it is not prime.

[QUOTE=Andi47;105331]Hmmmm... A prime is a positive [b]integer[/b] p with p > 1 and p is only divisible by one and itself.

Is infinity an integer?
Infinity is not a real number, so it is not an integer (the set of integers is a subset of real numbers), and so it is not prime.[/QUOTE]

Correct!!! Although I was actually joking, that is a pretty good proof.:tu:

Infinity would be the ultimate composite number if it were an integer.

[QUOTE=cheesehead;105334]Infinity would be the ultimate composite number if it were an integer.[/QUOTE]

:wink:

M332,192,831 update
 

Unfortunately due to the slowdown on P-1 factoring, stage 1 is only 30% complete. It will take until the end of September to finish Stage 1, and stage 2 will not be performed because Prime95 set B1=B2. Therefore, if no factor is found, I will finally be able to begin the LL test. This test will credit Team StarQwest with 900 CPU years, but it will be a few years before it completes. Anyone is welcome to join (you don't have to test 100 million digit numbers although you could if you wanted to). I am looking for anyone interested in trial factoring and LL testing or double checking "smaller" ten million digit numbers.

M332,192,831 Update
 

P-1 factoring is complete! No factor was found, so I began the LL test. I am currently at iteration 7500. It takes about 0.950 seconds per iteration, so I should complete about 90,000 per day. Check out [url]www.StarQwest.org[/url], as I have updated it with the new information.

Hmmm according to my calculations it will take almost EXACTLY 10 years at .950 per iteration

Let's all come back here again on 26-Sep-2017 and check out the result.

@StarQwest, such incredible patience, :respect:

We could make a nice poll, guessing the date when it finishes.
I will take a first shot:

21st april 2012

:cat:

With advancements in processors over time (he already upgraded to a dual core) I will go for
April 3, 2011 (my next prime number birthday)

Which by the way will also be the same day that all first time and double check tests are done for all mersenne numbers under 10 million digits

Which by the way will be when the 47th Mersenne prime will be found

P.S. It is exactly one year since this thread started

M332,192,831
 

I have now converted the testing of M332,192,831 over to the new v25.5 of Prime 95 because I can use both cores of my Pentium D to test this number. My per iteration time is now 0.640 seconds, and I should finish no later than Summer 2014. However, I will obviously be getting newer, faster computers between now and then, so I am hoping to finish by 2010 or 2011. This test will be worth around 900 P-90 CPU years when it finishes, but unfortantely, since I am the only one testing, I will fall dramatically in the rankings between now and then. I am looking for as many new members as possible. Anyone is welcome!

One way to look at this is:
If you take your 900 years of work and assume it takes 3 more years to complete ...

(I think I can fairly extrapolate on todays rate because while you will upgrade and get faster over time so will everyone else)

... you will be averaging about 74 P90 years each 90 days.

Based on the current TPR top 500 producers report that puts your throughput on par with about the 200th Top Producer.

But remember that he/she and everyone else will gain as well over the next 3 years some more, some less (excluding those who quit).

Coincidentally from Feb 06 to Feb 07 the top 200 person did 305 P90 years of work ... extrapolating 3 years of works out to just over 900 P90 Years.

Not sure this last math is relevant except to say you can't expect to get to top 200 with the current parameters.

Let's try another track ...

You are currently 1543 with 235 P90 years. All things being equal you will have 1,135 P90 years 3 years from now.

From Feb 06 to Feb 07 the #500 person gained 155 P90 Years (extrapolate to 465 in 3 years).

Today #500 has 616. Add to that 465 and you get 1081.

I guess I will sum this up with a wild @$$ guess that 500th place is reasonable for you when your current 100M digit exponents completes in 3 years.

M332,192,831 LL Testing
 

LL testing for M332,192,831 is well underway. I am using v25.5 of Prime95 and getting 1 iteration per 0.650 seconds, or roughly 130,000 iterations per day. I fully expect to reach 1,000,000 iterations late next Thursday night or early next Friday morning. I will post interim residues every 1,000,000 iterations on my website just as I did for M100,000,007, just in case anyone wants to doublecheck this. (WARNING: It will take nearly 10 years on a Pentium 4 processor, nearly 7 years on a Pentium D, or about 3-4 years on a Quad-Core Extreme Processor).

M332,192,831 Update
 

I have reached iteration 1,000,000! I am getting an average per iteration time of 0.6446 seconds, so I should complete one million iterations every 7.5 days or so. I updated [url]http://www.starqwest.org/page4.html[/url] with this and inserted a screenshot of Prime95 reaching this milestone. Note that I am using the NEW version 25.5 (currently a Beta version) so I can use both cores of my Pentium D on this test. The old v24.14 can only use one core and gets a per iteration time of 0.960 seconds.

How to register that large numbers to LL test?
 

Hello Starqwest. I am LL testing exponents in the range 50M-60M because I guess the next Mersenne prime will be in that range. But I don't know how to register such an exponent for Primenet. I always get an unknown error 3 in the client when it communicates with primenet and I even emailed George Woltman, but with no success until now. So I'd be glad if you could tell me how you managed to register that large numers. Since the current progress is shown on the status page I assume your client can communicate automatically after registration!? Thanks.

[QUOTE=multisync;118277]Hello Starqwest. I am LL testing exponents in the range 50M-60M because I guess the next Mersenne prime will be in that range. But I don't know how to register such an exponent for Primenet. I always get an unknown error 3 in the client when it communicates with primenet and I even emailed George Woltman, but with no success until now. So I'd be glad if you could tell me how you managed to register that large numers. Since the current progress is shown on the status page I assume your client can communicate automatically after registration!? Thanks.[/QUOTE]

I just put the line
"Test=332192831,77,1"
in the worktodo.ini file and it worked. Note that this line will trial factor 2^332192831-1 to 2^77 and do P-1 factoring before the LL test. If you just want to do a LL test, input the line "Test=332192831," or whatever exponent you want to test. I e-mailed George about this test and he put it into my account for testing, but you shouldn't need to do this for numbers less than 79,300,000.

[QUOTE=StarQwest;118693]"Test=332192831,77,1" ... this line will trial factor 2^332192831-1 to 2^77 and do P-1 factoring before the LL test.[/QUOTE]This is not true : the second parameter tells Prime95 that trial factoring HAS ALREADY BEEN DONE to 2^77 and the third parameter that the P-1 factoring HAS ALREADY BEEN DONE. To achieve what you want the line should be Test=332192831,29,0. (The smallest number of the form 2*k*332192831+1 is 29,3 bits so anything below 30 will test that potential factor of M332192831. This is assuming there has been no previous trial factoring you can rely on...)

Jacob

I think you're both right and wrong at the same time. I've noticed that version 24 and the beta version 25 do things a little differently. For instance, if you don't want to get work from the server, in version 24.14 you choose "Just Stress Testing," but in version 25.5 you say you're joining GIMPS, and then when it says "Get work from server(Y)" you say "N" and then you can do your own thing.

M332192831 Update
 

I am still running iterations on this number, but I am not running them 24/7 like I used to because it puts so much wear and tear on my machine. I try to run them whenever I have my computer on anyway, but it is incredibly slow going. I should hit 4 million iterations in a few days, but I won't hit 10 million until sometime in March. I'll try to keep posted more often, but trust me, this is going to take a loooooooooooooong time.:wink:

P.S. You can check my status on the status page of [url]www.mersenne.org[/url], but note that this test is at the very bottom and is represented by ********.

[QUOTE=StarQwest;123037]
P.S. You can check my status on the status page of [url]www.mersenne.org[/url], but note that this test is at the very bottom and is represented by ********.[/QUOTE]

That's too bad, because some people may have trouble viewing it at all. Some browsers stop if a webpage has not finished loading within 60 seconds; and the status of the highest exponent is always the last to load. So some people might not be able to view it at all.

Anyway, I was able to view it. At 14:17 UTC on Jan. 18, 2008, you were at iteration 3,446,919.

Interestingly, there are two other LL tests going on that are well above the main leading edge. S709129 is testing M59,999,999 and andreasbomke is testing M59,289,959. The next highest after that is less than M42,000,000.

[QUOTE=StarQwest;123037] I am not running them 24/7 like I used to because it puts so much wear and tear on my machine. [/QUOTE]

Unless it is a laptop, you don´t have to worry about that. Trust me, I have been crunching for GIMPS for 6 years, and have never damaged a machine, even though I have had them running 24/7. I am still using on a daily basis an Athlon XP1800 that is running the client from the very first day.
If your concern is the electricity bill, that´s another story. I have lately reduced the crunching effort for that reason.

I agree....I have been running my P4 2.4 Ghz Desktop non-stop (unless there is a power failure of I have to reboot for a Windows patch) on GIMPS for over 4 years without any sign of wear and tear.

As of 20:20 UTC, Jan. 27, 2008, Starqwest was at iteration 4,141,425. That comes out to 694,506 iterations in 13,323 minutes, a per iteration time of 1.151s. At that rate, it will take him until 14:36 UTC on Feb. 7, 2017 to finish the 100 million digit test.

I am currently running the test about 15 hours a day just to keep working on it, although I will likely get a new computer this summer which is much, much faster. I hope to finish by 2011 at the latest. Just out of curiosity, how much faster is the Quad-Core Processor than the Pentium D when running version 25?

It depends on which QuadCore and which Pentium D, but clock of clock, on a single core a core2 with adequate memory is at least 40 % quicker. (for M333000000 a core2 3GHz will take less than 5 years and a P4 3GHz 7 years.)

Jacob

AFAIK, the latest version of the Prime95 client has gone multithreaded, which allows you to run a single exponent in several cores. If you get a 4-core 45 nm Penryn, I think you will run the test MUUUCH faster than on a Pentium D, even running the Penryn at stock speed. Add the fact that Penryns are excellent overclockers, and you´ll end up with a top notch cruncher.
Or you might as well wait for Nehalem to come out...:yawn:

Just wanted to say that you might want to inform the [url=http://www.geocities.com/onehundredmdpp/index.htm]100 Million digit prefactor project[/url] of your intentions if you haven't already.

Also, if you are seriously considering getting a new CPU, wait just a little more until Intel releases some 45nm octocores, one of those will really move your test along.

[QUOTE=lavalamp;127865]...until Intel releases some 45nm octocores...[/QUOTE]

Does Intel indeed plan to release octocores in the near future?

Sadly, now I check again, it seems that 8 core Nehalems won't be here until 2009, however there should be some native quad core (far better then double dual core design) Nehalems this year.

[url=http://en.wikipedia.org/wiki/Nehalem_%28microarchitecture%29]Nehalem!!![/url]

[QUOTE=lavalamp;127900]Sadly, now I check again, it seems that 8 core Nehalems won't be here until 2009, however there should be some native quad core (far better then double dual core design) Nehalems this year.

[url=http://en.wikipedia.org/wiki/Nehalem_%28microarchitecture%29]Nehalem!!![/url][/QUOTE]

Does 4 cores / 8 threads for Bloomfield and Lynnfield processors mean, that these are hyperthreaded Quad-cores?

[QUOTE=Andi47;127905]Does 4 cores / 8 threads for Bloomfield and Lynnfield processors mean, that these are hyperthreaded Quad-cores?[/QUOTE]Yes, they brought back hyperthreading.

AMD FireStream
 

How will AMD's FireStream compare to the quad cores? I read it will process at up to 500 GFLOPs.

The M332192831 test has been silent for nearly two weeks. The last update was at 28-Feb-08 20:16. The iteration number was 6,036,045.

Are you planning to get the 8-core "Nehalem-EX" when it comes out next year? :D

Has there been any recent progress on this test?

His last update to Primenet was yesterday:[quote=http://mersenne.org/primenet/status.txt]******** 78 6817889 78.7 1942 60.3 18-Apr-08 15:53 31-Jan-08 14:22 StarQwest StarQwest1[/quote]He's 2.05% complete.

M332192831 Update
 

For those who are wondering why progress has stopped on M332192831, I have postponed further testing on this number until I get a faster computer. I am running a 2005 PC, which only contains a single Pentium D processor on this test. I will likely get a new computer later this year. To understand my reasoning for this, consider the following:

Imagine a spacecraft set to go to Alpha Centauri with an estimated travel time of 1,000 years. Over that time span, far faster spacecraft would be invented, and a craft invented 100 years later could possibly get there in 100 years, or 800 years before the original spacecraft. Similarly, my PC is simply to slow to effectively test M332192831, and any time lost in the meantime will more than be made up for when I someday get a faster computer. Eight core processors will come out early next year, which would shorten the testing time to approximately 2 years, as compared to 10 years on my current machine. So, the way I figure it, if I tested nonstop for the next 6 months on my machine, I would finish only 1.2 months sooner if I got an 8-core PC next year. It hardly seems worth it to me.

Please understand that I have NOT in any way given up on this test, and plan to resume it in the near future. It is just not practical at the current time, so I will resume testing 10-15 million digit numbers.

however,
[quote=retina;154483]No. You should always start your job now. Because when new hardware comes along you can transfer your work to the new machine and continue the job.

If you do nothing now then the job will take longer later, even though the new hardware may be faster you still lose the initial computing time.[/quote]
in the spacecraft example:

if you wait: it takes 1000 years to go to alpha centauri right now. in 100 years, it will take only 100 years to get to alpha centauri. if you start in 100 years, you get there in 200 years (100 years waiting, 100 years traveling)

whereas if you start now: in 100 years, you will have traveled 10% of the distance. you upgrade your spacecraft to a better one. as mentioned above, in 100 years the better spacecraft will take 100 years to get to alpha centauri from earth; however, you have already traveled 10% of the distance, leaving you with 90% left, or 90 years in the new spacecraft. so you continue in the upgraded spacecraft and arrive at alpha centauri 190 years after leaving earth now (100 years in first spacecraft, 90 years in upgraded one).

however, I do understand why you might stop testing; it is just too slow to wait months or even years for a result...

[QUOTE=starrynte;154596]however,

in the spacecraft example:

if you wait: it takes 1000 years to go to alpha centauri right now. in 100 years, it will take only 100 years to get to alpha centauri. if you start in 100 years, you get there in 200 years (100 years waiting, 100 years traveling)

whereas if you start now: in 100 years, you will have traveled 10% of the distance. you upgrade your spacecraft to a better one. as mentioned above, in 100 years the better spacecraft will take 100 years to get to alpha centauri from earth; however, you have already traveled 10% of the distance, leaving you with 90% left, or 90 years in the new spacecraft. so you continue in the upgraded spacecraft and arrive at alpha centauri 190 years after leaving earth now (100 years in first spacecraft, 90 years in upgraded one).
[/QUOTE]

Hmmm... maybe a spacecraft is a bad example:

When you start now, you can't upgrade the spacecraft halfway to Alpha Centauri, because you need to be on earth to buy a new warp engine and a dock to properly mount it. So you will be faster when you wait until the docks in Utopia Planitia are building spaceships with warp drives. :wink:

But with Mersenne numbers, you [i]can[/i] upgrade your hardware halfway through the LL-test and thus you will be faster when you start now.

[quote=Andi47;154616]But with Mersenne numbers, you [I]can[/I] upgrade your hardware halfway through the LL-test and thus you will be faster when you start now.[/quote]
Right, but at what cost? If you started it years ago on a 386 when it was state-of-the-art and upgraded it, say, every two years, you'd finish it some minute amount faster than if you started it on, say, state-of-the-art for 2010 hardware, but it would take a very long time, time that could be better spent searching for primes more reasonably sized for your hardware.

[QUOTE=Mini-Geek;154626]Right, but at what cost? If you started it years ago on a 386 when it was state-of-the-art and upgraded it, say, every two years, you'd finish it some minute amount faster than if you started it on, say, state-of-the-art for 2010 hardware, but it would take a very long time, time that could be better spent searching for primes more reasonably sized for your hardware.[/QUOTE]But the work still needs to be done. It makes no difference what order you do it. Just run the smaller numbers later on your faster hardware. The overall finish time for all the work will be about the same, only the order of completion is different.

(in the spacecraft example, upgrading is defined as:
new spacecraft is manufactured on earth (analogy of newer computers)
the spacecraft is teleported to as far as where the old spacecraft went (the interim files from the old computer is copied to the new computer)
the new spacecraft continues on the journey (the new computer continues processing the exponent)
)
related to Mini-Geek's last post:
(all completion times based on benchmarks, which may be slightly inaccurate but will do for the purpose)

on a pentium 3 996 MHz, M43112609 will take about 307 days to finish. M20996011 will take about 70 days

on a pentium 4, M43112609 will take about 42 days to finish. M20996011 will take about 10 days to finish.

if you begin on the pentium 3 with M43112609, and later do M20996011 on the pentium 4, it will take a total of 317 days
if you begin with M20996011, however, and then do M43112609 on the pentium 4, it will take a total of 112 days

so, if you are an average person running prime95, you should test exponents suited to your computer. however, if you really do want to test very large exponents, you should start right now, for you can always upgrade your computer, but continue where your old computer left off.

I've been posting about this in the "Will Any Current 100M Digit LL Tests Finish?" thread ([url]http://www.mersenneforum.org/showthread.php?p=154662#post154662[/url]). Of course, if you have an unlimited amount of money, it makes sense to start as soon as possible and upgrade whenever new hardware comes out. A more interesting question is what you should do if you have only a finite budget for purchasing new hardware. In that case, you should time your upgrades carefully for maximum effect...

[QUOTE=starrynte;154654]related to Mini-Geek's last post:
(all completion times based on benchmarks, which may be slightly inaccurate but will do for the purpose)

on a pentium 3 996 MHz, M43112609 will take about 307 days to finish. M20996011 will take about 70 days

on a pentium 4, M43112609 will take about 42 days to finish. M20996011 will take about 10 days to finish.

if you begin on the pentium 3 with M43112609, and later do M20996011 on the pentium 4, it will take a total of 317 days
if you begin with M20996011, however, and then do M43112609 on the pentium 4, it will take a total of 112 days[/QUOTE]This comparison is not correct : the time untill the PIII is upgraded is not the same (in one case you upgrade after 307 days, in the other after 70 days.) If you upgrade after 190 days you need about 216 days in the 43M followed by 21M case and 215 days in the 21M followed by 43M case ; the differences might be explained by the approximation of the benchmarks. (Benchmarks measure a best itteration time, not an average itteration time, over a few runs the standard deviation might can be as high as 5 %.)[QUOTE=starrynte;154654]so, if you are an average person running prime95, you should test exponents suited to your computer.[/QUOTE]I agree with this though.

Jacob

[QUOTE=StarQwest;133478]For those who are wondering why progress has stopped on M332192831, I have postponed further testing on this number until I get a faster computer. I am running a 2005 PC, which only contains a single Pentium D processor on this test. I will likely get a new computer later this year. To understand my reasoning for this, consider the following:

...

Please understand that I have NOT in any way given up on this test, and plan to resume it in the near future. It is just not practical at the current time, so I will resume testing 10-15 million digit numbers.[/QUOTE]

It seems like that this exponent has been assigned to another person.

[CODE]
Exponent Status Data
332192831 No factors below 2^77
Assigned LL testing to "KYOJI_KAMEI" on 2009-05-18
History no factor to 2^74 by "George Woltman" on 2008-06-16
[/CODE]

The website

[URL="http://www.starqwest.org"]www.starqwest.org[/URL]

also seems not to be available.

Gary

Confirmed: [url]http://downforeveryoneorjustme.com/www.starqwest.org[/url]