Join PrimeGrid
Returning Participants
Community
Leader Boards
Results
Other
drummers-lowrise
|
Message boards :
Generalized Cullen/Woodall prime search :
How are we doing on sieve?
| Author |
Message |
|---|
mackerel Volunteer tester
Send message
Joined: 2 Oct 08
Posts: 2652
ID: 29980
Credit: 570,442,335
RAC: 7,565
                              
|
|
I ran it a bit to compare CPUs recently, and haven't switched away yet. That reminded me, if my memory is correct we still need more work before the LLR project could be started effectively.
http://www.primegrid.com/stats_gcw_sieve.php
Looking at the above link I can see some bases have already reached optimal and no more work is running for now, and for most of the others there's around 4500 tests with 1500 outstanding. Is that indicative of remaining work before LLR can start, or is that just the work that is currently generated, and more will be generated until stopped? | |
|
JimBHonorary cruncher
Send message
Joined: 4 Aug 11
Posts: 920
ID: 107307
Credit: 989,666,286
RAC: 31,991
            
|
|
It's just the work that's currently generated. Some bases, like 69, will be sieving for a very long time yet. | |
|
|
|
|
A shame this "mysterious entity" soon to provide so much computing power didn't choose this sieve. | |
|
mackerelVolunteer tester
Send message
Joined: 2 Oct 08
Posts: 2652
ID: 29980
Credit: 570,442,335
RAC: 7,565
|
|
I'll see if I can keep some power on this then... a good use for Ryzen... | |
|
 Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14037
ID: 53948
Credit: 477,954,885
RAC: 338,184
                  
|
A shame this "mysterious entity" soon to provide so much computing power didn't choose this sieve.
It was our choice.
At the time decision was made, GCW sieve was not yet available, and therefore wasn't an option.
____________
My lucky number is 75898524288+1 | |
|
|
|
|
So bases 109, 41 and 73 were all turned off during the course of the challenge at the beginning of the year because they had been sieved far enough to allow LLR to start ("first optimal" depth.)
With the addition of the anonymous computers to this project while TRP finishes cleanup, I see that all three of those bases have many many tasks currently being processed.
Does this mean that all of the bases have now reached the "first optimal" depth that will allow LLR to start up soon? | |
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14037
ID: 53948
Credit: 477,954,885
RAC: 338,184
|
So bases 109, 41 and 73 were all turned off during the course of the challenge at the beginning of the year because they had been sieved far enough to allow LLR to start ("first optimal" depth.)
With the addition of the anonymous computers to this project while TRP finishes cleanup, I see that all three of those bases have many many tasks currently being processed.
Does this mean that all of the bases have now reached the "first optimal" depth that will allow LLR to start up soon?
That's exactly what it means. We're now sufficiently sieved to open up GCW-LLR. Jim and I are making the final preparations to bring it live. I don't have a date yet, but I expect it to be somewhere between a few days and a few weeks. Amongst other items, I believe there's still a handful of GCW tasks running on PRPNet. We need to let those finish or timeout.
____________
My lucky number is 75898524288+1 | |
|
|
|
|
There are still GCW Sieve tasks still available here at PrimeGrid. I'm currently running 8 of them. | |
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14037
ID: 53948
Credit: 477,954,885
RAC: 338,184
|
There are still GCW Sieve tasks still available here at PrimeGrid. I'm currently running 8 of them.
And there will be for quite a few years. GCW-Sieve is not shutting down anytime soon.
There's been 3, maybe 4 different flavors of GCW running in the last year, and perhaps that is causing some confusion:
1) GCW-LLR on PRPNet: This is now turned off after being replaced by (3) below.
2) GCW-Sieve on BOINC, initial sieving: The early part of the sieve was to make sure we were sufficiently sieved to start running GCW-LLR on BOINC. This meant a sufficient sieve depth to support testing GCW-LLR up to candidates of 4 million digits. Once that depth was met, GCW-LLR was stopped on PRPNet (1) and started on BOINC (3). The sieve kept running however -- see (4) below.
3) GCW-LLR on BOINC: Started once (2) above was sieved to a sufficient depth. Replaced (1) above.
4) GCW-Sieve on BOINC, continuing sieving: will continue to sieve GCW to a depth sufficient for testing candidates of 15 million digits.
(3) and (4) are currently active and will be active for years to come.
____________
My lucky number is 75898524288+1 | |
|
Message boards :
Generalized Cullen/Woodall prime search :
How are we doing on sieve? |
