| Author |
Message |
|---|
Reggie Volunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 10 May 14
Posts: 162
ID: 311759
Credit: 115,534,067
RAC: 0
                   
|
|
People have been asking about the "life" expectancy of the current projects. Below is a comparison of project groups. The time-frames are just references so you can compare the different groups. Take all of this with a grain of salt. :)
The conjectures in the 10+ group could easily take 100's of years if ever proven. Most of the other primality test projects (except for GFN) could technically go on forever simply by adding more work (n max increased.)
The sieves will finish before their LLR counterparts. However, they will be classified as suspended. The sieves could start back up if there are further advances in sieving efficiencies in the future or if more work is added (n max increased).
The Sophie Germain Search is by itself since a Twin and/or SG could be found at any minute. Since this is a quad sieve, there is a chance of finding both a Twin and an SG. However, our current plan is to continue to search for additional Twins and SG pairs, so this project will continue until we've tested all the candidates in our sieve file.
The Generalized Fermat Number searches will continue until the searches reach the limits of the GPU software.
I hope this helps.
10+ years
Extended Sierpinski Problem (LLR) will be tested to at least 50M or until all remaining k's have been proven prime.
Prime Sierpinski Problem (LLR) will be tested to at least 50M or until all remaining k's have been proven prime.
Seventeen or Bust (LLR) will be tested to at least 50M or until all remaining k's have been proven prime.
The Riesel Problem (LLR) will be tested to at least 50M or until all remaining k's have been proven prime.
The Sierpinski/Riesel Base 5 Problem (LLR) will be tested to at least 50M or until all remaining k's have been proven prime.
The Proth Prime Search (LLR), in some combination of PPS, PPSE, and PPS-MEGA, will be tested to at least 5M.
Proth Prime Search (Sieve) is currently sieving 6M-9M which will be sieved to 900P. We will then transition to 9M-12M which will be sieved to 1600P.
Generalized Fermat Number (except 17-Low) will be tested to the limits of the GPU applications, currently b=400 million.
GCW LLR: All candidates in the remaining bases of up to 15 million digits will be tested until a prime is found in each base.
321 Prime Search (LLR) will be tested to at least n=50M.
5 to 10 years
Cullen Prime Search (LLR) will be tested to at least n=25M.
Woodall Prime Search (LLR) will be tested to at least n=25M.
Sophie Germain Prime Search (LLR) will be tested at least until a Twin and/or SG prime is found, but will probably continue and search our entire SGS sieve file.
1 to 5 years
Sometime in 2021, give or take a year, PPS-LLR will catch up with where PPS-MEGA started. At that point, PPS will skip over the candidates already tested by PPS-MEGA, and continue searching k<1200.
Generalized Fermat Number 17-Low will be tested until b reaches the starting point of 17-Mega.
Less than a year
WW will run to its software limit which is currently 2^62. However, there is a decent chance the software limit can be increased to 2^63 or 2^64. If so, the lifetime of the project will double or quadruple, respectively.
PPS-DIV will at least test all the candidates currently loaded (up to n=9M), though we may decide to continue the project with a larger range of k's or larger n.
Unknown
AP27: Currently, the plan is to continue to run this until a program optimized for finding AP28s is available. If this changes, we will give the standard 30 day notice before suspending the project.
Suspended 1 May 2019
GCW Sieve has been sieved to optimal depth.
Suspended 22 April 2017
The Riesel Problem (Sieve) has been sieved to optimal depth.
Closed 13 May 2010 and Reopened 24 August 2016
AP26/AP27.
Suspended 1 September 2012
Cullen/Woodall (Sieve) has been sieved to optimal depth.
Suspended 22 April 2011 - Earth Day, reopened 16 March 2019, and suspended again 31 October 2020.
321 Prime Search (Sieve) has been sieved to optimal depth.
Suspended 13 October 2010, Reopened 28 June 2014, and suspended again 31 July 2016
PSP/SoB (Sieve) sieved to optimal depth. Restarted to sieve ESP. |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
                    
|
|
Thanks, John!
Can you inform us when one of the 1+ year projects is looking closer to completion than the others?
____________
|
|
|
|
|
|
When the PSP/SoB sieve end, any chance that you could startup some sieving in collaboration with the non-base 2 conjectures, in order to help them greatly advance? If no non-base 2 conjectural sieving is considered, would you do anything special, to encurrage sieving for TRP?
Also I've not really been able to get an answer anywhere, but are PG considering a collaboration or overtaking of some of the harder CRUS conjectural work, in order to push the amount of proven conjectures? Maybe you could if you doesn't want to do sieving, find a way to help (if Gary accept) start up the untested ranges... Anyway this is just my 2 cents, but I sure would like to know how PG stands on these questions :)
Regards
KEP |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
               
|
|
Ken_g6 wrote: Can you inform us when one of the 1+ year projects is looking closer to completion than the others?
Sure thing. :)
KEP wrote: When the PSP/SoB sieve end, any chance that you could startup some sieving in collaboration with the non-base 2 conjectures, in order to help them greatly advance? If no non-base 2 conjectural sieving is considered, would you do anything special, to encurrage sieving for TRP?
PrimeGrid currently has several projects in the PSA that can replace the PSP/SoB sieve. However, we are always open to suggestions...such as non-base 2 conjectures. As for TRP, it is already approaching 8P which is excellent progress for 6 month's work. :)
KEP wrote: Also I've not really been able to get an answer anywhere, but are PG considering a collaboration or overtaking of some of the harder CRUS conjectural work...
Although there is the possibility of collaboration with other projects, I'm not aware of anything with CRUS.
____________
|
|
|
|
|
|
Thanks for this thread =) It is very useful. I'm now moving my PC over to PSP sieve so I can get my bronze badge before the project ends!!
____________
My Top 5000 Primes:
110059!+1 is prime! (FPS)
14493618614235*2^666666-1 is prime! (SGS) |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1893
ID: 352
Credit: 3,141,488,980
RAC: 0
                          
|
Thanks for this thread =) It is very useful. I'm now moving my PC over to PSP sieve so I can get my bronze badge before the project ends!!
+1
Expect a bit less from me on PRPNet...going for ruby on PSP Sieve.
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
|
same here thanks for the thread for this. was a big question i was going to ask. sooner or later |
|
|
|
|
|
Let me dream here and correct me if I go astray.
These estimates are based on past history I assume over how many past years but with new entrants will bring more faster clockspeeds and cpus with multiple cores. Was (or is) this taking into consideration the increase in efficincy of newer cpus entering market places?
Am I correct that some sieves lend themselves to possible compliation for GPU (graphic cards) crunching that will greatly speed up solution times thus reducing project longevities?
Will it be possible to make LLR available to GPU's also.?
I haven't kept up with new cpus being developed but cpu's with 5G hertz or higher clockspeeds can't be to far in the future.
What I'm asking is, "Are we using projections based solely on PAST expierences and/or usng ANY predicted new (not too distant future) forecasts of new hardware that we soon be using.
Thoughts or comments anyone? |
|
|
|
|
|
I assume the estimates are based on the average amount of work completed per day(perhaps per week or per month) in the past for a certain sub project as of recently and the life expectancy is calculated based on the LLR range/Sieve depth that is desired. Trying to accurately adjust the forecast for future hardware/rush of new users/hosts would be difficult and perhaps unnecessary. I believe that CPU manufacturers are aiming for more cache and more cores instead of higher clock speeds as more cores is an easier/better solution. GPU crunching on other sieves besides PPS is possible but the question of whether using a GPU for a particular sieve is worth it(does the GPU complete a WU in a satisfactory amount of time?). GPUs perform better on protein folding/MilkyWay than they do on PrimeGrid (GPUs are better suited for those other tasks). LLRing on a GPU may be possible, but again the question is: is the completion time of a task reasonable?
____________
 |
|
|
|
|
Let me dream here and correct me if I go astray.
These estimates are based on past history I assume over how many past years but with new entrants will bring more faster clockspeeds and cpus with multiple cores. Was (or is) this taking into consideration the increase in efficincy of newer cpus entering market places?
Am I correct that some sieves lend themselves to possible compliation for GPU (graphic cards) crunching that will greatly speed up solution times thus reducing project longevities?
Will it be possible to make LLR available to GPU's also.?
I haven't kept up with new cpus being developed but cpu's with 5G hertz or higher clockspeeds can't be to far in the future.
What I'm asking is, "Are we using projections based solely on PAST expierences and/or usng ANY predicted new (not too distant future) forecasts of new hardware that we soon be using.
Thoughts or comments anyone?
You are correct--sieve programs lend themselves much more readily to being written for GPUs. Sieving mostly consists fo integer operations, at which GPUs excel; the double-precision floating point operations required by LLR are rather trickier.
Recent higher-end GPUs from nVidia, with compute capacity >2, have double-precision floating point support. In fact, an application has been developed for performing LL tests (the granddaddy of the LLR test--used by GIMPS for 2^n-1 numbers) on CUDA GPUs over at mersenneforum.org. This application currently runs at a speed approximately equivalent to 8 fast CPU cores working as one--which while very fast, is not nearly as big a speedup relative to the CPU as can be gained with sieving (as much as 50x on a high-end GPU).
I, among others, am hopeful that a CUDA LLR application will be developed (possibly based on the existing LL application, since the algorithms are similar) in the near future. However, this is by no means guaranteed since it depends on the willingness and ability of the developers to do so. |
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2165
ID: 1178
Credit: 8,777,295,508
RAC: 0
                                  
|
Recent higher-end GPUs from nVidia, with compute capacity >2, have double-precision floating point support.
This is also true of NVidia GPUs with compute capability 1.3.
____________
141941*2^4299438-1 is prime!
 |
|
|
|
|
|
Why in PSP sieve project is whole in range 84P - 90P? Will we not crunching this range?
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Why in PSP sieve project is whole in range 84P - 90P? Will we not crunching this range?
A test range was completed in 90P-91P. At this time, we do not intend to complete the 84P-90P range.
____________
|
|
|
|
|
Recent higher-end GPUs from nVidia, with compute capacity >2, have double-precision floating point support. In fact, an application has been developed for performing LL tests (the granddaddy of the LLR test--used by GIMPS for 2^n-1 numbers) on CUDA GPUs over at mersenneforum.org. This application currently runs at a speed approximately equivalent to 8 fast CPU cores working as one--which while very fast, is not nearly as big a speedup relative to the CPU as can be gained with sieving (as much as 50x on a high-end GPU).
I, among others, am hopeful that a CUDA LLR application will be developed (possibly based on the existing LL application, since the algorithms are similar) in the near future. However, this is by no means guaranteed since it depends on the willingness and ability of the developers to do so.
I am also hoping this will be possible in the near future. Even at speeds of 8 times faster than a single core CPU it will be a big boost for the LLR projects.
____________
 |
|
|
 warddrVolunteer tester
Send message
Joined: 7 Feb 08
Posts: 254
ID: 18735
Credit: 10,383,352
RAC: 0
        
|
|
Are the optimal-depths calculated before or after the recent llr speed improvements?
I assume with the faster llr, the sieves will reach their optimal-depths much faster.
____________
 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Are the optimal-depths calculated before or after the recent llr speed improvements?
I assume with the faster llr, the sieves will reach their optimal-depths much faster.
PSP (Sieve) was approaching the "optimal zone". With the LLR speed improvements, it definitely put the project well within the "zone".
____________
|
|
|
|
|
|
I'm curious if there are any updated estimates? Specifically I'm asking about 321 Prime Search (Sieve). Factors are becoming harder to find and frankly I'm wondering whether or not to "go for the gold". (which is still a long way off for me)
tks!
|
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1893
ID: 352
Credit: 3,141,488,980
RAC: 0
|
|
I would suggest to continue 321 Sieve if you want higher badge - this should be the first one to suspend in a few months.
(going for Ruby before it gets suspended).
____________
My stats
Badge score: 1*1 + 5*1 + 8*3 + 9*11 + 10*1 + 11*1 + 12*3 = 186 |
|
|
|
|
|
Well, that's my plan if 321 Sieve will indeed stay active for 3 months. I guess I was just concerned that it might be terminated sooner. |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 785
ID: 18447
Credit: 262,879,636
RAC: 0
         
|
|
I'm also trying to get 321 Sieve to pink before it stops - I think ruby is beyond me.
The irony of all of us trying to get the next badge up is that more work gets done, which brings forward the date where the sieve is optimal, and therefore the date of suspension moves ever earlier... :-)
____________
 |
|
|
|
|
|
More work in sieves dosen't mean that project will be closed earlier, it's only go deeper. At least that was in case of psp/sob sieve, where the date was set.
____________
Polish National Team |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 785
ID: 18447
Credit: 262,879,636
RAC: 0
|
|
Afaik, PSP/SoB Sieve end date did indeed move slightly earlier than was originally assumed.
I suppose it depends on if the decision is to pick a date to stop, or a depth - the latter makes more sense to me, and is movable based on how quickly that depth is reached. Being optimal doesn't care about the calendar (or badges for that matter).
____________
|
|
|
|
|
|
True. :)
I need about 52 days to reach a gold badge ( 321 sieve ).
____________
Polish National Team |
|
|
|
|
|
John,
How about an update on expectancies
especially 321 sieve and
C/W sieve since it has gone GPU?
____________
 |
|
|
|
|
|
321 sieve is only on cpu, right?
____________
Polish National Team |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
321 sieve is only on cpu, right?
Correct. And as far as I know that's not going to change in the foreseeable future.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
First post has been updated again.
4-6 weeks
321 Prime Search (Sieve) will be sieved to optimal depth.
____________
|
|
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
           
|
First post has been updated again.
4-6 weeks
321 Prime Search (Sieve) will be sieved to optimal depth.
John, is there an option of doubling this time frame? I try to reach my silver badge...
____________
Best wishes. Knowledge is power. by jjwhalen
 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
First post has been updated again.
4-6 weeks
321 Prime Search (Sieve) will be sieved to optimal depth.
John, is there an option of doubling this time frame? I try to reach my silver badge...
Back when the first post in this thread was established in Sept 2010, 321 (Sieve) was listed as 4-6 months. The list has been updated monthly since then. We are already towards the end of those 6 months and well within the "optimal sieve zone". An extra month time extension was added to allow those who were on the "cusp" of the next badge level to achieve their goal. It seems a bit of a stretch to extend it an additional month.
I noticed that you have several 64 bit machines running 32 bit Windows. I highly recommend Wubi which installs and uninstalls like a regular Windows application. It's free, easy to install, and would definitely help you in achieving your silver badge goal.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
First post has been updated again. 321 Prime Search (Sieve) will be suspended on 22 April 2011 - Earth Day!
____________
|
|
|
|
|
|
I have not done the math but I'm shooting for a 321sieve pink. I should be a shoe-in for gold though.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
|
7 days remain!!! The 321 Prime Search (Sieve) will be suspended on 22 April 2011 - Earth Day!
____________
|
|
|
STE\/EVolunteer tester
Send message
Joined: 10 Aug 05
Posts: 573
ID: 103
Credit: 3,625,446,192
RAC: 0
    
|
|
I'm still on track to get my Ruby Badge on the 18'th so I have room to spare yet ... :)
____________
 |
|
|
|
|
|
Good luck Steve.
got my silver! a few hours ago. that will be if for me...
needed 2 weeks to get 150k..
____________
|
|
|
|
|
|
John or Rytis,
Now it''s two down (PSP sieve,
and 321 sieve) and 11 indians to go.
Which is next sub-project suspension
and time frame.
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 785
ID: 18447
Credit: 262,879,636
RAC: 0
|
|
See the subproject life expectancy thread - in summary, gcwsieve and trpsieve within a year or so, and sgsllr at any moment if an sg prime is found, all others are much longer term.
____________
|
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
See the subproject life expectancy thread - in summary, gcwsieve and trpsieve within a year or so, and sgsllr at any moment if an sg prime is found, all others are much longer term.
This is true. I suspect that Cullen/Woodall (Sieve) will finish before The Riesel Problem (Sieve). And yes, probably "within a year or so."
However, Sophie Germain Prime Search (LLR) may be extended even if a twin or SG is found. The probabilities favor two SG's and one twin in our data set. If we found one form, we'd probably continue towards the other.
____________
|
|
|
|
|
|
Does anyone know if any of the other sieve projects are slated to have gpu enabled apps anytime soon on PG and boinc? Specifically the trp sieve. |
|
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
|
Does anyone know if any of the other sieve projects are slated to have gpu enabled apps anytime soon on PG and boinc? Specifically the trp sieve.
I talked with Ken_g6 in february over a gpu-app for 321sieve and tested some 321sieve-units with his TPsieve/PPsieve cuda-app.
Conclusion:
- 321sieve with TPsieve/PPsieve-cuda was much slower than sr2sieve on cpu.
- Ken has a vague outline of a way if you want to learn how to do discrete logarithms with Pollard's Rho method and write a new app.
____________
Best wishes. Knowledge is power. by jjwhalen
|
|
|
|
|
|
I have a doubt: in GPU-enabled sieves, will the CPU's become unable to crunch the WU's sooner than GPU's?
____________
 |
|
|
|
|
|
I have some questions.
When will the Proth prime search (sieve) subproject suspend?
Do you plan to make any other GPU subprojects with ppd like PPS (sieve)?
____________
  |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 785
ID: 18447
Credit: 262,879,636
RAC: 0
|
|
The PPS Sieve will transition to the next n-range of 3M-6M once the current range is optimal.
____________
|
|
|
|
|
The PPS Sieve will transition to the next n-range of 3M-6M once the current range is optimal.
Work for at least another 1 year? ;) |
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 785
ID: 18447
Credit: 262,879,636
RAC: 0
|
|
It all depends on how efficient those GPUs are versus the LLR on CPUs - we could always move to 6M-9M but that's only worthwhile if folks are starting to test that range within the forseeable future. The admins will have the final say as usual.
Of course, if llrCUDA is proven to work well with BOINC, we could alter the rules quite a lot in this respect...
____________
|
|
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
|
Of course, if llrCUDA is proven to work well with BOINC, we could alter the rules quite a lot in this respect... But this need a DP-capable card like GTX260-295 or the 400 and 500 series...
____________
Best wishes. Knowledge is power. by jjwhalen
|
|
|
|
|
But this need a DP-capable card like GTX260-295 or the 400 and 500 series...
Yes. And of course some more optimization up to a rollout at boinc for the masses. I'm running cudallr on prpnet. In my mind, the CPU-Core is still to much in use and die running time only is moderate. As Comparison: With the same card, in time of one PPS Sieve at Boinc I only can crunch 2 up to 3 PPSHigh on prpnet. I know, the comparision is flawed slightly, but if I compare the CPU times of these workunits, the llr on cpu is much more efficiently.
Regards Odi
____________
 |
|
|
|
|
The PPS Sieve will transition to the next n-range of 3M-6M once the current range is optimal.
And how I can find out, what n-range is computing at this moment? Is that on subproject statistics? If yes, where? |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
The PPS Sieve will transition to the next n-range of 3M-6M once the current range is optimal.
And how I can find out, what n-range is computing at this moment? Is that on subproject statistics? If yes, where?
You can see current p ranges in Subproject status in range status for PPS Sieve.
n<3M is finishing up (high p ranges) while 3M<n<6M is just beginning (low p ranges).
____________
|
|
|
|
|
The PPS Sieve will transition to the next n-range of 3M-6M once the current range is optimal.
And how I can find out, what n-range is computing at this moment? Is that on subproject statistics? If yes, where?
A look into WUs or the result files is the fastest way ;) I recognized the switch because the size of the upload files increased to over 25K per WU (ca. 500-600 factors found) from a few byte (ca. 0-5 factors per WU found at the end of the 0-3M phase) before.
____________
|
|
|
|
|
|
Is TRP Sieve near that magical "optimal depth"? It seems like it is for me, if I understand this correctly. I can run a TRP Sieve unit in about 3 hours, but each one returns only about 0.24 hits (composites). So it takes me an average of about 4 WUs to eliminate a candidate, or about 12 hours. I can run a TRP LLR task faster than that (or at least I could, as of a month or two ago).
My sieving benefits from being on a 64-bit machine; maybe we're not considered "optimal" until 32-bit machines are also break-even?
--Gary |
|
|
|
|
Is TRP Sieve near that magical "optimal depth"? It seems like it is for me, if I understand this correctly. I can run a TRP Sieve unit in about 3 hours, but each one returns only about 0.24 hits (composites). So it takes me an average of about 4 WUs to eliminate a candidate, or about 12 hours. I can run a TRP LLR task faster than that (or at least I could, as of a month or two ago).
My sieving benefits from being on a 64-bit machine; maybe we're not considered "optimal" until 32-bit machines are also break-even?
--Gary
64 bit machines have a higher optimal sieve depth compared with 32 bit machines.
____________
|
|
|
VatoVolunteer tester
Send message
Joined: 2 Feb 08
Posts: 785
ID: 18447
Credit: 262,879,636
RAC: 0
|
|
You also need to compare against the time to LLR a candidate with a higher n-value, e.g 75% of the range covered by the sieve, rather than just current leading-edge n-values. That moves the optimal point much further out. The reason for this is the factors found are saving tests over the entire range, and the time saved per factor found is also twice the sieve runtime equivalent because of the doublechecking of LLR WUs.
____________
|
|
|
|
|
|
Thanks Micro and Vato for the explanations... looks like I can press on with the sieve for a while then before the well runs dry.
--Gary |
|
|
|
|
|
Is it possible we can get an overall update to this list since it has been a few months? I'm generally curious as to when or whether any of the projects I do not yet have badges in will be shutting down.
Thanks!
____________
 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
Is it possible we can get an overall update to this list since it has been a few months?
First post has been updated. Only change is that Cullen/Woodall (Sieve) is confirmed a 1+ year project...because of the introduction of GPU sieving.
____________
|
|
|
|
|
Is it possible we can get an overall update to this list since it has been a few months?
First post has been updated. Only change is that Cullen/Woodall (Sieve) is confirmed a 1+ year project...because of the introduction of GPU sieving.
As that was one of the projects I was most wondering about (since I want to get it ruby soon) that was helpful!
Thanks,
Robert
____________
|
|
|
|
|
|
So when will work no longer be added to SGS?
____________
 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
            
|
So when will work no longer be added to SGS?
From the official announcement of the twin prime discovery:
The Sophie Germain Prime Search will continue to search the remainder of the sieved search space to hopefully find a Sophie Germain Prime.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
I don't understand.
Could some one explain why it should stop if a twin prime was found.
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
I don't understand.
Could some one explain why it should stop if a twin prime was found.
It's not going to stop, so no, nobody can explain why it should. ;-)
The reason people were confused (myself included) was that the "official" stance on this, from the first post in this thread said this:
Sophie Germain Prime Search (LLR) will be tested until a Twin and/or SG prime is found.
That seems to imply that the project would stop if either a Sophie Germain pair or a twin pair were found. That's the source of the confusion.
However, the same post also says this:
However, should one be found at this n, another search will be started at a different n so this project would still be around.
That unambiguously states that even if we find a Sophie Germain (let alone a twin), the SGS project would continue, but in a different search range.
All of that is moot, however, since it's been clearly and unambiguously stated in today's announcement that the SGS project will continue, at least until we exhaust the sieved range or until we find a Sophie Germain prime.
____________
My lucky number is 75898524288+1 |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
The reason people were confused (myself included) was that the "official" stance on this, from the first post in this thread said this:
Sophie Germain Prime Search (LLR) will be tested until a Twin and/or SG prime is found.
That seems to imply that the project would stop if either a Sophie Germain pair or a twin pair were found. That's the source of the confusion.
Yes, ambiguous...but the "and" carries just as much weight as the "or". ;)
However, should an SG be found now, the current search would definitely be stopped. This would only mean stopped at the current n. A new SGS search would resurface at a different n reaching for a higher goal. :)
____________
|
|
|
|
|
Yes, ambiguous...but the "and" carries just as much weight as the "or". ;)
However, should an SG be found now, the current search would definitely be stopped. This would only mean stopped at the current n. A new SGS search would resurface at a different n reaching for a higher goal. :)
Thanks. :) |
|
|
|
|
|
John, are we close to suspension of Cullen/Woodall and Riesel sieves? |
|
|
JohnHonorary cruncher
Send message
Joined: 21 Feb 06
Posts: 2875
ID: 2449
Credit: 2,681,934
RAC: 0
|
John, are we close to suspension of Cullen/Woodall and Riesel sieves?
The TRP sieve still has plenty of life in it. However, the CW sieve is approaching the end. The upcoming Leap Day Challenge will only shorten this time. An official shut down date will probably be announced after the Challenge and will be based on the work that was completed.
At most, there's only 6 months left. At the very least, a 4-6 week window will be given as advance notice before any "pausing of the sieve" is made.
____________
|
|
|
|
|
At most, there's only 6 months left. At the very least, a 4-6 week window will be given as advance notice before any "pausing of the sieve" is made.
yay! the sooner the better. Then I can move on to something else. ;-)
|
|
|
|
|
|
I need 34 days to reach 25kk. 5 weeks of 6, it will be on the edge. |
|
|
|
|
|
Any news on when CWsieve will be suspended?
I want the ruby badge ;)
Since there has been no ancouncement yet I assume there is still at least 4 weeks left?
____________
|
|
|
|
|
|
No news yet! As John suggested, there will be an announcement at least a month before the sub-project is suspended.
Cheers
- Iain
Any news on when CWsieve will be suspended?
I want the ruby badge ;)
Since there has been no ancouncement yet I assume there is still at least 4 weeks left?
____________
Twitter: IainBethune
Proud member of team "Aggie The Pew". Go Aggie!
3073428256125*2^1290000-1 is Prime! |
|
|
|
|
I need 34 days to reach 25kk. 5 weeks of 6, it will be on the edge.
I need 196 days for gold, just too much to make the try worth it. I will abandon it and leave the rest to others who can use the GCW credits better than I can.
Good luck in the hunt.
____________
Member team AUSTRALIA
My lucky number is 9291*2^1085585+1 |
|
|
|
|
|
Any progress in recalculation of CW sieve? |
|
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
|
Any progress in recalculation of CW sieve?
I hope they will be there for the next years. The CW sieve units run smooth and you get results and credits in a shortest time...
____________
Best wishes. Knowledge is power. by jjwhalen
|
|
|
|
|
|
Yes, they are nice and goes smooth, but according past information, it's going to an end. Question is, when. :)
My last checkpoint in this project is on 25 000 000. I'll reach it within 3 days. |
|
|
|
|
|
It is time to do SG ( sgs llr ) if you need a badge or a higher badge.
As it looks now we have about 2-4 weeks work left.
If we keep the speed we have now I expect it to be done before 1 May.
Good hunting :)
Lennart
|
|
|
|
|
It is time to do SG ( sgs llr ) if you need a badge or a higher badge.
As it looks now we have about 2-4 weeks work left.
If we keep the speed we have now I expect it to be done before 1 May.
Good hunting :)
Lennart
Lennart, why do we stop SG? I thought it's infinite. We still haven't found SG prime. If we stop the search, then when will we see the next range (1M)?
What about GCWSieve? I though it would be stopped first.
Cheers.
P.S. it's very sad!
|
|
|
|
|
|
And what happend with that:
The Sophie Germain Search is by itself since a Twin and/or SG could be
found at any minute. Since this is a quad sieve, there is a chance of finding both a
Twin and an SG. However, should one be found at this n, another search will be
started at a different n so this project would still be around.
You consider to stop it for a while or for couple of years? |
|
|
|
|
It is time to do SG ( sgs llr ) if you need a badge or a higher badge.
As it looks now we have about 2-4 weeks work left.
If we keep the speed we have now I expect it to be done before 1 May.
Good hunting :)
Lennart
Lennart, why do we stop SG? I thought it's infinite. We still haven't found SG prime. If we stop the search, then when will we see the next range (1M)?
What about GCWSieve? I though it would be stopped first.
Cheers.
P.S. it's very sad!
We stop it because we don't have more work sieved.
We started with N=666666 k1-40T we are at 37T on Boinc and we have soon done all from 38.5T in PRPNet.
Lennart
Read here.
http://www.primegrid.com/forum_thread.php?id=1450#17454 |
|
|
|
|
...
Lennart, why do we stop SG? I thought it's infinite. We still haven't found SG prime. If we stop the search, then when will we see the next range (1M)?
What about GCWSieve? I though it would be stopped first.
...
SG may be infinite, but the quadratic sieve at 200700-digits
was finite, and already produced a record twin prime pair.
I think I recall someone saying that if there's going to be
a new quad sieve, at something like 300000+ digits, that
it would be Lennart that would be doing the crunching.
That would get the SG and Twin candidate primes back on
the top5000 list, and perhaps improve user participation.
Besides, a record SG at 200700 is such a rare event, in
terms of winning lottery tickets, that one might as well try
for 300000+ and a record that might last 5 years, or a decade.
I keep attempting to remind PG users fond of searching for
SG's that the present record is under 80000-digits. As in,
no one knows any with between 80000 and 200000 digits.
Bumping that up to 300000 might actually improve the odds if
more PG users are willing to by a ticket ... err, that's accept running
a SG/LLR WU, I mean. -bdodson*
PS --- If I'm reading recent history correctly, the previous search,
before 200700-digits with 2^666666 was was in c. 100355-digits,
with 2^333333, which also produced a twin and no Sophie
(pending the next 2-4 weeks). That might suggest 2^999999
for a prospective 300000+ digit search (although I don't know
whether that's a PG preference or part of the quad sieve program). |
|
|
rroonnaallddVolunteer developer
Volunteer tester
Send message
Joined: 3 Jul 09
Posts: 1213
ID: 42893
Credit: 34,634,263
RAC: 0
|
We stop it because we don't have more work sieved.
We started with N=666666 k1-40T we are at 37T on Boinc and we have soon done all from 38.5T in PRPNet.
Lennart
Read here.
http://www.primegrid.com/forum_thread.php?id=1450#17454
Any chance for deeper sieving?
____________
Best wishes. Knowledge is power. by jjwhalen
|
|
|
|
|
It is time to do SG ( sgs llr ) if you need a badge or a higher badge.
As it looks now we have about 2-4 weeks work left.
If we keep the speed we have now I expect it to be done before 1 May.
Good hunting :)
Lennart
Lennart, why do we stop SG? I thought it's infinite. We still haven't found SG prime. If we stop the search, then when will we see the next range (1M)?
What about GCWSieve? I though it would be stopped first.
Cheers.
P.S. it's very sad!
We stop it because we don't have more work sieved.
We started with N=666666 k1-40T we are at 37T on Boinc and we have soon done all from 38.5T in PRPNet.
Lennart
Read here.
http://www.primegrid.com/forum_thread.php?id=1450#17454
So, is it being suspended until more sieve work is done? If so why hasn't new sieve work started already knowing this day was nearing? Or is it suspended indefinitely? And I thought that John said somewhere that we'd have 6 weeks notice before projects would be suspended? Please, more current info and not just a redirect to general info.
Boy do I miss John with his clear, concise and current info. Aside from Micheal Goetz' great updates and info, this project is sinking into an info abyss. We need more PROactive adnims. and less REactive. I thought we'd see more of Rytis with John gone, but he's still rarely seen. Rytis, please interact more. And I know Lennart has big shoes to fill, but with his English as a second language it's hard to get all the info we need clearly, and when we need it. Sorry Lennart, but sometimes it's hard to clearly understand what you're meaning (example: ...and we have soon done all from 38.5T in PRPNet.).
Something this big should be on the main page, or at least it's own thread.
NM*
____________
Largest Primes to Date:
As Double Checker: SR5 109208*5^1816285+1 Dgts-1,269,534
As Initial Finder: SR5 243944*5^1258576-1 Dgts-879,713
 |
|
|
|
|
We stop it because we don't have more work sieved.
We started with N=666666 k1-40T we are at 37T on Boinc and we have soon done all from 38.5T in PRPNet.
Lennart
Read here.
http://www.primegrid.com/forum_thread.php?id=1450#17454
Thank you, Lennart!
1. Could you also please explain about higher n's? (666667-666685). We haven't reached k=40 on those n's. Then why are we stopping only because we have reached k=40 at n=666666?
2. When will we start a new sieve for Sophie? For higher or current n?
3. What about GCWSieve? Any news? I see that it's hardly producing any new factors regardless of huge amount of work done.
Thanks and Cheers! |
|
|
|
|
1. Could you also please explain about higher n's? (666667-666685). We haven't reached k=40 on those n's. Then why are we stopping only because we have reached k=40 at n=666666?
There is only one n ;) Here is an example:
8 * 2^666666 + 1 = 4 * 2^666667 +1 = 2 * 2^666668 + 1 = 1 * 2^666669 + 1
So the limits are:
k=40T for n=666666
k=20T for n=666667
k=10T for n=666668
.
.
.
k=77M for n=666685
____________
|
|
|
|
|
We stop it because we don't have more work sieved.
We started with N=666666 k1-40T we are at 37T on Boinc and we have soon done all from 38.5T in PRPNet.
Lennart
Read here.
http://www.primegrid.com/forum_thread.php?id=1450#17454
So, is it being suspended until more sieve work is done? If so why hasn't new sieve work started already knowing this day was nearing? Or is it suspended indefinitely? And I thought that John said somewhere that we'd have 6 weeks notice before projects would be suspended? Please, more current info and not just a redirect to general info.
Boy do I miss John with his clear, concise and current info. Aside from Micheal Goetz' great updates and info, this project is sinking into an info abyss. We need more PROactive adnims. and less REactive. I thought we'd see more of Rytis with John gone, but he's still rarely seen. Rytis, please interact more. And I know Lennart has big shoes to fill, but with his English as a second language it's hard to get all the info we need clearly, and when we need it. Sorry Lennart, but sometimes it's hard to clearly understand what you're meaning (example: ...and we have soon done all from 38.5T in PRPNet.).
Something this big should be on the main page, or at least it's own thread.
NM*
It's exacly what I wanted to write today. |
|
|
|
|
|
Thanks! Didn't think about it )) |
|
|
Scott BrownVolunteer moderator
Project administrator
Volunteer tester
Project scientist
Send message
Joined: 17 Oct 05
Posts: 2165
ID: 1178
Credit: 8,777,295,508
RAC: 0
|
Sorry Lennart, but sometimes it's hard to clearly understand what you're meaning (example: ...and we have soon done all from 38.5T in PRPNet.).
Something this big should be on the main page, or at least it's own thread.
NM*
I believe that what Lennart said was that PG is soon to complete SGS currently sieved work up to 40T. This is divided between BOINC and PRPnet such that the BOINC side is completing up to 38.5T anr PRPnet will soon be done from 38.5T upward (i.e., from 38.5T to 40T).
Also, let's all give Lennart a bit of a break. Several people are working on filling pieces of the job that John was doing, and from conversations with some of them as well as beginning to fill a small part of his work role myself, it is obvious that John was a visiting alien with the ability to warp both time and space! ;)
Joking aside, these roles/tasks previously performed by John that others are learning take time to master. And as completely volunteer work, mastering things will take a bit longer than for paid work since the latter often takes up time itself. Personally, given the huge amount of work that John performed here at PG, I am amazed that things have continued on as well as they have...which is a testament to how well Lennart has done (as well as to how much impact the new volunteers, such as Mike and Iain, have done as well).
____________
141941*2^4299438-1 is prime!
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
|
First off, let me say that the following comments are mine alone, and do not necessarily represent the views of anyone else at PrimeGrid -- although I suspect that they will agree with what I'm going to say. :)
One thing everyone needs to remember is that PrimeGrid is entirely a volunteer organization, and that the various volunteers speak different languages and live in different time zones.
Plus there's several volunteers still climbing up the learning curve. And at least one role that I just discovered has fallen through the cracks and nobody is doing yet. Bonus points if you can figure out what the unfilled role is. :)
Therefore, coordination and decisions will sometimes take a bit longer than I (or anyone) would like.
The future of SGS is one of those things that's going to take a little while to figure out. It would have been nice to have already worked our way through that decision, but we've been busy with several time consuming tasks, some of them completely unexpected. (A LOT of behind the scenes scrambling has been going on to deal with the LLR bug problem, for example.)
Should more notice of the impending work stoppage have been given? Absolutely. That was a mistake on our part, and was not intentional. We'll try to do better.
What I can say is that a discussion has been started about the future of SGS. I know people want to continue to hunt for a new record SGS pair. I know people want to get the next SGS badge.
Speaking of badges: here's a suggestion to help those who are close to getting the next SGS badge: If you're not going to get a new SGS badge before we run out of work (or don't care about the badge), stop crunching SGS so that more WUs are available for those who are close to getting the badge.
One more thing to remember: we just finished the SGS challenge. It's entirely possible that a lot more WUs were crunched than expected. The impending end of SGS may have been just as much of a surprise to Lennart as the rest of us!
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
And like the Germans say: "Alles hat ein Ende, nur ein Wurst hat zwei!"
(Literal translation: Everything has an end, only a sausage has two.)
Hope I can make it to the next badge. It's gonna be a close call!
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
 |
|
|
|
|
|
Lennart,
Thank you for all the work and time you put into this project! It can be a rewarding experience but can really take its toll. Be sure to give yourself justly deserved breaks and rest to help prevent burn out. I do not want to think of PG without you!
--
I am one of the *two* admins for DistrRTgen (freerainbowtables). When I joke that the project is my 2nd job, it literally takes all the time I can give it outside of my day job. Having to balance running a server, boinc, the project, responding to users, keeping up communication, finding time to improve or fix code, etc. means a lot of work for two people to somehow manage.
To others that are pecking at Lennart as if he had abundant free time that he was just wasting instead of answering every post with all details desired, shame on you! The time demands from such work are extreme and burn out is always something very close for many trying to juggle all the roles as best they can. Yes, abundant open communication from project staff and crunchers is vital *but* please realize time spent on a task such as message board responses is time taken away from other areas. |
|
|
|
|
|
My suspicion of the missing role is the creation and posting of official announcements (as I have had a prime that this should have happened), plus keeping up the home page with the significant primes. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
My suspicion of the missing role is the creation and posting of official announcements (as I have had a prime that this should have happened), plus keeping up the home page with the significant primes.
BZZZZZZT!!
Nope, that one's actually recently filled, and you should see things starting to pick up in that area once he gets up to speed.
____________
My lucky number is 75898524288+1 |
|
|
|
|
What I can say is that a discussion has been started about the future of SGS. I know people want to continue to hunt for a new record SGS pair. I know people want to get the next SGS badge.
I think that the hunt for any SGS project prime is what interests people. Perhaps the admins do not realise that getting onto the records table is of no interest to many. Consider that there has been little loss of interest now that the primes are out of the top 5000. No one cares about it!!
Speaking of badges: here's a suggestion to help those who are close to getting the next SGS badge: If you're not going to get a new SGS badge before we run out of work (or don't care about the badge), stop crunching SGS so that more WUs are available for those who are close to getting the badge.
Good idea, I have done that for GCW SV and I am 8.3 days away from my silver in SGS.
Good Luck all fellow badge chasers!
____________
Member team AUSTRALIA
My lucky number is 9291*2^1085585+1 |
|
|
|
|
|
The key question is: when sgs will be available again. |
|
|
|
|
|
My post Message boards : News : World Record Twin Primes Found! (Message 45708) http://www.primegrid.com/forum_thread.php?id=3873&nowrap=true#45708 mentioned this problem. Several people offered to do sieving if required in subsequent posts in this thread.
I am just reminding Admins of the offers in case they can be of use now.
As to when will SGS return, I hope it is ASAP.
____________
Member team AUSTRALIA
My lucky number is 9291*2^1085585+1 |
|
|
|
|
|
What I said about Lennart was only meant to be constructive criticism, hoping a little more thought would go into what he says in English . It was in no way meant as an insult, and I hope it wasn't taken that way. I know what he means 90% of the time but it does lack details often that would make it clearly understandable 100% of the time. I know he is doing a lot ATM and should be commended for that.
So thank you, Lennart, for all that you do.
As far as a role that's missing/needed is and Information Officer/PR kind of person, who knows what going on in the back ground by other admins. and volunteers. A GO TO GUY. This would be his/her only job, reporting info frequently in the forums and making sure the main page is always up to date. Someone who can direct hard or unanswered question to the appropriate person(s). And someone to make announcements too, when needed. I know Mike and Lennart has been the most visible at this at the moment, but we need a PR like person, but not one who glosses over things but gives it to us straight, raw and detailed, at least for this kind of position. Mike would be good at it but he's already doing more than than could be asked of him, answering all kinds of questions outside of GFN rather than him focusing on GFN alone, which is kind of his baby ATM.
Is this what you had in mind Mike? Or is it you don't mind spreading your wings more? Would you mind this kind of role?
NM*
____________
Largest Primes to Date:
As Double Checker: SR5 109208*5^1816285+1 Dgts-1,269,534
As Initial Finder: SR5 243944*5^1258576-1 Dgts-879,713
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
As far as a role that's missing/needed is and Information Officer/PR kind of person, who knows what going on in the back ground by other admins. and volunteers. A GO TO GUY. This would be his/her only job, reporting info frequently in the forums and making sure the main page is always up to date. Someone who can direct hard or unanswered question to the appropriate person(s). And someone to make announcements too, when needed. I know Mike and Lennart has been the most visible at this at the moment, but we need a PR like person, but not one who glosses over things but gives it to us straight, raw and detailed, at least for this kind of position. Mike would be good at it but he's already doing more than than could be asked of him, answering all kinds of questions outside of GFN rather than him focusing on GFN alone, which is kind of his baby ATM.
Is this what you had in mind Mike? Or is it you don't mind spreading your wings more? Would you mind this kind of role?
Hmmm. Half credit for that one. We DO need someone to fill that role. I call it "Head Cheerleader", but also includes the "Go-To Guy" and "Chief Public Relations" functions. I'm not too worried about the go-to guy function, however. Even if I don't say anything, if I see a problem that needs Rytis's or Lennart's attention, I make sure they know about it. That's probably not necessary as I suspect they read it all anyway.
As for me being that person -- I think we need someone more versed in all the projects. Plus, I'd rather be spending my time working on Genefer.
But that's not the role I was thinking of. :)
____________
My lucky number is 75898524288+1 |
|
|
|
|
Thanks! Didn't think about it ))
http://www.primegrid.com/stats_sgs_llr.php
provides some additional info.
____________
|
|
|
|
|
|
Oh I'm going to be so close to my Bronze badge on Sophie looks like I'm short 300 pts right now. I'm NNT all other projects and both cores will be on Sophie and I will keep my fingers crossed. |
|
|
|
|
|
Lennart, please update us how many WUs left till full stop of SGS in BOINC and how long will GCWSieve last? |
|
|
|
|
Lennart, please update us how many WUs left till full stop of SGS in BOINC and how long will GCWSieve last?
222,095 total + Buffer ~2 000
Lennart |
|
|
|
|
|
So SGS in BOINC is going to be over in less than a week I guess?
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
|
|
So SGS in BOINC is going to be over in less than a week I guess?
113,250 total
It seems like that.
Lennart
|
|
|
|
|
|
64,430 total
Lennart |
|
|
|
|
64,430 total
Lennart
dropping like a rock and based on the posted results less than a day left. |
|
|
|
|
Lennart, please update us how many WUs left till full stop of SGS in BOINC and how long will GCWSieve last?
222,095 total + Buffer ~2 000
Lennart
Thank you! But you haven't answered about GCWSieve. Please update us.
Thanks. Cheers!
|
|
|
|
|
Lennart, please update us how many WUs left till full stop of SGS in BOINC and how long will GCWSieve last?
222,095 total + Buffer ~2 000
Lennart
Thank you! But you haven't answered about GCWSieve. Please update us.
Thanks. Cheers!
I have todo some times/factor first.
I shall try fix it to the weekend.
Lennart |
|
|
|
|
|
No available WU's for SGS. I guess the project is over in BOINC. Very sad.
Impatiently waiting for its restart. |
|
|
|
|
No available WU's for SGS. I guess the project is over in BOINC. Very sad.
Impatiently waiting for its restart.
Very sad indeed..came within less than 10k credit of the final badge. :-(
____________
 |
|
|
RytisVolunteer moderator
Project administrator
Send message
Joined: 22 Jun 05
Posts: 2649
ID: 1
Credit: 26,363,112
RAC: 0
     
|
|
There still be resends, so you may just get the tasks you need.
____________
|
|
|
|
|
Thank you! But you haven't answered about GCWSieve. Please update us.
Thanks. Cheers!
I have todo some times/factor first.
I shall try fix it to the weekend.
Lennart
any news? |
|
|
|
|
|
All I know is I have just started the push to get my CWSieve badge up to Ruby. Should take about a month of running CUDA only at night (got to love that nice cool night air keeping the computer nice and cool).
____________
|
|
|
|
|
|
Any news about GCWSieve?
____________
 |
|
|
|
|
Any news about GCWSieve?
There is about 4 weeks left on GCW Sieve. !!
Lennart |
|
|
|
|
|
Eep! Time to get my computer going on those 24/7 in hopes of making it to Ruby.
____________
|
|
|
|
|
Any news about GCWSieve?
There is about 4 weeks left on GCW Sieve. !!
Lennart
Are we down to 2 weeks now? Will there be an official announcement? |
|
|
|
|
Any news about GCWSieve?
There is about 4 weeks left on GCW Sieve. !!
Lennart
Are we down to 2 weeks now? Will there be an official announcement?
I will tell you when it is 2 weeks left.
Lennart
|
|
|
|
|
|
Looks like 4 weeks estimate was way too optimistic.
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
|
I added GFN life expectancies to the first post.
____________
My lucky number is 75898524288+1 |
|
|
|
|
Any news about GCWSieve?
There is about 4 weeks left on GCW Sieve. !!
Lennart
So what's happened to this estimate? We're about 6 weeks on and we haven't seen the 2 week update yet...
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
1+ years
Proth Prime Search (Sieve) will be sieved to optimal depth. (will transition to 3M-6M sieve)
Hm, didn't this happen already? (Edit: The transition to 3M-6M, I mean.)
____________
|
|
|
|
|
|
Might also want to update "Sophie Germain Prime Search (LLR)" since it is currently suspended and Sieving is being done for the next set. |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
Might also want to update "Sophie Germain Prime Search (LLR)" since it is currently suspended and Sieving is being done for the next set.
I didn't update that because I'm not 100% sure of its exact status right now.
____________
My lucky number is 75898524288+1 |
|
|
|
|
Might also want to update "Sophie Germain Prime Search (LLR)" since it is currently suspended and Sieving is being done for the next set.
I didn't update that because I'm not 100% sure of its exact status right now.
"Once an SG is found" should be changed to reflect it has been found and that the project is on hiatus/suspended currently. More info to come at a later date.
Since the Forum section itself has not been place into retirement at least people can see that the last set finished. |
|
|
ardo
Send message
Joined: 12 Dec 10
Posts: 168
ID: 76659
Credit: 1,690,471,713
RAC: 0
          
|
Any news about GCWSieve?
There is about 4 weeks left on GCW Sieve. !!
Lennart
So what's happened to this estimate? We're about 6 weeks on and we haven't seen the 2 week update yet...
And now we're two more weeks further.
Can one of the project scientists give any information here? Even a rough guestimate would do.
____________
Badge score: 2*5 + 8*7 + 3*8 + 3*9 + 1*10 + 1*11 + 1*13 = 151
 |
|
|
|
|
|
I got a few SGS LLR - reactivation?
____________
 |
|
|
|
|
I got a few SGS LLR - reactivation?
No news on the front page, but it seems sgs is back with a new range:
http://www.primegrid.com/stats_sgs_llr.php
____________
676754^262144+1 is prime |
|
|
|
|
|
I am testing SGS now.
Only a few wu loaded.
Lennart
|
|
|
|
|
Any news about GCWSieve?
There is about 4 weeks left on GCW Sieve. !!
Lennart
So what's happened to this estimate? We're about 6 weeks on and we haven't seen the 2 week update yet...
And now we're two more weeks further.
Can one of the project scientists give any information here? Even a rough guestimate would do.
any updates now that we're 10 weeks into the 4 week estimate? |
|
|
|
|
Any news about GCWSieve?
There is about 4 weeks left on GCW Sieve. !!
Lennart
Are we down to 2 weeks now? Will there be an official announcement?
I will tell you when it is 2 weeks left.
Lennart
GCW Sieve will be loaded last time September 1.
If you need credit or badges it is time to put your core on GCW sieve now ! :)
Lennart |
|
|
|
|
GCW Sieve will be loaded last time September 1.
Lennart
That's kind of a funny date to pick or maybe there's some meaning to it but why not say August 31. Just wondering.. no big deal just kinda struck me as odd :)
Also, when you say loaded, about how many wu's will that be?
____________
@AggieThePew
 |
|
|
|
|
GCW Sieve will be loaded last time September 1.
Lennart
That's kind of a funny date to pick or maybe there's some meaning to it but why not say August 31. Just wondering.. no big deal just kinda struck me as odd :)
Also, when you say loaded, about how many wu's will that be?
Will only last for some hr :)
Lennart |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
That's kind of a funny date to pick or maybe there's some meaning to it but why not say August 31. Just wondering.. no big deal just kinda struck me as odd :)
There's absolutely no significance to that date other than that we thought we should give you about a month's notice with a definitive date.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Can someone please explain the difference between the startpost and Lennarts post on GCWSieve life time expectancy?
Startpost says this:
Cullen/Woodall (Sieve) has been sieved to optimal depth and will be closed on September 1st, 2012.
Where as Lennarts post says:
GCW Sieve will be loaded last time September 1.
The first sounds like: from September 1st onwards no new work for GCWSieve can be downloaded anymore.
The second sounds like: On September 1st we will put (more) work in the queue for the last time and when that's out, the project is over.
Now I know the latter does not necessarily exclude the former, but as I have no idea how much work gets done in a day or how much work would/could potentially be loaded on September 1st. None the less, I'm curious as to what is what :)
As I have my ruby badge I'm going to leave the points alone for whomever needs them.
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
|
All of the above are correct.
Prior to September 1st, the GCW queue will be kept filled with work. On September 1st, the last batch of work will be loaded for GCW sieve. That should only last a few hours.
Bottom line: GCW will shut down on, or very shortly after, September 1st.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Ah great. Thanks for clearing that up :)
____________
PrimeGrid Challenge Overall standings --- Last update: From Pi to Paddy (2016)
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
|
I updated the first post by moving the TRP-Sieve to the 10+ year category based on calculations of where the optimal sieve depth occurs.
____________
My lucky number is 75898524288+1 |
|
|
|
|
GCW Sieve will be loaded last time September 1.
If you need credit or badges it is time to put your core on GCW sieve now ! :)
Lennart
Any substitute sieve project when this is shutdown?
____________
92*10^1439761-1 REPDIGIT PRIME :) :) :)
314187728^131072+1 GENERALIZED FERMAT
31*332^367560+1 CRUS PRIME
Proud member of team Aggie The Pew. Go Aggie! |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
Any substitute sieve project when this is shutdown?
Not at this time.
____________
My lucky number is 75898524288+1 |
|
|
Ken_g6Volunteer developer
Send message
Joined: 4 Jul 06
Posts: 915
ID: 3110
Credit: 183,164,814
RAC: 0
|
Any substitute sieve project when this is shutdown?
If by "substitute" you mean "what sieve project should I run when GCW Sieve is shut down", Proth Prime Search sieve uses a very similar codebase on GPUs. And there's still TRP Sieve which is CPU-only.
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
Any substitute sieve project when this is shutdown?
If by "substitute" you mean "what sieve project should I run when GCW Sieve is shut down", Proth Prime Search sieve uses a very similar codebase on GPUs. And there's still TRP Sieve which is CPU-only.
There's also a CUDA GFN sieve available for manual sieving. For CPU sieving, there are manual Factorial and PPS/RSP sieves. BOINC credits are awarded for the manual sieves.
The easiest sieve to switch to is the PPS sieve on BOINC.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Is the work that is currently available the last work that will ever be available for CW Sieve?
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
Is the work that is currently available the last work that will ever be available for CW Sieve?
Yes, that is correct, but there are two caveats:
1) Lennart is supposed to load work today for the last time. I do not know if that's been done yet.
2) The sieve isn't finished; it's suspended. It's possible it may be resumed sometime in the future if conditions warrant.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
GCW sieve work load script is stoped !!!!!!!
Should be empty in 1 hr.
Lennart |
|
|
|
|
|
Thanks for the fun, GCW Sieve. I've moved on; the remaining units are for those in search of a badge. Cheers!
--Gary |
|
|
|
|
|
My CPU managed to pick one up, I will go ahead and let it finish.
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
|
With respect to GCW, or any subproject that is shutting down, it should be noted that there will continue to be small amounts of tasks available for a short time. Tasks that are in progress that eventually fail (either because they return an error or when they exceed the deadline) will be resent out until a valid result is returned. Since sieves usually have a quorum of 1 and don't require verification, this is less likely than with LLR or GFN. Although there won't be wingman verification errors with a sieve, there will still be computation errors and timeouts.
Therefore, if you're just shy of a badge or some other milestone you may wish to continue to try to get more tasks, even if the queue seems to be empty. There may be a few more that trickle in over time.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
So, now that 3 sub-projects have been suspended/closed (since I srarted in 2010),
what is the NEXT likely sub-project to be suspended/closed and time interval to such.
Thanx.
Mike
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
So, now that 3 sub-projects have been suspended/closed (since I srarted in 2010),
what is the NEXT likely sub-project to be suspended/closed and time interval to such.
Thanx.
Mike
All of the primality tests (LLR & Genefer) will be running for quite a few years, unless we get crazy lucky with the conjecture projects (SoB, PSP, & TRP). The other primality testing projects (SGS, PPS, etc.) we can always come up with more numbers to test if we exhaust the range we're testing now (similar to SGS's hiatus earlier this year.) The short GFN tasks (n=19) will probably be exhausted around the end of this year, but I suspect we'll start crunching n=20 instead.
The two remaining sieves (PPS/TRP) I'm not certain about, but since there's no real end in sight for any other project, it's likely to be one of these two next. "Next" might be years from now, however.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
I know that some will miss badges and be a little upset but the sieves ending is really an achievment for us.
We have successfully completed one part of the job.
Well done all.
____________
Member team AUSTRALIA
My lucky number is 9291*2^1085585+1 |
|
|
|
|
|
Any possibility of new sub-projects being added?
____________
 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 13513
ID: 53948
Credit: 236,960,660
RAC: 0
|
Any possibility of new sub-projects being added?
Did you have anything particular in mind?
____________
My lucky number is 75898524288+1 |
|
|
HonzaVolunteer moderator
Volunteer tester
Project scientist
Send message
Joined: 15 Aug 05
Posts: 1893
ID: 352
Credit: 3,141,488,980
RAC: 0
|
