80 million goal achieved!
Nice to tick off those x0 increments.

[QUOTE=rudy235;493936]Would you venture a guess as to when we could reach the limit of 79.3 million for double checked? 10 Years?[/QUOTE]

I'm going to draw a terrible comparison here, so brace yourself. :smile:

It took the first-time checks 5 years to go from 45M to 79.3M so maybe the DC's will take the same amount of time.

The reason I say it's a terrible comparison is because not as many people are doing DC work, but then again as time goes on and throughput improves because of new CPU/GPU designs, maybe it would stick to that pace.

Another way to look at it is that the DC "million" milestones are being crossed every ~4 months, based on a very small sampling of recent dates. Again assuming increases in CPUs that may help keep that pace, we could expect to hit 79.3M for double-checks in (34 * 4) months = 11.3 years. :smile:

[QUOTE=Madpoo;494044]I'm going to draw a terrible comparison here, so brace yourself. :smile:

It took the first-time checks 5 years to go from 45M to 79.3M so maybe the DC's will take the same amount of time.

The reason I say it's a terrible comparison is because not as many people are doing DC work, but then again as time goes on and throughput improves because of new CPU/GPU designs, maybe it would stick to that pace.

Another way to look at it is that the DC "million" milestones are being crossed every ~4 months, based on a very small sampling of recent dates. Again assuming increases in CPUs that may help keep that pace, we could expect to hit 79.3M for double-checks in (34 * 4) months = 11.3 years. :smile:[/QUOTE]

Yeah i think this last estimate may be a tad optimistic.. may some apathy among the people that volunteer with the double checking will be somewhat compensated by better processors and even by better programming. I might not be able to see this milestone accomplished, but at the very least I with to see it get to 60 million.

Roughly 62K double checks was done in the last year, so that would be 9-10 years, IF the increasing speed of the hardware keeps up with the increasing LL runtime from 45M to 79M.

[QUOTE=ATH;494067]Roughly 62K double checks was done in the last year, so that would be 9-10 years, IF the increasing speed of the hardware keeps up with the increasing LL runtime from 45M to 79M.[/QUOTE]

We might be doing better actually....

Correct my if I'm wrong but I seem to recall that a few years back George changed the rules that all new PCs must successfully complete at least 2 DC's per core before doing LL....and/or the default work type for new PCs was also changed from LL to DC.

[QUOTE=petrw1;494113]We might be doing better actually....

Correct my if I'm wrong but I seem to recall that a few years back George changed the rules that all new PCs must successfully complete at least 2 DC's per core before doing LL....and/or the default work type for new PCs was also changed from LL to DC.[/QUOTE]

The default is to do 1 DC per year. New clients haven't done 1 DC in the last year, so they get assigned a DC to start by default.

September 09, 2018 All tests below 45 million verified. :smile:

It looks like we will have cleared up to 81M by the end of October. This will be a fine year of milestones. The "new" assignment rules and class system seems to be helping out. Poaching has all but disappeared.
:thread:

16,000 to go under 1Million
 

That is, not yet factored. Just a little over 20% left.

Milestones this year:
2018-01 None
2018-02-22 All tests below M(42 643 801) are verified, officially making it the 46th Mersenne prime.
2018-02-27 All exponents below 77 million tested at least once.
2018-03-02 All exponents below M(77 232 917) tested at least once. (All first times are now WR).
2018-04-08 All tests below 43 million verified.
2018-04-08 All tests below M(43 112 609) are verified, officially making it the 47th Mersenne prime.
2018-04-18 All exponents below 78 million tested at least once.
2018-05 None
2018-06-13 All tests below 44 million verified.
2018-06-20 All exponents below 79 million tested at least once.
2018-07 None
2018-08-14 All exponents below 79.3 million tested at least once. This was the original upper limit for early versions of Prime95.
2018-08-16 All exponents below 80 million tested at least once.
2018-09-09 All tests below 45 million verified.
2018-10-21 All exponents below 81 million tested at least once.
2018-11 ?
2018-12 ?

Looking at projected completion dates on PrimeNet, all of the DC's [below the next milestone] -should be- done by the end of Nov. and all of the first time [below the next milestone] by the end of Dec.

If that happens it would be most favourable.