Newer milestone thread - Page 291

kriesel · 2019-10-08, 06:50

Quote:

Originally Posted by petrw1

As I read it 1st and 3rd are current. The middle one is 17 days late.

Update dates are changing, yes. But % completed?

rudy235 · 2019-10-08, 16:41

Quote:

Originally Posted by kriesel

Update dates are changing, yes. But % completed?

That's what i meant by stuck.

There are 3 assignments

Code:

85526383	LL	LL, 10.80%	2	8	2019-07-12	2019-10-08	2019-10-09	2019-10-16	c0nd0r
85767821	LL	LL, 93.30%	0	-17	2019-09-08	2019-09-20	2019-09-21	2019-09-21	OscarX
85925303	LL	LL, 63.70%	35	11	2019-08-14	2019-10-08	2019-10-09	2019-10-19	Michael Quevillon

kriesel · 2019-10-09, 17:03

Quote:

Originally Posted by rudy235

That's what i meant by stuck.

There are 3 assignments

Code:

85526383    LL    LL, 10.80%    2    8    2019-07-12    2019-10-08    2019-10-09    2019-10-16    c0nd0r
85767821    LL    LL, 93.30%    0    -17    2019-09-08    2019-09-20    2019-09-21    2019-09-21    OscarX
85925303    LL    LL, 63.70%    35    11    2019-08-14    2019-10-08    2019-10-09    2019-10-19    Michael Quevillon

One expired and Jon Pace is now on the case. Unfortunate that it was so close to done before stalling out. Maybe someday it will complete as a double check.
C0nd0r's expires in a day.
Quevillon's has quite a while to go yet, and will probably determine the milestone timing.

ATH · 2019-10-14, 23:06

Anonymous results overtook CurtisC within the last 30 days for the #1 Lifetime Ghz-days all categories combined and "TheJudger" is "pretty close" behind CurtisC in 3rd place:
https://www.mersenne.org/report_top_...1-01&end_date=

CurtisC has 97.9% in LL first tests while TheJudger has 93.4% in TF :-) Anonymous divided between LL, LL-D and TF.

In Lifetime "LL first tests" CurtisC still reigns supreme with ~2.5 times as much as all the Anonymous results combined and over 10 times as much as #3:
https://www.mersenne.org/report_top_...1-01&end_date=

kriesel · 2019-10-20, 13:31

Expected completion a week away, Oct 27
https://www.mersenne.org/assignments...xtf=1&exdchk=1

Current projection is 87M milestone in early December. That's without the possible delaying effect of expire/reassign/complete the overdue assignments.

dcheuk · 2019-11-02, 14:29

Quote:

Originally Posted by kriesel

Expected completion a week away, Oct 27
https://www.mersenne.org/assignments...xtf=1&exdchk=1

Current projection is 87M milestone in early December. That's without the possible delaying effect of expire/reassign/complete the overdue assignments.

Looks like it's stuck at 58.50%.

rudy235 · 2019-11-02, 15:05

Quote:

Originally Posted by dcheuk

Looks like it's stuck at 58.50%.

YEP...Very much so...and on the second verification to 49 Million, 4 of the 5 exponents are stuck. Only this one here is advancing!

kriesel · 2019-11-02, 17:34

Quote:

Originally Posted by rudy235

YEP...Very much so...and on the second verification to 49 Million, 4 of the 5 exponents are stuck. Only this one here is advancing!

Four of the five are reporting in recently, with indicated ETAs of 2 days or less. The other expires within a week. At worst case, all expire within 30 days, so it's likely to complete in December.

One way exponents can get stalled is if the number of prime95 workers on a system gets reduced. Some exponents in progress will get queued behind others from that, and won't see any more compute cycles until the exponents in front of them in the prime95 worktodo queue per worker get cleared, or the user manually intervenes.

In the run to 87M first test milestone, https://www.mersenne.org/assignments...xtf=1&exdchk=1 there are 134 left. An interesting one is https://www.mersenne.org/report_expo...6954009&full=1 which is sitting at 100.0% complete PRP but not submitted by rx yet; 12 days overdue, 56 days to expiration, hasn't reported in since Oct 21. It's looking iffy for December completion of this milestone, with some projected to take to mid December to complete, and others overdue and not reporting in, and requiring 8 weeks or more to expire.

kriesel · 2019-11-03, 14:56

A recent update shows the last exponent to 86M now at 61.9%
https://www.mersenne.org/assignments...xtf=1&exdchk=1
But it, and Yoon's other exponent on node tulip, are projected to expire before their ETAs.
https://www.mersenne.org/M86838679

retina · 2019-11-04, 14:48

I notice the count for the OneLL range 40,250,000 - 50,000,000 on the classic report is off by one. It always shows one extra.

rudy235 · 2019-11-04, 23:48

Quote:

Originally Posted by retina

I notice the count for the OneLL range 40,250,000 - 50,000,000 on the classic report is off by one. It always shows one extra.

Code:

40,250,000 	50,000,000 	553,232 	2 	357,490 	193,438 	2,302 	0 	0.000 	0 	1.640 	2560

Code:

Countdown to verifying all tests below 50 million: 2 301 (1 960 available)

True, but there is still a lot of time to get this fixed.

2019-10-14, 23:06	#3194
ATH Einyen Dec 2003 Denmark 2·1,579 Posts	Anonymous results overtook CurtisC within the last 30 days for the #1 Lifetime Ghz-days all categories combined and "TheJudger" is "pretty close" behind CurtisC in 3rd place: https://www.mersenne.org/report_top_...1-01&end_date= CurtisC has 97.9% in LL first tests while TheJudger has 93.4% in TF :-) Anonymous divided between LL, LL-D and TF. In Lifetime "LL first tests" CurtisC still reigns supreme with ~2.5 times as much as all the Anonymous results combined and over 10 times as much as #3: https://www.mersenne.org/report_top_...1-01&end_date= Last fiddled with by ATH on 2019-10-14 at 23:07

2019-10-20, 13:31	#3195
kriesel "TF79LL86GIMPS96gpu17" Mar 2017 US midwest 2²×5×271 Posts	1 to go to 86M first primality test milestone Expected completion a week away, Oct 27 https://www.mersenne.org/assignments...xtf=1&exdchk=1 Current projection is 87M milestone in early December. That's without the possible delaying effect of expire/reassign/complete the overdue assignments. Last fiddled with by kriesel on 2019-10-20 at 13:35

2019-11-03, 14:56	#3199
kriesel "TF79LL86GIMPS96gpu17" Mar 2017 US midwest 2²×5×271 Posts	Progress! A recent update shows the last exponent to 86M now at 61.9% https://www.mersenne.org/assignments...xtf=1&exdchk=1 But it, and Yoon's other exponent on node tulip, are projected to expire before their ETAs. https://www.mersenne.org/M86838679 Last fiddled with by kriesel on 2019-11-03 at 15:00

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Newer X64 build needed	Googulator	Msieve	73	2020-08-30 07:47
Performance of cuda-ecm on newer hardware?	fivemack	GMP-ECM	14	2015-02-12 20:10
Cause this don't belong in the milestone thread	bcp19	Data	30	2012-09-08 15:09
Newer msieves are slow on Core i7	mklasson	Msieve	9	2009-02-18 12:58
Use of large memory pages possible with newer linux kernels	Dresdenboy	Software	3	2003-12-08 14:47

2019-11-04, 14:48	#3200
retina Undefined "The unspeakable one" Jun 2006 My evil lair 6224₁₀ Posts	I notice the count for the OneLL range 40,250,000 - 50,000,000 on the classic report is off by one. It always shows one extra.