Countdown to < 40M DCs
 

We're almost there... only 2 left.

Unfortunately I have my doubts about the machine running M39890261

It reports only 7 days left (and 12 days left to expiration), however this machine is very sporadic.

On a good day it'll move the needle nearly 4.5% daily. However it has inexplicable gaps where it goes several days at a time with nearly zero progress.

It recently went 12 days between check-ins during which it only progressed 1.6% (between May 4 and May 16).

Then it went another 5 days silent from May 18 to May 23 and only progressed 1.5%

Anyway, just wanted to say it could be a close one, and we'll just hope this user keeps it going at it's top speed of 4.5% per day in which case it may actually finish in 5-6 days.

[QUOTE=Madpoo;459727]We're almost there... only 2 left.

Unfortunately I have my doubts about the machine running M39890261[/QUOTE]

Only one left now.

However, the original result for [M]39890261[/M] might turn out to be bad, based on the track record of the machine in question.

It's probably good though, because the bad results date from several years before this particular result was turned in.

[QUOTE=GP2;459817]Only one left now.

Ok we'll be ready!

[QUOTE=GP2;459817]Only one left now.

However, the original result for [M]39890261[/M] might turn out to be bad, based on the track record of the machine in question.

It's probably good though, because the bad results date from several years before this particular result was turned in.[/QUOTE]

That system does have a spotty track record... lifetime of 12 bad, 27 good.

That result came in June 2008 (just about the end of the Primenet v4 era) and it's most recent bad result was a couple years earlier in Jan 2006.

Besides this exponent, there's only one other "unknown" by that CPU: [M]41443327[/M]

[QUOTE=Madpoo;459727]We're almost there... only 2 left.

Unfortunately I have my doubts about the machine running M39890261

It reports only 7 days left (and 12 days left to expiration), however this machine is very sporadic.
[/QUOTE]

User "pi" got poached by "ANONYMOUS", so we're done ! :tu:

[QUOTE=proxy2222;459852]User "pi" got poached by "ANONYMOUS", so we're done ! :tu:[/QUOTE]

ok lets just update the new count for 42,000,000 which is around 2000:smile:

[QUOTE=proxy2222;459852]User "pi" got poached by "ANONYMOUS", so we're done ! :tu:[/QUOTE]

Someone jumped the gun. Maybe. The computer failed to check in yesterday which, historically, was an indicator that it might not check in for a few more days and it would have zero progress done.

I guess we'll know if/when it finishes up.

[QUOTE]2017-03-27 All exponents below 40,000,000 double-checked[/QUOTE]

It was pretty warm today for March.

[QUOTE=ATH;459887]It was pretty warm today for March.[/QUOTE]

Whoops... the dangers of cut-and-paste. :smile: Thanks, good catch.

... AND THEN THERE WERE FIVE!
 

So, just about 3-1/2 month after the 71,000,000 milestone, we are about to get to the 72 million.

[URL="https://www.mersenne.org/assignments/?exp_lo=71770529&exp_hi=72000000&execm=1&exp1=1&extf=1&exdchk=1"]http://https://www.mersenne.org/assignments/?exp_lo=71770529&exp_hi=72000000&execm=1&exp1=1&extf=1&exdchk=1[/URL]

[QUOTE=rudy235;461591]So, just about 3-1/2 month after the 71,000,000 milestone, we are about to get to the 72 million.

[URL="https://www.mersenne.org/assignments/?exp_lo=71770529&exp_hi=72000000&execm=1&exp1=1&extf=1&exdchk=1"]http://https://www.mersenne.org/assignments/?exp_lo=71770529&exp_hi=72000000&execm=1&exp1=1&extf=1&exdchk=1[/URL][/QUOTE]

Yeah, and we'd be further along on the next one as well except the user Xolotl had a range of those in the 72M range checked out but seems to have let them all sit a while and they've been expiring the past few days. Oh well... they're cat0 now so they should hopefully get picked up by machines that will finish quick.

[QUOTE=Madpoo;461606]Yeah, and we'd be further along on the next one as well except the user Xolotl had a range of those in the 72M range checked out but seems to have let them all sit a while and they've been expiring the past few days. Oh well... they're cat0 now so they should hopefully get picked up by machines that will finish quick.[/QUOTE]

Lame... I checked the last 5 in the <72M countdown and 4 of the 5 probably won't finish before they expire. I could be wrong, they could pick up some steam, but they're poking along at a snails pace. One of them hasn't even checked in for 12 days.

The only one I think will be okay is M71907713.

Here are my projections for those 4:
[CODE]exponent Expiry DaysToGo
71770529 2 31.9
71892109 14 18.0
71913659 19 54.4
71963201 23 110.6[/CODE]

(Note: this isn't an encouragement to poach... they'll expire and get reassigned, hopefully by someone who finishes quick, and these assignments will crawl across later and be a double-check)

so.. if not poached 71963201 will take about a month to be completed.. sheesh!

[QUOTE=rudy235;461737]so.. if not poached 71963201 will take about a month to be completed.. sheesh![/QUOTE]

Updated:
[CODE]exponent Expiry DaysToGo
71770529 0 33.0
71892109 12 18.9
71907713 15 5.2
71913659 17 52.8
71963201 21 116.9[/CODE]

I included 71907713 to show that I'm projecting it's completion later than it's self-reported value (2 days).

[QUOTE=Madpoo;461825]Updated:
[CODE]exponent Expiry DaysToGo
71770529 0 33.0
71892109 12 18.9
71907713 15 5.2
71913659 17 52.8
71963201 21 116.9[/CODE]

I included 71907713 to show that I'm projecting it's completion later than it's self-reported value (2 days).[/QUOTE]

They seem to be almost going backwards!

[QUOTE=rudy235;461827]They seem to be almost going backwards![/QUOTE]

I would be very impressed if you work out how to go backwards in a LL test efficiently.

[QUOTE=henryzz;461911]I would be very impressed if you work out how to go backwards in a LL test efficiently.[/QUOTE]
[SIZE="3"]
ok, I'll bite.:cat:[/SIZE]

They were, lets say, supposed to end in according to Madpoo like this

Here are my projections for those 4:
[CODE]exponent Expiry DaysToGo
71770529 2 31.9
71892109 14 18.0
71913659 19 54.4
71963201 23 110.6[/CODE]

Now 1 day and 14 hours later (1.6 days) they are looking like this:

Updated:

[CODE]exponent Expiry DaysToGo
71770529 0 33.0
71892109 12 18.9

71913659 17 52.8
71963201 21 116.9[/CODE]

So 3 of the 4 exponents are getting further away from completion.
The other one is advancing but barely: 1.6 days. But that 1.6 days is the 1.6 days interval between the posts, so it is static.

One of the five still pending 71770529 got reassigned and that is good news because I believe this new user will actually finish [URL="https://www.mersenne.org/report_exponent/?exp_lo=71770529&full=1"]on time.[/URL]

[QUOTE=rudy235;461921]One of the five still pending 71770529 got reassigned and that is good news because I believe this new user will actually finish [URL="https://www.mersenne.org/report_exponent/?exp_lo=71770529&full=1"]on time.[/URL][/QUOTE]

Yes, ETA is 46 hours.

[QUOTE=rudy235;461914]...
Now 1 day and 14 hours later (1.6 days) they are looking like this:
...
So 3 of the 4 exponents are getting further away from completion.
The other one is advancing but barely: 1.6 days. But that 1.6 days is the 1.6 days interval between the posts, so it is static.[/QUOTE]

It's because progress on those assignments has significantly slowed (or halted, even though a couple are still reporting in each day, but with zero progress made).

That has the effect of making their "% per day" calculation go down, thus extending how long it'll take to finish at that estimated daily rate.

In the case of M71913659 it hasn't updated since June 9. My calculations only take intervals between updates to calculate the daily average... if they don't update, their daily average remains the same but without any progress made at all, the "days to complete" will actually go down with the expectation that it's still calculating but not reporting in. Which probably isn't true in many cases. :smile:

For those other ones, if they check in day after day and haven't made any progress at all from one to the next, their daily average starts to plummet and the "days to complete" will rise and rise...

Here's the latest on the 3 slow pokes:
[CODE]exponent Points RealEta Expiry DaysToGo
71892109 75 2017-07-14 17:24 10 19.7
71913659 54 2017-08-15 00:05 14 51.0
71963201 45 2017-10-23 16:47 19 120.6[/CODE]

Some info on each of those:[LIST][*]M71892109 has been stuck at 78.9% since April 15 and it's current daily avg is 1.0398[*]M71913659 has been stuck at 48.6% since April 20 and it's current daily avg is 0.7741[*]M71963201 is currently at 38.5% and it reports in about every 2 days and lately it only progresses about 1% every 3-4 days (most days it reports zero progress made). Funny, because it started out strong and a couple times it progressed 6% in a single day. But mostly it just putters along. It's current daily avg is 0.5079[/LIST]
These machines clearly should not be doing cat 0 work, but at this point they wouldn't qualify for the lower categories because they're not turning in sufficient GHz/day and they'll have some recent expirations once these actually hit that point.

[QUOTE=Madpoo;461999]

These machines clearly should not be doing cat 0 work, but at this point they wouldn't qualify for the lower categories because they're not turning in sufficient GHz/day and they'll have some recent expirations once these actually hit that point.[/QUOTE]

All of them (except for [URL="https://www.mersenne.org/report_exponent/?exp_lo=71770529&full=1"]71770529[/URL]) and [URL="https://www.mersenne.org/report_exponent/?exp_lo=71907713&full=1"]71907713[/URL] are not going to finish in their allotted times. And that means best case scenario (barring poaches) about 28 days from now.

[ATTACH]16325[/ATTACH]

We are getting closer to the end of the "classical view"

proxy2222 got rid of [URL="https://www.mersenne.org/report_exponent/?exp_lo=71913659&exp_hi=&full=1"]71913659[/URL] and now there only 3 left!

[QUOTE=rudy235;462057][ATTACH]16325[/ATTACH]

We are getting closer to the end of the "classical view"[/QUOTE]

1 LL .... yes
2 LL (DC) ... a LONG LONG time to go.

Two to go
 

[CODE]There are 2 assignments

Exponent

71892109 LL LL, 78.90% 8 1 2017-04-06 2017-06-26 2017-06-27 2017-06-28 -Anonymous-
71963201 LL LL, 38.70% 17 13 2017-04-15 2017-06-26 2017-06-27 2017-07-10 overclocked_by_default
[/CODE]

[QUOTE=rudy235;462166][CODE]There are 2 assignments

Exponent

71892109 LL LL, 78.90% 8 1 2017-04-06 2017-06-26 2017-06-27 2017-06-28 -Anonymous-
71963201 LL LL, 38.70% 17 13 2017-04-15 2017-06-26 2017-06-27 2017-07-10 overclocked_by_default
[/CODE][/QUOTE]

Both of which I expect will expire before finishing.

M71963201 had an amazing burst of energy a week ago... it went from 38.5% to 38.7% where it had been stuck for the previous week or so. Maybe in the next few days it'll jump another 0.2%.

M71892109 is still stuck at 78.9%. If the owner of that assignment ever figured out what the problem is, it should be able to finish in just a couple more days, but that's a big if. When it was progressing normally it was able to do 13.7% per day.

[QUOTE=Madpoo;462541]Both of which I expect will expire before finishing.

M71963201 had an amazing burst of energy a week ago... it went from 38.5% to 38.7% where it had been stuck for the previous week or so. Maybe in the next few days it'll jump another 0.2%.

M71892109 is still stuck at 78.9%. If the owner of that assignment ever figured out what the problem is, it should be able to finish in just a couple more days, but that's a big if. When it was progressing normally it was able to do 13.7% per day.[/QUOTE]

One expired and already finished by user ATH and one poached.

So, we're done with 72M! :tu:

[QUOTE=proxy2222;462955]One expired and already finished by user ATH and one poached.

So, we're done with 72M! :tu:[/QUOTE]

I just noticed that (much later) and updated the milestone page. :smile:

We are getting close to a mini-milestone of sorts.

In the [URL="https://www.mersenne.org/report_classic/"]"colorful stats report"[/URL] the range 35,100,000 to 40,250,000 is about to finish and [B]only 4 numbers remain to be double checked.[/B]

[ATTACH]16486[/ATTACH]

[QUOTE=rudy235;463702]We are getting close to a mini-milestone of sorts.

In the [URL="https://www.mersenne.org/report_classic/"]"colorful stats report"[/URL] the range 35,100,000 to 40,250,000 is about to finish and [B]only 4 numbers remain to be double checked.[/B]

[ATTACH]16486[/ATTACH][/QUOTE]Looking just now it shows [b]0[/b].

Time to merge in another line.

[ATTACH]16530[/ATTACH]

There are only 7 remaining exponents to complete the 73M milestone

However this one will expire before completion.

72989101 LL LL, 63.10% 14 [COLOR="Red"]29[/COLOR] 2017-05-18 2017-08-02 2017-08-03 2017-08-31 Kent Freeze
©1996-2017 Mersenne Research, Inc. Current time: 2017-08-02 17:49 UTC - Page rendered in 0.0518s

My DCs have crossed over from 40M to 41M.

[QUOTE=rudy235;464741]There are only 7 remaining exponents to complete the 73M milestone

However this one will expire before completion.

72989101 LL LL, 63.10% 14 [COLOR="Red"]29[/COLOR] 2017-05-18 2017-08-02 2017-08-03 2017-08-31 Kent Freeze
©1996-2017 Mersenne Research, Inc. Current time: 2017-08-02 17:49 UTC - Page rendered in 0.0518s[/QUOTE]

There are 4 left, 3 of which won't finish in time (good job, ATH, on the one that actually will finish). That's based on my analysis of their past progress... they fall into that category of "mystery machines that check in daily but report no progress for days or weeks on end"

I'm not saying I'm poaching those 3 exponents, but don't be surprised if someone magically turns in results for those just before they expire. :smile:

Here's my analysis of the *actual* days-to-complete for these 3, just based on the average %/day of those assignments.
72831991 - 113.3 days (expires in 4 days)
72893621 - 394.2 days (expires in 10 days)
72989101 - 34 days (expires in 11 days)

[QUOTE=Madpoo;464863]There are 4 left, 3 of which won't finish in time (good job, ATH, on the one that actually will finish). That's based on my analysis of their past progress... they fall into that category of "mystery machines that check in daily but report no progress for days or weeks on end"

I'm not saying I'm poaching those 3 exponents, but don't be surprised if someone magically turns in results for those just before they expire. :smile:

Here's my analysis of the *actual* days-to-complete for these 3, just based on the average %/day of those assignments.
72831991 - 113.3 days (expires in 4 days)
72893621 - 394.2 days (expires in 10 days)
72989101 - 34 days (expires in 11 days)[/QUOTE]

I am not poaching those either, however, I will not cry or decry if (when) someone else does.

The server has no faith in us :smile: It expect us to take 26 months just for 521 exponents:

[QUOTE]Countdown to first time checking all exponents below 75M: 521 (Estimated completion : 2019-10-07)[/QUOTE]

[QUOTE=ATH;464886]The server has no faith in us :smile: It expect us to take 26 months just for 521 exponents:[/QUOTE]

LOL... it's just the latest "estimated time to completion" of all the exponents in that range.

Those client estimates are usually wrong anyway, and if a client stops reporting in then it's even more useless so I'm not entirely sure that milestone page should even report that. Oh well.

[QUOTE=Madpoo;464896]Those client estimates are usually wrong anyway, and if a client stops reporting in then it's even more useless so I'm not entirely sure that milestone page should even report that. Oh well.[/QUOTE]It was suggested many posts ago that those useless estimated completion dates be removed. They serve no purpose whatsoever. Might as well just post random numbers, they would make as much sense IMO.

I'm all in favour of showing more details (instead of whitespace and so-called "clean" layouts) but when those details are useless then they should be removed, or replaced with something else that actually conveys information.

[QUOTE=retina;464912]It was suggested...[/QUOTE]
+1

[QUOTE=retina;464912]It was suggested many posts ago that those useless estimated completion dates be removed. They serve no purpose whatsoever. Might as well just post random numbers, they would make as much sense IMO.

I'm all in favour of showing more details (instead of whitespace and so-called "clean" layouts) but when those details are useless then they should be removed, or replaced with something else that actually conveys information.[/QUOTE]

I'll take a look at it, maybe think about whether there's anything else useful I could put there instead.

FYI, I turned in the last 3 results for the < 73M exponents. They weren't going to finish in time, that much was clear. Now they can just turn in their double-check in however many days or years they'll actually take.

[QUOTE=Madpoo;464984]I'll take a look at it, maybe think about whether there's anything else useful I could put there instead.

FYI, I turned in the last 3 results for the < 73M exponents. They weren't going to finish in time, that much was clear. Now they can just turn in their double-check in however many days or years they'll actually take.[/QUOTE]

Just a few suggestions - such as a countdown to n=100M and a countdown to n=332192809 (100M digits)

:)

The top two lines of [URL=https://www.mersenne.org/report_classic/]this[/URL] report can now be merged.

[QUOTE=NBtarheel_33;465429]The top two lines of [URL=https://www.mersenne.org/report_classic/]this[/URL] report can now be merged.[/QUOTE]

I'm not sure what the original thinking was of those breakpoints in the chart. I thought maybe it was based on the average FFT size used for those exponents or something like that (except for the large 0-35.1M section).

They have merged in the past. I remember vaguely when what is now the line up to 35'100,000 was divided.:smile:

[QUOTE=rudy235;465477]They have merged in the past. I remember vaguely when what is now the line up to 35'100,000 was divided.:smile:[/QUOTE]

Oh, hmm... looks like that's the case.

Okay, I merged it and made a few minor layout tweaks.

While you are busy with layout tweaks on that page, the column "TwoLL" needs to accommodate one digit more, it should have the same width as the column "Facored" or "Numbers".

Jacob

[QUOTE=S485122;465493]While you are busy with layout tweaks on that page, the column "TwoLL" needs to accommodate one digit more, it should have the same width as the column "Facored" or "Numbers".

Jacob[/QUOTE]

I see no reason why they can't fix [u]both[/u] issues simultaneously.:smile:

[QUOTE=Madpoo;465450]I'm not sure what the original thinking was of those breakpoints in the chart. I thought maybe it was based on the average FFT size used for those exponents or something like that (except for the large 0-35.1M section).[/QUOTE]

That is exactly how it was: one line for each FFT size, going back to the days of low seven-figure exponents. As each FFT size bracket was "eliminated" (i.e. all of the exponents requiring a given FFT size were double-checked), they were merged together into the larger line at the top of the report, and the FFT size for all of the merged exponents was recorded as "Many" or "Various".

Historically, the rationale likely was that each larger FFT size meant an increase in the number of P90 CPU years required to test a given group of exponents, and since this report gives the number of P90 CPU years left in each interval, it made sense to break the exponents down by FFT size.

Incidentally, imagine testing exponents near 79.3M on a P90! The fact that we are completing this table (though completing the double-check stage is still probably 10+ years off) is a testament to both the massive increases in computing power that we have witnessed in the last 20 years, and the amazing effort that both administrators and users bring to GIMPS day after day, year after year (and now, decade after decade!).

[QUOTE=Madpoo;465486]Oh, hmm... looks like that's the case.

Okay, I merged it and made a few minor layout tweaks.[/QUOTE]

The TwoLL column does not add up at the bottom; it looks as though the top cell in that column is not being added into the total.

[QUOTE=NBtarheel_33;465614]The TwoLL column does not add up at the bottom; it looks as though the top cell in that column is not being added into the total.[/QUOTE]

Whoops, yeah. Fixed.

changes to milestone page
 

I was making some tweaks to the milestone page.

First off was removing the "ETA" since they're wildly inaccurate. I kept in the "unassigned" count but gave it a friendlier moniker: "available"

Next up was re-wording past milestones... I wanted to remove some ambiguity, especially around "proving" the past Mersenne primes in regards to the official order of them when all below have been double-checked.

And speaking of "double checking", that's also a misnomer because it may be more than double-checked... the proper term, to me, is "verified" so I've adjusted everything accordingly.

Plus, rather than talk about verifying all "exponents" below XYZ I opted for the phrase "verified all tests below XYZ" because that's really what's being verified/double-checked... the test, not the exponent.

Anyway, it probably matters to nobody here since we all know, but when primes are found or bigger milestones reached like those that "prove" the order of a known prime, we don't want to be confusing to anyone on the outside.

The wording changes can be found on the mockup page...they're not live yet since I thought it'd be good to get people's input:
[URL="https://www.mersenne.org/report_milestones/default.mock.php"]Milestone page -
test[/URL]

[QUOTE=Madpoo;466133]...
[URL="https://www.mersenne.org/report_milestones/default.mock.php"]Milestone page -
test[/URL][/QUOTE]There are still many place where the word "double" appears.

Plus, since you are making everything consistent then the three different discovered texts could be altered: "discovered!!", "discovered!" and "is discovered!"

Ugh! It is a mess ... of "exponents tested" and "tests verified". Never noticed how much of a mess it was.

While we're at it, why the two different notations, one using exponent and one using Mersenne number?
Countdown to verifying all tests below 41M
Countdown to verifying all tests below M(42643801)

[QUOTE=axn;466158]Ugh! It is a mess ... of "exponents tested" and "tests verified". Never noticed how much of a mess it was.

While we're at it, why the two different notations, one using exponent and one using Mersenne number?
Countdown to verifying all tests below 41M
Countdown to verifying all tests below M(42643801)[/QUOTE]

Because while 42643801 is an exponent of a proven Mersenne prime, the 41M only reflects "a round number" which it is only to represent a symbolic milestone.
AFAIK

[QUOTE=rudy235;466165]Because while 42643801 is an exponent of a proven Mersenne prime, the 41M only reflects "a round number" which it is only to represent a symbolic milestone.
AFAIK[/QUOTE]

Better to change it to 41,000,000 for consistency with all the sections below.

Or possibly change everything in all sections to "41 million".

[QUOTE=GP2;466166]Better to change it to 41,000,000 for consistency with all the sections below.

Or possibly change everything in all sections to "41 million".[/QUOTE]

Scientific notation maybe? 41e6? That's what I personally use in casual conversation, probably because that's what I use in SQL queries. :smile:

[QUOTE=Madpoo;466190]Scientific notation maybe? 41e6? That's what I personally use in casual conversation, probably because that's what I use in SQL queries. :smile:[/QUOTE]

Since there's one entry per line, there's no need for horizontal compaction.

Also, the milestones page should be straightforward and understandable by non-participants and non-mathematicians, since reporters or the general public might occasionally view it, for instance when there is a burst of publicity surrounding a new prime discovery.

[QUOTE=GP2;466218]Since there's one entry per line, there's no need for horizontal compaction.

Also, the milestones page should be straightforward and understandable by non-participants and non-mathematicians, since reporters or the general public might occasionally view it, for instance when there is a burst of publicity surrounding a new prime discovery.[/QUOTE]

Fine, I'll give it a go. Changed to "million" in the text. I didn't want to do like 41,000,000 since comma/period is a locale specific thing. I was going to do something like "41 000 000" but it just looked weird, especially when the known primes are just lumped together, no thousands separator at all.

Then again, all of the historical entries are using comma for the thousands separator. Hmm... maybe I should be consistently US biased with those. :smile: I'll mull it over anyway.

[QUOTE=Madpoo;466254]...
Then again, all of the historical entries are using comma for the thousands separator. Hmm... maybe I should be consistently US biased with those. :smile: I'll mull it over anyway.[/QUOTE]For dates the site already adopted the international standard of yyyy-mm-dd, one could adopt the international standard for thousands separators in numbers as well : the space separator.

Jacob

(Now if mprime / prime95 could use the yyyy-mm-dd date format on screen, in its logs and result files... But that wish is for another thread.)

[QUOTE=Madpoo;466254]Then again, all of the historical entries are using comma for the thousands separator. Hmm... maybe I should be consistently US biased with those. :smile: I'll mull it over anyway.[/QUOTE]

Actually, the historical entries could use "million" too, since all of them are round numbers. All those zeros are a visual distraction from the truly relevant digits.

[QUOTE=GP2;466268]Actually, the historical entries could use "million" too, since all of them are round numbers. All those zeros are a visual distraction from the truly relevant digits.[/QUOTE]

On the [URL="https://www.mersenne.org/report_milestones/default.mock.php"]https://www.mersenne.org/report_milestones/default.mock.php[/URL] page I made a few changes...

Changed the thousands separator to a &thinsp; (thin space) entity and replaced ",000,000" with " million"

Some of that data comes from SQL itself so even on the regular page you'll see some thin spaces.

I should be able to do some stuff in PHP where it's already using the formatting functions to add commas, but I foresee a problem with people who crawl various pages to collect data (rather than using the convenient XML data generated daily, or they crawl huge sections of the exponent report pages and aren't using the XML option there either). Their crawlers/parsers are likely to choke on a bunch of html entities where there used to be commas or no separator at all.

In light of that, I'll take baby steps if using thin spaces as a separator is acceptable (which sounds more and more like a good international standard).

[QUOTE=Madpoo;466383]On the [URL="https://www.mersenne.org/report_milestones/default.mock.php"]https://www.mersenne.org/report_milestones/default.mock.php[/URL] page I made a few changes...[/QUOTE]In some places it says "All exponents below" and others is says "All tests below".

In some places is says "discovered!!" and others is says "discovered!" and others it says "is discovered!".

Re the space separator: Maybe other [i]languages[/i] use different conventions for number digit grouping, but I've not seen any English language usage that uses anything but commas. I would have thought that since the site is in English that normal English groupings should apply. That way any automated translations can them reformat accordingly. If you make it non-standard without commas then translations won't recognise the numbers and they won't get reformatted.

[QUOTE=retina;466384]Re the space separator: Maybe other [i]languages[/i] use different conventions for number digit grouping, but I've not seen any English language usage that uses anything but commas. I would have thought that since the site is in English that normal English groupings should apply. That way any automated translations can them reformat accordingly. If you make it non-standard without commas then translations won't recognise the numbers and they won't get reformatted.[/QUOTE]

Why would translations need to reformat the number?

[URL="https://en.wikipedia.org/wiki/Thin_space"]According to Wikipedia[/URL], "Thin spaces are recommended for use as a thousands separator for measures made with SI units." I wonder if "dimensionless" is an SI unit. :smile:

In addition to "Thin space" at U+2009 or &thinsp; (e.g., 70 000 000), there is also a "Narrow no-break space" at U+202F (e.g., 70 000 000), if it's necessary to avoid splitting the number across more than one line.

[QUOTE=GP2;466396][URL="https://en.wikipedia.org/wiki/Thin_space"]According to Wikipedia[/URL], "Thin spaces are recommended for use as a thousands separator for measures made with SI units."[/QUOTE]Well some overzealous editor at WP doesn't mean much to me. WP also recommends using MiB and GiB, but in practice "no one" ever does. I am referring to [i]normal[/i] English usage. IME [i]normal[/i] English usage uses commas. Plus, exponents aren't "measures made with SI units", they're integers, they aren't measured, they aren't units, so there that problem with using a space (thin or otherwise).

[QUOTE=retina;466397]Well some overzealous editor at WP doesn't mean much to me. WP also recommends using MiB and GiB, but in practice "no one" ever does.[/QUOTE]

It sounds silly when pronounced out loud, but it's not a problem in print, and actually it does get used. Amazon, for instance, quotes the memory size for its cloud instances in GiB, but the SSD storage size in GB ([URL="https://aws.amazon.com/ec2/instance-types/#m3"]for example[/URL]). I imagine that would be because manufacturers quote specs that way. It's nice to know the exact figure and not have to rely on an approximation which diverges ever more, as we start to talk about terabytes and petabytes (e.g., a ten percent difference between a TB and a TiB).

[QUOTE]I am referring to [i]normal[/i] English usage. IME [i]normal[/i] English usage uses commas. Plus, exponents aren't "measures made with SI units", they're integers, they aren't measured, they aren't units, so there that problem with using a space (thin or otherwise).[/QUOTE]

It looks perfectly fine with thin spaces. There is no loss of clarity. Not sure why you have a bee in your bonnet about this.

[QUOTE=GP2;466402]I...
It looks perfectly fine with thin spaces. There is no loss of clarity. Not sure why you have a bee in your bonnet about this.[/QUOTE]

LOL - I don't know either. I think I'm happy with the thin spaces there... at least on the pages most likely to be viewed by outsiders (i.e. journalists or new folks when new primes are found, and no, there isn't one waiting to be announced, it's just proactive).

Where I work, I deal with websites in a variety of languages and locales, and it's amazing how many quirky things there are with measurements, even between the English speaking countries.

As the old joke goes, England and America are two countries divided by a common language. :smile: I'm just imagining how the discussion will evolve when we start talking about billion digit numbers, because English speakers in some parts of the world will be agog that we're talking about 10^12 so soon, when really I would mean 10^9.

[QUOTE=GP2;466402]It looks perfectly fine with thin spaces. There is no loss of clarity. Not sure why you have a bee in your bonnet about this.[/QUOTE]Like I mentioned, the translations are not so clear, and can be garbled. If all you deal with is English then you probably won't encounter any problems.

[QUOTE=retina;466416]Like I mentioned, the translations are not so clear, and can be garbled. If all you deal with is English then you probably won't encounter any problems.[/QUOTE]

Whenever I run the text through Google Translate, from English to various other languages, it leaves the numbers unaltered, except it turns the thin spaces into regular spaces.

The discussion about units has been moved to another thread:
[URL]http://www.mersenneforum.org/showthread.php?t=22548[/URL]

Regarding the 41 million milestone
 

I see that there are 14 exponents that need to be cleared for that milestone (41 million) to be completed.
However, almost half of those exponents are not moving and they are past due and are not moving at all. I believe they are six of those there.

41 million milestone. There are only 10 exponents left.

However, 6 of them seem to be stuck.

This one, for example, has been totally abandoned. (apparently)

40992157 D LL, 91.80% 29 0 2017-08-01 2017-09-01 2017-09-02 2017-09-01 samol

[QUOTE=rudy235;466852]41 million milestone. There are only 10 exponents left.

However, 6 of them seem to be stuck.

This one, for example, has been totally abandoned. (apparently)

40992157 D LL, 91.80% 29 0 2017-08-01 2017-09-01 2017-09-02 2017-09-01 samol[/QUOTE]

Probably not that one since, as you show, the assignment updated today. As for the others...could be... it's hard to tell sometimes if they're abandoned because some people (like George himself) may kick a bunch of work off and then go on a vacation. :smile:

4 of the 6 that haven't checked in lately are expiring soon anyway and I'm guessing someone will probably pick off the others if the assignee still hasn't updated their progress.

[QUOTE=Madpoo;466894]Probably not that one since, as you show, the assignment updated today. [/QUOTE]
I'm very sorry! I meant this one!
[code]
40986997 D LL, 99.80% 27 -11 2017-07-31 2017-08-22 2017-08-23 2017-08-22 Kish[/code]

1 left for milestone 41 million.
[URL="https://www.mersenne.org/report_exponent/?exp_lo=40992617&full=1"]40992617[/URL]

[QUOTE=rudy235;467106]1 left for milestone 41 million.
[URL="https://www.mersenne.org/report_exponent/?exp_lo=40992617&full=1"]40992617[/URL][/QUOTE]

C'mon ATH, pick up the pace. :smile: (I know, only assigned yesterday or whatever...)

[QUOTE=Madpoo;467134]C'mon ATH, pick up the pace. :smile: (I know, only assigned yesterday or whatever...)[/QUOTE]

Done.

:bow wave:

OK, so our next BIG milestone is the first time checking of the mersenne prime of exponent 74,207,281
There are 39 exponents left to get there but one of them [URL="https://www.mersenne.org/report_exponent/?exp_lo=74101927&full=1"]74101927[/URL] is suppossed to take ~28 months to complete!

There must be a serious glitch in the program that estimates the completion which is supposed to end on 2019-12-21, because even a simple "rule of three" suggests around 10 months (293 days). Of course, it will expire in 21 days and let's hope it is given to ATH :smile:

[QUOTE=rudy235;467222]OK, so our next BIG milestone is the first time checking of the mersenne prime of exponent 74,207,281
There are 39 exponents left to get there but one of them [URL="https://www.mersenne.org/report_exponent/?exp_lo=74101927&full=1"]74101927[/URL] is suppossed to take ~28 months to complete!

There must be a serious glitch in the program that estimates the completion which is supposed to end on 2019-12-21, because even a simple "rule of three" suggests around 10 months (293 days). Of course, it will expire in 21 days and let's hope it is given to ATH :smile:[/QUOTE]

18 of those 39, I'm estimating will take longer to complete than it will to reach the expiration point.

These in particular are estimating to take a REALLY long time:
74181521
74051921
74055439
74097707
74101927
74202361

That last one ... based on it's past 47 data points of check-ins, I'm showing it's actual days-to-complete as 2800 days. It's moved an awesome 0.0348% per day since it first checked in with any progress.

74181521 is estimating 1210 days, the rest are 220-550 days, somewhere in that big range. They'll all expire in due course.

The one you mentioned, 74101927, I'm estimating 213 days to go with an average 0.3581% per day, with 55 data points. The fact that the client is now reporting a much longer ETA means it's probably slowed down even more than that. I could check, but I'm too lazy right now. :smile: The client is saying 847 days. Meh.

When the milestone of 42M is reached in a few months, this will simultaneously result in the double-checking of all remaining exponents whose first time test was in 2007. All exponents from earlier years have already been double-checked, as well as all of the very early exponents with no timestamp.

Currently there are maybe four 2007-era exponents in the 41.1M range, three of which will complete the double-check in a few days, and one that is assigned but hasn't started yet ([M]41156153[/M]); and there is also the exponent [M]41979199[/M] which is moving along slowly and will probably expire before the current assignee finishes it.

If it matches, 41719147 will complete in 21.5 hours. :smile: Then I'll be back to Cat 3 stuff. :blush:

[QUOTE=Madpoo;466383]On the [URL="https://www.mersenne.org/report_milestones/default.mock.php"]https://www.mersenne.org/report_milestones/default.mock.php[/URL] page I made a few changes...[/QUOTE][QUOTE=retina;466384]In some places is says "discovered!!" and others is says "discovered!" and others it says "is discovered!".[/QUOTE]Now that the new page appears to be up, it still has the weird unexpected factorials, weirder unexpected double factorials, and the unexpected "is".

[QUOTE=GP2;467245]When the milestone of 42M is reached in a few months, this will simultaneously result in the double-checking of all remaining exponents whose first time test was in 2007. All exponents from earlier years have already been double-checked, as well as all of the very early exponents with no timestamp.

Currently there are maybe four 2007-era exponents in the 41.1M range, three of which will complete the double-check in a few days, and one that is assigned but hasn't started yet ([M]41156153[/M]); and there is also the exponent [M]41979199[/M] which is moving along slowly and will probably expire before the current assignee finishes it.[/QUOTE]

Here are all 8 of the exponents, still unverified, from earlier than '2008-01-01'
[CODE]exponent Date
41979199 2007-07-04
41386967 2007-12-31
41404529 2007-12-31
41188673 2007-12-31
41166991 2007-12-31
41156153 2007-12-31
41184851 2007-12-31
41439883 2007-12-31[/CODE]

And for fun, there are 11,038 unverified from < '2009-01-01' ... largest one is M48452233, so when that becomes our new "all verified below", then we can cross off the 2008 and below stuff as done-and-done.

[QUOTE=retina;467259]Now that the new page appears to be up, it still has the weird unexpected factorials, weirder unexpected double factorials, and the unexpected "is".[/QUOTE]

Well, I didn't fix any other text besides what I mentioned. :smile:

Okay, I made those consistent ... sorry, but I think a new prime discovery deserves at least one exclamation point (but only one). I also removed the extraneous "is" from the earlier primes (they didn't show up in the later listings).

Changes on the "mock" page for now and I'll probably just roll those out when the next milestone gets checked off the list.

Since we're quibbling about punctuation...

[CODE]
Countdown to first time checking all exponents below 74 million
[/CODE]

We could add a hyphen, to make it "first-time checking".

Makes it a little less ambiguous. Otherwise, a reporter or other non-participant visiting this page might wonder what "time checking" is.

The number of users with more than 100,000 GHz days credit is now too gross.

Edit: I meant two gross. :smile:

[QUOTE=cuBerBruce;467293]The number of users with more than 100,000 GHz days credit is now too gross.

Edit: I meant two gross. :smile:[/QUOTE]

Hey, just because some of the participants may have poor hygiene, you don't have to call them gross.

Oh... you meant the unit of measure. :smile:

74 million

DONE!

[code]
ATH 5960X 73968971 C 2017-09-19 20:21 5.9 208.5514
0B24A31B1BAD71__
[/code]

M74095187 -- waiting...
 

Still waiting on M74095187 now. The history of check-ins from the cpu doing it is a little spotty.

Lately it's only been checking in every 3 days or so, and only doing < 1% per day, if it even progressed at all.

Strange, because it started out okay and even managed to do 18% in a day, but that was a month ago.

Basically, in the past 30 days it's only moved from 68.4% to 80.2% ... avg 0.4% daily.

I'm kind of waiting to see if it checks in again in the next 24 hours since it should be soon, if it sticks to that every 3 days thing. I guess we'll see.

Madpoo: there is a previous result AE5C73EB430E2D__ for 74095187 from July 2017 by mikeblas that is considered "suspect". Do you know why is that?

[QUOTE=rudy235;468860]is considered "suspect". Do you know why is that?[/QUOTE]
Suspect results have an error code when submitted. Basicly the program ran into trouble at some point of the calculation and detected that. Prime95 has pretty decent error-detection and can sometimes do the iterations with errors again (or with a different FFT size). So results that have an error code when reported can still produce the right residue, but we're less confident of the results (which is why it is handed out for an extra LL test immediately).

By the way, if a LL result has zero error code it doesn't automaticly mean no errors occurred. About 1-2% of the LL tests without error codes are still proven wrong when they are DoubleCheck/TripleChecked.

[QUOTE=VictordeHolland;468874]Suspect results have an error code when submitted. Basicly the program ran into trouble at some point of the calculation and detected that. Prime95 has pretty decent error-detection and can sometimes do the iterations with errors again (or with a different FFT size). So results that have an error code when reported can still produce the right residue, but we're less confident of the results (which is why it is handed out for an extra LL test immediately).

By the way, if a LL result has zero error code it doesn't automaticly mean no errors occurred. About 1-2% of the LL tests without error codes are still proven wrong when they are DoubleCheck/TripleChecked.[/QUOTE]

Ditto all that. Last time I checked, results marked suspect had about a 50% chance of being wrong, so it's a great idea to treat those as if they're bad and re-assign them as a first time check again.

As you mentioned, even a "clean" run still has that potential to be bad. Some of the machines out there that generated almost all bad results were coming in as "clean" runs, so even the typical error detection routines (roundoff errors, etc) weren't catching it. Probably just lousy RAM in those cases, not CPU related, if I had to guess.

[QUOTE=Madpoo;468855]Still waiting on M74095187 now. The history of check-ins from the cpu doing it is a little spotty.

Lately it's only been checking in every 3 days or so, and only doing < 1% per day, if it even progressed at all.
.[/QUOTE]

[SIZE="4"]This a real joke! It progressed from 80.20% to 80.30% in three days.[/SIZE]

If cancer progressed this fast, nobody would die of cancer. (Something else would get them first!)

[QUOTE=rudy235;468945]If cancer progressed this fast, nobody would die of cancer. (Something else would get them first!)[/QUOTE]

really so you think under 8.3 years is enough for something else to kill them first ?

[QUOTE=rudy235;468945]This a real joke! It progressed from 80.20% to 80.30% in three days.
If cancer progressed this fast, nobody would die of cancer. (Something else would get them first!)[/QUOTE]

Perhaps you should check your math, sir: 0.1% in 3 days is 1% per month is 12% per year. If cancer is diagnosed even at 20% of the way to the end, that's around 7 years to reach 100% of the end. If I were told cancer would get me in 7 years, I'm pretty sure I'd feel like I'd die of cancer.

[QUOTE=VBCurtis;468959]Perhaps you should check your math, sir: 0.1% in 3 days is 1% per month is 12% per year. If cancer is diagnosed even at 20% of the way to the end, that's around 7 years to reach 100% of the end. If I were told cancer would get me in 7 years, I'm pretty sure I'd feel like I'd die of cancer.[/QUOTE]

Ohh my G*d. They took me literally.

I was simply making a 'figure of speech". I expect that given the level of the forum participants most would be familiar with a "trope". i.e. a figurative or metaphorical use of a word or expression.

I suspect if I mention that a process is "as slow as watching paint dry" someone would come out and argue that you can't actually notice any change in the appearance of the paint while it dries and thusly there is no value in actually watching.

But yeah it was a lot of fun to read the responses. [B][U]In the meantime, the exponent it remains at 80.30%[/U][/B]

[QUOTE=rudy235;468969]Ohh my G*d. They took me literally.

I was simply making a 'figure of speech". I expect that given the level of the forum participants most would be familiar with a "trope". i.e. a figurative or metaphorical use of a word or expression.

I suspect if I mention that a process is "as slow as watching paint dry" someone would come out and argue that you can't actually notice any change in the appearance of the paint while it dries and thusly there is no value in actually watching.

But yeah it was a lot of fun to read the responses. [B][U]In the meantime, the exponent it remains at 80.30%[/U][/B][/QUOTE]

they also know their fair share of tripe when they read it.