mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Marin's Mersenne-aries (https://www.mersenneforum.org/forumdisplay.php?f=30)
-   -   Strategic Double Clicking (https://www.mersenneforum.org/showthread.php?t=20372)

Madpoo 2015-07-22 21:47

Strategic Double Clicking
 
So here's the thought, if you missed the discussion in the "Lounge" thread:

I'm doing an analysis of [I]computers[/I] to see which of them have had a sketchy track record, and return unusually high #'s of bad residues.

I've done my own pilot program, picking some 34M exponents that had only been tested once by machines with spectacularly low good/bad ratios and it's been pretty effective. Off the top of my head I'd say that in the first 15 or so of these that I've targeted, ~12 of them ended up with my result NOT matching that first test by the iffy machine.

Here's the plan then:
I generate lists of exponents that have only been checked once by these suspect machines, and people volunteer to take them on.

George has suggested either doing a manual assignment request which should work okay for exponents above 40M. For smaller exponents (I'll figure out the exact threshold) his suggestion is to just go ahead and add a "DoubleCheck=xx" line to your worktodo and get it going. It'll apparently create the assignment next time it contacts the server (is that correct, George?)

I'm not quite sure the best way to hand out any exponents... I could just post a list here like I was doing for the triple-checking stuff, and let people request blocks of work (first come, first serve)? Or would people prefer if they volunteer for X exponents and I just PM those, to avoid problems like multiple people accidentally starting on the same things? Not sure how things like that have been done previously.

I probably shouldn't generate a big list ahead of time because it might get auto-assigned to a random participant in the meantime.

So, any comments, questions, ideas?

petrw1 2015-07-22 21:50

How you handled the several hundred triple checks last time seemed to work....

Unless GPU72 can create a new "reserved" work-type for this. :)

Madpoo 2015-07-22 21:52

[QUOTE=petrw1;406321]How you handled the several hundred triple checks last time seemed to work....

Unless GPU72 can create a new "reserved" work-type for this. :)[/QUOTE]

That's kind of my thought too. Not make it too complicated, and trust people to lookup the exponent first and see if it's assigned to someone before they start any work on it.

If two people did end up working on it, that may just be okay because I predict a good amount of these will need 2 more checks anyway, considering I'm guessing the first one is bad... so it wouldn't be the end of the world, but then again some of these will be right.

As the saying goes, even a broken clock is right twice a day, and sometimes these computers manage to get it right.

chalsall 2015-07-22 22:23

[QUOTE=petrw1;406321]How you handled the several hundred triple checks last time seemed to work....

Unless GPU72 can create a new "reserved" work-type for this. :)[/QUOTE]

I would rather not get involved with the DC'ing assignments for two reasons. First, work that doesn't have long-term value. And second, I don't want this effort to be constrained to only GPU72 workers.

On the other hand, it would be trivial for me to ensure the candidates are appropriately TF'ed. The code already exists to prioritize certain candidates.

Aaron, if you PM me the list (or post it here) I'll put them in the system for workers to process.

Madpoo 2015-07-22 22:54

1 Attachment(s)
[QUOTE=chalsall;406323]I would rather not get involved with the DC'ing assignments for two reasons. First, work that doesn't have long-term value. And second, I don't want this effort to be constrained to only GPU72 workers.

On the other hand, it would be trivial for me to ensure the candidates are appropriately TF'ed. The code already exists to prioritize certain candidates.

Aaron, if you PM me the list (or post it here) I'll put them in the system for workers to process.[/QUOTE]

I'm attaching a rough list here. These are all of the currently unassigned exponents, and I used a factor of "bad > good * 1.5" which is pretty loose. But I wanted to get as many in here *solely* for purposes of making sure they're all TF'd to a high enough level.

This also only includes exponents below 58M since I started this out as a way to fill in some vital gaps between where the DC'ing is, and M(48).

I wouldn't start on doing any DC work yet... wait on the TF'ing and then we'll have new lists of the DC eligible stuff.

For those lists, I'll include the # bad, # good, # unknown, # suspect that the same machine has done which people can use as a guide to picking the exponents they're interested in checking.

Mark Rose 2015-07-22 23:56

[QUOTE=chalsall;406323]On the other hand, it would be trivial for me to ensure the candidates are appropriately TF'ed. The code already exists to prioritize certain candidates.[/QUOTE]

What "Option" is needed to get these prioritized assignments? It would be nice if there were a new one that wouldn't return any assignments if the whole list is assigned, so I can prioritize the work I do.

chalsall 2015-07-23 01:17

[QUOTE=Madpoo;406328]I wouldn't start on doing any DC work yet... wait on the TF'ing and then we'll have new lists of the DC eligible stuff.[/QUOTE]

OK. This is now underway. Should take about a week or so.

Note that the first 356 in your list (to 41228987 inclusive) are already appropriately TF'ed, so DC'ing could start on these immediately.

chalsall 2015-07-23 01:26

[QUOTE=Mark Rose;406331]What "Option" is needed to get these prioritized assignments? It would be nice if there were a new one that wouldn't return any assignments if the whole list is assigned, so I can prioritize the work I do.[/QUOTE]

No special option needed; these are simply handed out first to those who choose "WMS" or "LG72D". You, in fact, are already being assigned some of these to take from 70 to 71.

Prime95 2015-07-23 02:29

I've taken these:


48435221,72
48465463,72
48587491,72
48659089,72
49229179,72
49270057,72
49380893,72
49450519,72
49451071,72

All reserved by prime95 after prefixing with "Test=" and adding a suffix of ",1"

Prime95 2015-07-23 02:43

I took these:

Test=37110529,71,1
Test=37126321,71,1
Test=37127599,71,1
Test=37133377,71,1
Test=37139287,71,1
Test=37170901,71,1
Test=37188343,71,1
Test=37188829,71,1
Test=37191487,71,1
Test=37207999,71,1
Test=37227653,71,1
Test=37240271,71,1

Mark Rose 2015-07-23 03:25

[QUOTE=chalsall;406336]No special option needed; these are simply handed out first to those who choose "WMS" or "LG72D". You, in fact, are already being assigned some of these to take from 70 to 71.[/QUOTE]

Awesome. I've now switched 2/3 of my current power to take things to 72 for this project.

Madpoo 2015-07-23 05:15

It also occurs to me that we can help resolve the bad/good ratio by looking at the "suspect" results returned by these systems that are now in need of triple-checking.

First off, it's fun to test these single-checked exponents and find out they were (probably) wrong. That means you might just find a hidden prime. So I get that, and that's one part of this.

Second though is to check the ones those machines did so poorly that the error code was bad enough for the server to mark it suspect and hand it out as a "first time check" again. In those cases we can probably guess that the suspect result is bad, when the 2nd residue didn't match, but that's not always the case, odd as it sounds.

If we go after those, we're (probably) adding more to the "bad" tally of these computers, which helps identify the ones that are doing lousy.

In fact, I had in mind a project I was going to tackle on my own of going after all the 34M exponents that currently need triple checks. There are 281 unassigned exponents needing TCs. Of those, ~ 182 were done by a machine with at least one other known bad residue.

Either taking those 182, or the full set of 281, 34M-35M exponents that need TC's... well, we'd be doing the needed checks on them anyway, and building up the statistical set at the same time.

I *could* do it myself given enough time, but I would love having help. My systems are doing a mix of some of these targeted checks, but I'm also going through *all* of the exponents where Curtis did both a 1st and 2nd check but mismatched. I figured I'd help his team out by figuring out which one was good/bad. That'll take me another couple weeks to complete, I think (there were quite a few).

Sound interesting?

frmky 2015-07-23 05:19

I took the first 8:

Test=34654643,71,1
Test=34665307,71,1
Test=34665871,71,1
Test=34671661,71,1
Test=34682537,71,1
Test=34683613,71,1
Test=34686947,71,1
Test=34691149,71,1

I'll also switch my ~1000 GHz-day/day of TF over to this once my current assignment expires.

Madpoo 2015-07-23 05:21

[QUOTE=Madpoo;406344]Of those, ~ 182 were done by a machine with at least one other known bad residue.[/QUOTE]

Oh, and along the way we may find a few that wind up needing quadruple checks. I've hit a few of those myself. I just noticed that the "182" figure is inflated a little because some of the exponents are showing up twice... both checks done on them were by somewhat unreliable machines, to some degree or another.

For instance, M34670303, M34757869, M34764817 etc.

These are great because we know one way or another we're increasing the "good" count of one and the "bad" of another, which help both systems "reliability scores".

Madpoo 2015-07-23 05:48

[QUOTE=Madpoo;406344]If we go after those, we're (probably) adding more to the "bad" tally of these computers, which helps identify the ones that are doing lousy.[/QUOTE]

By way of example, consider:
[URL="http://www.mersenne.org/M34919821"]M34919821[/URL]

The "suspect" result was returned by a machine that has done 15 bad, 12 good, 11 unknown, and 24 suspect results.

15 to 12 for the bad/good ratio isn't great, but since I was initially targeting higher ratios than that, I wouldn't focus on this right away. Resolving a few more of those unknown or suspect results would help one way or another, to either exclude them or pull them into the query. 4 of those unknown/suspect are under 35M.

Assuming the < 35M stuff goes well and people enjoy doing that kind of triple-checking work, it could go on to higher ranges too. There's no shortage of exponents below 58M needing triple-checks where an "iffy" machine provided one of the results. 1049, give or take (based on a metric of the current "bad > good*0.5" for the systems).

I mean, you get some real doozies like:
[URL="http://www.mersenne.org/M44880113"]M44880113[/URL]

No disrespect at all towards ArtfulDodger, but his systems are some that I've been deliberately doing triple-checks on already for a while. I id'd those machines months ago as some of the more prolific contributors of bad residues. The system in that case has 110 bad results, 17 good ones, 3 unknown, and 20 suspect. That's just not a good track record, and it's done so much work that it seems really weird to me for some reason. :smile:

My analogy I just came up with (so it won't be that great) is like a window washer who spent all this time cleaning a high rise, only to finish and realize he'd been using oil instead of soap and water. :smile:

In fact that might be a good case study to look at the distribution of good/bad over time to see if there's any sort of pattern. But I digress. It's late... my wind tends to wander. :smile:

Or we might find ones like:
[URL="http://www.mersenne.org/M41732771"]M41732771[/URL]

Here's a case of a machine with just 1 known bad result, zero good or unknown, but 15 marked "suspect" because of some error code or another. Hard to say how good/bad it is until we triple-check them... but in these cases we know that the double-checks on each of those suspect results didn't match, so I have a hunch all 15 will be bad.

chalsall 2015-07-23 12:14

[QUOTE=Madpoo;406348]Or we might find ones like:
[URL="http://www.mersenne.org/M41732771"]M41732771[/URL][/QUOTE]

And I notice that these weren't in your original list; I've just had Spidy bring them in.

Please feel free to post (or PM) me additional lists as you generate them. Takes me almost no time to import the candidates for distribution, and we'd might as well get them done now so the DC/TC'ing can start in earnest.

Edit: Oh, and don't worry about duplication -- Spidy is smart enough to know it already holds a candidate, and will not assign for TF'ing anything still owned by another worker assigned through Primenet.

Madpoo 2015-07-23 15:36

1 Attachment(s)
[QUOTE=chalsall;406359]And I notice that these weren't in your original list; I've just had Spidy bring them in.

Please feel free to post (or PM) me additional lists as you generate them. Takes me almost no time to import the candidates for distribution, and we'd might as well get them done now so the DC/TC'ing can start in earnest.

Edit: Oh, and don't worry about duplication -- Spidy is smart enough to know it already holds a candidate, and will not assign for TF'ing anything still owned by another worker assigned through Primenet.[/QUOTE]

Cool (about the dupe checking on your side).

Yeah, the original exponents I listed had only been checked once. The examples I just mentioned had been checked twice, but both times by suspect machines.

What I probably should do is just generate a list of all exponents and their current TF level where they've had 2 checks already with no matches (under 58M). I imagine in those cases there are going to be some that could use an extra bit or two of TF before someone does the triple check.

And by that I mean *all* those needing triple-checks, not just the exponents where one or two of the tests were done by historically flaky systems.

Going by "somewhat flaky" as a metric, there are around 1,050 of them, but there are currently a total 5,428 unassigned exponents below 58M that have been checked twice with no match.

Which list would you prefer? I'm attaching the full list... I sorted by the current TF bit level and I imagine you'd only have to test half of them to an extra bit or two?

Madpoo 2015-07-23 15:39

In case anyone was confused about the need to also strategically check exponents that have been tested twice with no match...

I checked in 5 exponents this morning... 1 match and it's now verified. 1 mismatch of an exponent already tested once.

And then these 3 beauties that didn't match the first 2 runs (by the same user for first/second check):
[URL="http://www.mersenne.org/report_exponent/?exp_lo=55570813&full=1"]M55570813[/URL]
[URL="http://www.mersenne.org/report_exponent/?exp_lo=55864219&full=1"]M55864219[/URL]
[URL="http://www.mersenne.org/report_exponent/?exp_lo=56302951&full=1"]M56302951[/URL]

chalsall 2015-07-23 16:07

[QUOTE=Madpoo;406364]Which list would you prefer? I'm attaching the full list... I sorted by the current TF bit level and I imagine you'd only have to test half of them to an extra bit or two?[/QUOTE]

OK, I'm bringing in everything now which isn't already "owned", and which need additional TF'ing.

Might as well get these "off our books", and people seem to really enjoy these little specialized sub-projects.

petrw1 2015-07-23 18:50

1 of my own that did not make the list....
 
[url]http://www.mersenne.org/report_exponent/?exp_lo=35555123&full=1[/url]

Madpoo 2015-07-23 22:04

[QUOTE=petrw1;406376][url]http://www.mersenne.org/report_exponent/?exp_lo=35555123&full=1[/url][/QUOTE]

Yeah, it's assigned to someone so I didn't include it. Through all of this I'm going to try and avoid the assigned ones.

I'll check periodically and see if assignments expire so we can pick them up and assign them to our task force (or whatever we'd call ourselves). :smile:

frmky 2015-07-23 22:39

I just submitted the first two results of the eight I reserved. Both matched the previous runs. :smile:

[url]http://www.mersenne.org/report_exponent/?exp_lo=34654643&full=1[/url]
[url]http://www.mersenne.org/report_exponent/?exp_lo=34682537&full=1[/url]

Madpoo 2015-07-24 03:35

[QUOTE=frmky;406384]I just submitted the first two results of the eight I reserved. Both matched the previous runs. :smile:

[url]http://www.mersenne.org/report_exponent/?exp_lo=34654643&full=1[/url]
[url]http://www.mersenne.org/report_exponent/?exp_lo=34682537&full=1[/url][/QUOTE]

Cool. I didn't know anyone was going to start double-checking any, but that's cool.

The list I put up for extra TF work was using looser rules for figuring out if work was done by a bad machine or not. Once all the extra TF'ing is done, I'll generate a list of work done where it's far more likely the first residue will be bad.

I don't know what it is, but it's kind of fun and satisfying to do a test and realize your result (which is hopefully the correct one) is different from whomever ran it earlier. I had 6-7 more today. :smile:

chalsall 2015-07-24 15:04

New report, a "hmmm..."
 
Just for those "playing", I've made a quick [URL="https://www.gpu72.com/reports/available/sdc/"]report showing the Strategic Double Check TF'ing[/URL] status. This isn't linked anywhere from the site as this is temporary and of special interest only.

Separately, Mark [URL="http://www.mersenne.org/report_exponent/default.php?exp_lo=57909959&full=1"]found a factor a little while ago[/URL], and I noticed that as soon as it was factored the status changed to show the full residue marked as "Verified (Factored)".

Since this whole exercise is to find bad machines, perhaps showing the full residue isn't a good idea, since there's a chance the residue isn't in fact correct.

Thoughts?

Mark Rose 2015-07-24 15:29

[QUOTE=chalsall;406403]Just for those "playing", I've made a quick [URL="https://www.gpu72.com/reports/available/sdc/"]report showing the Strategic Double Check TF'ing[/URL] status. This isn't linked anywhere from the site as this is temporary and of special interest only.[/quote]

Awesome! I was wondering if the 70-bit 50-52M was assigned or not. I'll scoop those.

Edit: Or not. What Makes Sense seems to ignore the exponent range. No big deal :)

[quote]
Separately, Mark [URL="http://www.mersenne.org/report_exponent/default.php?exp_lo=57909959&full=1"]found a factor a little while ago[/URL], and I noticed that as soon as it was factored the status changed to show the full residue marked as "Verified (Factored)".

Since this whole exercise is to find bad machines, perhaps showing the full residue isn't a good idea, since there's a chance the residue isn't in fact correct.

Thoughts?[/QUOTE]

I agree.

I also found a factor for [url=http://www.mersenne.org/report_exponent/?exp_lo=47549387&full=1]M47549387[/url], which is in the list.

chalsall 2015-07-24 15:53

[QUOTE=Mark Rose;406404]Edit: Or not. What Makes Sense seems to ignore the exponent range. No big deal :)[/QUOTE]

Ah... Thanks for pointing that out. A legacy conditional. Please try again.

Madpoo 2015-07-24 16:40

[QUOTE=chalsall;406403]Just for those "playing", I've made a quick [URL="https://www.gpu72.com/reports/available/sdc/"]report showing the Strategic Double Check TF'ing[/URL] status. This isn't linked anywhere from the site as this is temporary and of special interest only.

Separately, Mark [URL="http://www.mersenne.org/report_exponent/default.php?exp_lo=57909959&full=1"]found a factor a little while ago[/URL], and I noticed that as soon as it was factored the status changed to show the full residue marked as "Verified (Factored)".

Since this whole exercise is to find bad machines, perhaps showing the full residue isn't a good idea, since there's a chance the residue isn't in fact correct.

Thoughts?[/QUOTE]

"Factored" status on LL tests is tricky... it's kind of a no-go status in terms of determining computer quality. We may never know in those cases.

Sometimes people go back and do extra factoring on an exponent that was already verified. I guess I could look for any cases where there was a reject residue that should be included in the "known bad" column.

For the factors found when it was still unverified, yeah, just no way of knowing. I could at least look at their error code and see if it was originally marked "suspect".

For now I'm not even bothering to include the "factor found later" runs in my tallies.

EDIT: By the way, I'm going to employ a little bit of a cheat when counting how many bad results a computer has done.

In the cases where I've done a double (or triple+) check and there's still no match, I'm going to claim technical superiority and say my result is correct even though it's unverified. After all, I'm running systems with ECC at stock speeds in climate controlled datacenters. I have a mere 3 results that were bad, and in all 3 I'm inclined to chalk them up to some kind of Prime95 issue. If you look at them you'll see why (the first few bits of the residue are zero, but the rest match). Plus, they were all part of my manic "triple check everything below 2M" thing:
[URL="http://www.mersenne.org/M8291"]M8291[/URL]
[URL="http://www.mersenne.org/M12281"]M12281[/URL]
[URL="http://www.mersenne.org/M801883"]M801883[/URL]

I think that gives me solid footing to claim I'm right, they're wrong, etc. :smile: I wonder if I could expand that to do the same for other mismatches, and if one is from a very solidly reliable computer, assume that's right and the other is bad, for purposes of this project.

Mark Rose 2015-07-24 17:48

[QUOTE=chalsall;406406]Ah... Thanks for pointing that out. A legacy conditional. Please try again.[/QUOTE]

Didn't work. I asked for 96 assignments to 72, 50M to 53M, WMS, and it gave me back a bunch of 41M assignments. I tried the same with to 71, and got a bunch of 43M assignments.

I'm giving myself the sub-sub-sub project of taking the sdc exponents above 50M to 72, then to 73.

chalsall 2015-07-24 17:52

[QUOTE=Mark Rose;406411]Didn't work. I asked for 96 assignments to 72, 50M to 53M, WMS, and it gave me back a bunch of 41M assignments. I tried the same with to 71, and got a bunch of 43M assignments.[/QUOTE]

Weird. Yes, I saw (in the logs) that the code is changing your 50M to be 40M.

Could you please give it one more go? I've removed every case where the "Low" is modified to be 40M.

Mark Rose 2015-07-24 18:29

[QUOTE=chalsall;406412]Weird. Yes, I saw (in the logs) that the code is changing your 50M to be 40M.

Could you please give it one more go? I've removed every case where the "Low" is modified to be 40M.[/QUOTE]

Same result. Oh well, it's not like there's a huge amount of TF'ing to be done. I'll just do it in the order GPU72 currently wants to give it :)

chalsall 2015-07-24 18:48

[QUOTE=Mark Rose;406413]Same result. Oh well, it's not like there's a huge amount of TF'ing to be done. I'll just do it in the order GPU72 currently wants to give it :)[/QUOTE]

Grrrr... I hate Newton's Method of software debugging....

There was yet another legacy conditional, which this time brought your low down to 35M. This has been removed.

Sorry about this; the next time you try it /should/ work as requested.

Mark Rose 2015-07-24 19:05

[QUOTE=chalsall;406414]Grrrr... I hate Newton's Method of software debugging....

There was yet another legacy conditional, which this time brought your low down to 35M. This has been removed.

Sorry about this; the next time you try it /should/ work as requested.[/QUOTE]

Qapla'!

Thanks for the bug fixes :)

Madpoo 2015-07-25 00:12

Let's get this party started
 
Okay, let's do some strategic double checking. :smile:

To start out here I've generated a list of exponents that are pretty likely to wind up with one of our double checks NOT matching the first one. The criteria here is:
Bad > 6 and Bad > Good*6 and Unknown > 0

Essentially meaning they have at least one unknown (so there'd actually be something to check), and they either have 6 times as many bad as good, or if good happens to be zero they have 6+ bad ones.

The list includes the exponent, and then for each exponent there's a worktodo entry plus some stats on the computer that did the first time check: # of bad, # of good, # of unknown, and # of suspect

I have a hunch that in these cases, if they have a bunch of proven bad results, the suspect results are probably also bad but they're waiting on triple checks to find out for sure. I don't factor them into the calculation, but if you're eyeballing some likely suspects then feel free to use that # as a guide as well.

These are all currently unassigned, so I'd say you can try what George mentioned and simply add it to your worktodo and I guess the server will create an assignment for it when it phones home.

I'm starting out small with these 16 just to make sure the process works (plus there's a good # assigned out to GPU72 for carrying them forward the few extra bits).

Oh, and before picking any of these up, look at the exponent report page and see if it's already assigned to someone.
The URL for that is simply:
[COLOR="Red"]www.mersenne.org/M<exponent>[/COLOR]
and look in the assignment section for any active ones.

Be sure to report results back here for feedback... like did your residue match, or did you get a different (and hopefully correct) one? That'll guide us going forward.
[CODE]exponent Bad Good Unk Sus worktodo
35064059 10 1 5 0 DoubleCheck=35064059,71,1
36497473 14 2 3 2 DoubleCheck=36497473,71,1
36517909 19 3 4 0 DoubleCheck=36517909,71,1
36572957 19 3 4 0 DoubleCheck=36572957,71,1
36702241 9 1 1 10 DoubleCheck=36702241,71,1
36779077 7 1 4 1 DoubleCheck=36779077,71,1
37053197 14 2 3 2 DoubleCheck=37053197,71,1
37303081 9 1 4 1 DoubleCheck=37303081,71,1
37521863 16 1 4 1 DoubleCheck=37521863,71,1
40564793 9 1 1 1 DoubleCheck=40564793,72,1
41216807 13 2 4 8 DoubleCheck=41216807,72,1
41350963 9 1 4 1 DoubleCheck=41350963,71,1
41636321 13 1 3 6 DoubleCheck=41636321,72,1
42159511 7 1 4 1 DoubleCheck=42159511,71,1
43207327 7 1 3 4 DoubleCheck=43207327,72,1
49520501 15 2 30 14 DoubleCheck=49520501,72,1[/CODE]

frmky 2015-07-25 00:14

The second pair of my original 8 finished. One matched and one didn't.
[url]http://www.mersenne.org/report_exponent/?exp_lo=34665307&full=1[/url]
[url]http://www.mersenne.org/report_exponent/?exp_lo=34683613&full=1[/url]

Edit: I grabbed the first pair from the new list and added them to the top of the queue. I should know tomorrow if they match.

frmky 2015-07-25 00:26

[QUOTE=Madpoo;406408]"Factored" status on LL tests is tricky... it's kind of a no-go status in terms of determining computer quality. We may never know in those cases.
[/QUOTE]
This one looks funny. Two different residues as "Verified (Factored)." I guess it simply means "verified" as composite. :smile:
[url]http://www.mersenne.org/report_exponent/?exp_lo=41277389&full=1[/url]

Madpoo 2015-07-25 03:21

[QUOTE=frmky;406431]This one looks funny. Two different residues as "Verified (Factored)." I guess it simply means "verified" as composite. [/QUOTE]

You guessed right. "Verified" in that case simply means that it is indeed composite, no matter that the LL tests didn't match.

The only way I know the first one is actually "suspect" is because I can look in the DB and see that it's error code is non-zero "00000100"

In a sense it's kind of a bummer not to have more info on the reliability of the LL testers in these cases... it would help in some ways. But in the broader sense it's about proving things composite or not and a factor is a good way of doing just that. :smile:

Madpoo 2015-07-25 06:56

For my part, I just checked out a big chunk of ~400 exponents that need triple-checks. That includes all of them below 35M and a few above 35M as well.

That should keep my systems churning through the end of August. The nice thing with those is that I know one way or another I'll identify which result was bad, to better predict which single-checked exponents would benefit from these strategic DCs.

And speaking of those DC's, I saw that some from the list I put up earlier have been assigned (the first two), so that's good. I'll check again and when they're all gone I'll dig up some more.

chalsall 2015-07-25 12:58

[QUOTE=Madpoo;406428]To start out here I've generated a list of exponents that are pretty likely to wind up with one of our double checks NOT matching the first one.[/QUOTE]

OK, I've taken the last 14 candidates in the list (read: all remaining); registered with Primenet. Will report back once completed (~ the 29th for the largest candidate on a relatively slow machine).

Better generate another list Aaron! :smile:

chalsall 2015-07-25 13:01

[QUOTE=Madpoo;406440]You guessed right. "Verified" in that case simply means that it is indeed composite, no matter that the LL tests didn't match.[/QUOTE]

And this was my earlier point. Yes, the candidate has been verified as being composite by being factored, but the residue(s) has NOT been verified. I again suggest that the second two hexs be masked in cases where two matching residues don't exist.

lycorn 2015-07-25 13:21

Very funny...

I wonder what the status of [URL="http://www.mersenne.org/report_exponent/?exp_lo=36807349&full=1"]http://www.mersenne.org/report_exponent/?exp_lo=36807349&full=1[/URL] was before madpoo checked in his result! Was Ian CoburnĀ“s result considered the bad one? Or... something is hiding in the masked bits of the 2 apparently matching residues?
Interesting, by any measure.

Madpoo 2015-07-25 17:42

[QUOTE=lycorn;406466]Very funny...

I wonder what the status of [URL="http://www.mersenne.org/report_exponent/?exp_lo=36807349&full=1"]http://www.mersenne.org/report_exponent/?exp_lo=36807349&full=1[/URL] was before madpoo checked in his result! Was Ian CoburnĀ“s result considered the bad one? Or... something is hiding in the masked bits of the 2 apparently matching residues?
Interesting, by any measure.[/QUOTE]

There are a few oddball cases in the database where the LL results section has 2 "matching" residues (and I use the term lightly), but the shift counts are identical.

Most of the time the duplicates are hidden (because they were from the same account submitting the result twice, under the old v4 server). In this case the same residue came from two different v4 users.

The server is smart enough to know they're the same and won't count it as verified, but they still show up in there.

There are fortunately only 88 of these still unverified. For example:
[URL="http://www.mersenne.org/M35079829"]M35079829[/URL]

If at least one of the submissions is from a version 5 user, the dupes are merged in the display, for example:
[URL="http://www.mersenne.org/M37074337"]M37074337[/URL]

You still see the 2 same residues in the history section, but only one in the LL section.

I've talked to George about going through and simply clearing out these duplicates. I think most (all?) of them were probably from manual submissions that George may have added in based on users emailing their results directly.

I went through a little bit ago and ran my own tests on any where the same user had submitted it more than twice, so those are done, and a few of the "only" twice.

lycorn 2015-07-25 18:10

OK. Makes sense. Thx.

science_man_88 2015-07-25 18:19

[QUOTE=Madpoo;406440]You guessed right. "Verified" in that case simply means that it is indeed composite, no matter that the LL tests didn't match.

The only way I know the first one is actually "suspect" is because I can look in the DB and see that it's error code is non-zero "00000100"[/QUOTE]

in theory you can use an LL test with the known factor in place of the mersenne it's a factor of and the full residue mod that factor in place of the end residue and check that it works out it should if the factor isn't fake and the full residue is correct.

frmky 2015-07-25 18:24

[QUOTE=frmky;406429]I grabbed the first pair from the new list and added them to the top of the queue. I should know tomorrow if they match.[/QUOTE]
Neither matched. :smile:

[url]http://www.mersenne.org/report_exponent/?exp_lo=35064059&full=1[/url]
[url]http://www.mersenne.org/report_exponent/?exp_lo=36497473&full=1[/url]

Mark Rose 2015-07-25 21:18

.

Madpoo 2015-07-25 21:56

[QUOTE=frmky;406476]Neither matched. :smile:

[url]http://www.mersenne.org/report_exponent/?exp_lo=35064059&full=1[/url]
[url]http://www.mersenne.org/report_exponent/?exp_lo=36497473&full=1[/url][/QUOTE]

Isn't that kind of fun, in a weird way? :smile:

Be sure and pick on James for his test on M36497473 ...

(of course there's always that chance *yours* is bad... :devil: )

Madpoo 2015-07-26 05:19

[QUOTE=chalsall;406464]Better generate another list[/QUOTE]

Okay, here we go.

For this list, I used the criteria of unassigned exponents below 58M (as usual) where:
- The computer has zero good results
- 3 or more bad results

There are 47 in the list. Some of the computers involved have a large # (14 or 28) unknowns... once those start getting double/triple checked it'll change their good/bad count. But those same computers also have a couple suspect results as well, so I'm still not convinced they're all that reliable until more data is collected.

[CODE]exponent Bad Good Unk Sus worktodo
34938793 3 0 2 0 DoubleCheck=34938793,71,1
35006537 3 0 1 1 DoubleCheck=35006537,71,1
35090593 3 0 4 3 DoubleCheck=35090593,71,1
35159191 3 0 3 0 DoubleCheck=35159191,71,1
35379403 3 0 3 0 DoubleCheck=35379403,71,1
35662219 5 0 2 5 DoubleCheck=35662219,71,1
35863103 3 0 2 3 DoubleCheck=35863103,71,1
35963243 5 0 5 0 DoubleCheck=35963243,71,1
36009163 3 0 5 2 DoubleCheck=36009163,71,1
36026687 4 0 4 8 DoubleCheck=36026687,71,1
36086483 3 0 2 1 DoubleCheck=36086483,71,1
36138559 5 0 8 5 DoubleCheck=36138559,71,1
36186791 5 0 5 0 DoubleCheck=36186791,71,1
36273859 3 0 1 0 DoubleCheck=36273859,71,1
36381031 4 0 1 2 DoubleCheck=36381031,71,1
36392009 5 0 4 7 DoubleCheck=36392009,71,1
36435929 5 0 6 3 DoubleCheck=36435929,71,1
36503917 3 0 5 2 DoubleCheck=36503917,71,1
36532913 3 0 1 1 DoubleCheck=36532913,71,1
36536707 4 0 1 0 DoubleCheck=36536707,71,1
36576769 5 0 4 7 DoubleCheck=36576769,71,1
36648539 3 0 5 2 DoubleCheck=36648539,71,1
36670561 5 0 4 7 DoubleCheck=36670561,71,1
36688991 4 0 4 8 DoubleCheck=36688991,71,1
36743743 3 0 3 1 DoubleCheck=36743743,71,1
36892939 4 0 3 2 DoubleCheck=36892939,71,1
36904081 5 0 8 5 DoubleCheck=36904081,71,1
36938339 5 0 6 3 DoubleCheck=36938339,71,1
36955621 3 0 2 1 DoubleCheck=36955621,71,1
37064927 4 0 1 0 DoubleCheck=37064927,71,1
37260571 4 0 4 8 DoubleCheck=37260571,71,1
37359779 3 0 4 3 DoubleCheck=37359779,71,1
40922261 3 0 7 1 DoubleCheck=40922261,72,1
41294441 3 0 2 0 DoubleCheck=41294441,72,1
41364503 4 0 14 2 DoubleCheck=41364503,72,1
41364541 4 0 14 2 DoubleCheck=41364541,72,1
41442407 4 0 14 2 DoubleCheck=41442407,72,1
41485517 4 0 3 2 DoubleCheck=41485517,71,1
41501191 3 0 1 2 DoubleCheck=41501191,71,1
41979727 4 0 28 1 DoubleCheck=41979727,72,1
41992207 3 0 7 1 DoubleCheck=41992207,72,1
42187843 4 0 28 1 DoubleCheck=42187843,72,1
43671757 4 0 28 1 DoubleCheck=43671757,72,1
44876599 3 0 1 1 DoubleCheck=44876599,72,1
55583947 4 0 6 0 DoubleCheck=55583947,73,1
57783881 4 0 6 0 DoubleCheck=57783881,73,1
57859121 4 0 6 0 DoubleCheck=57859121,73,1[/CODE]

Madpoo 2015-07-26 05:25

And here's a shorter list with:
- Good results between 1 and 10
- At least 5 times as many bad as good

As you can see, some of these also have a high # of suspect results which is a pretty good sign (combined with their known bad) that those will turn out bad as well.

Enjoy!
[CODE]exponent Bad Good Unk Sus worktodo
34769731 6 1 2 9 DoubleCheck=34769731,71,1
35200849 6 1 6 0 DoubleCheck=35200849,71,1
35997649 6 1 6 0 DoubleCheck=35997649,71,1
36278761 6 1 6 0 DoubleCheck=36278761,71,1
36589351 6 1 1 0 DoubleCheck=36589351,71,1
36717713 11 2 1 1 DoubleCheck=36717713,71,1
45951173 11 2 6 10 DoubleCheck=45951173,72,1[/CODE]

chalsall 2015-07-26 14:30

[QUOTE=Madpoo;406503]There are 47 in the list.[/QUOTE]

I've grabbed the first 24 (34938793 to 36688991 inclusive) from this list. Results should be in late Tuesday / early Wednesday.

And you're correct, this is fun! A bit like weeding a garden! :smile:

Oh, and three are back from my previous batch; 0 for 3. Will provide the full list once they're all completed.

Madpoo 2015-07-26 17:18

[QUOTE=chalsall;406532]And you're correct, this is fun! A bit like weeding a garden! :smile:[/QUOTE]

Or popping zits... LOL

I just found 2 more today, like this doozy that will now need a quintuple check:
[URL="http://www.mersenne.org/M35759939"]M35759939[/URL]

Although technically that one was from a batch I'd picked where exponents needed a 3rd/4th check, and wasn't from the lists of possibly suspect work... but still satisfying somehow. :smile:

Madpoo 2015-07-26 17:27

Verifying a residue if a factor is known?
 
Hey all... flagrantflowers mentioned to me that there's some kind of method for verifying the LL residue if we happen to know a factor. He suggested reaching out to science_man_88 about it, but before I bug him I thought I'd group think this thing.

If this were possible, it'd be nice to go through all of the cases where someone did an LL test and it was later factored... might be nice to split the singular "factored" status of such an LL test into "factored - verified" and "factored - bad".

Basically treating the factor as a confirming or disproving 2nd LL test.

Before that even took place I could at least go through and find all the cases where it was already verified by a 2nd LL test. Just in case the residue confirmation math is less than trivial, saving some time.

Anyone have any thoughts on that, or should I just check with science_man_88 as suggested?

Mark Rose 2015-07-26 18:17

[QUOTE=Madpoo;406547]I just found 2 more today, like this doozy that will now need a quintuple check:
[URL="http://www.mersenne.org/M35759939"]M35759939[/URL]
[/QUOTE]

I should have it done in a few days.

science_man_88 2015-07-26 18:17

[QUOTE=Madpoo;406549]Hey all... flagrantflowers mentioned to me that there's some kind of method for verifying the LL residue if we happen to know a factor. He suggested reaching out to science_man_88 about it, but before I bug him I thought I'd group think this thing.

If this were possible, it'd be nice to go through all of the cases where someone did an LL test and it was later factored... might be nice to split the singular "factored" status of such an LL test into "factored - verified" and "factored - bad".

Basically treating the factor as a confirming or disproving 2nd LL test.

Before that even took place I could at least go through and find all the cases where it was already verified by a 2nd LL test. Just in case the residue confirmation math is less than trivial, saving some time.

Anyone have any thoughts on that, or should I just check with science_man_88 as suggested?[/QUOTE]

see my post 43 of this thread and I'm not the only one to suggest this is possible there have been other threads talking about this. here's the easiest example in PARI:

[CODE]f=23;s=4;b=1736;for(x=3,11,s=Mod(s,f)^2-2);if(lift(s)==lift(Mod(b,f)),print("factor found:"f))[/CODE]

basically all this is using is the fact that assuming we have a number y of form ax+b (that is b mod x) assume we know a factor of x, call this factor f we can rewrite this as y=azf+(b\f)*f +b%f where \ between two numbers indicates the floor of integer division and % means modular arithmetic,this can be simplified by saying all multiples of f that are present sum to c*f we then get y=c*f+b%f and so b%f is our residue mod f of our original number if f is a factor of anything but b in the original equation y=ax+b. of course this isn't only potentially useful to use factors to check residues but if we know the residues are correct we can then use the residues from numbers higher than x ( in this case out mersenne number to be factored), and all residues for numbers greater than the mersenne number have to match their residue mod f for f to be a factor of the mersenne number x.

Madpoo 2015-07-26 19:27

[QUOTE=science_man_88;406555]here's the easiest example in PARI:

[CODE]f=23;s=4;b=1736;for(x=3,11,s=Mod(s,f)^2-2);if(lift(s)==lift(Mod(b,f)),print("factor found:"f))[/CODE]

basically all this is using is the fact that assuming we have a number y of form ax+b (that is b mod x) assume we know a factor of x, call this factor f we can rewrite this as y=azf+(b\f)*f +b%f where \ between two numbers indicates the floor of integer division and % means modular arithmetic,this can be simplified by saying all multiples of f that are present sum to c*f we then get y=c*f+b%f and so b%f is our residue mod f of our original number if f is a factor of anything but b in the original equation y=ax+b. of course this isn't only potentially useful to use factors to check residues but if we know the residues are correct we can then use the residues from numbers higher than x ( in this case out mersenne number to be factored), and all residues for numbers greater than the mersenne number have to match their residue mod f for f to be a factor of the mersenne number x.[/QUOTE]

So, pretend I don't know the math very well (okay, enough laughing...)

For the PARI syntax, I followed along with M11 having the factor 23, but what is b= in that example? That's not the residue for M11 (it's 0x48), and that's where I lost ya. :smile:

Let's say we have this as an example: [URL="http://www.mersenne.org/M2000143"]http://www.mersenne.org/M2000143[/URL]:
M2000143 has the factor 6174103888966755151 and a verified residue of 0x7AFB854059F08BEF (well, technically the last 64 bits of it, but you know what I mean).

How would that look if I wanted to use the factor to verify the LL test/residue?

science_man_88 2015-07-26 19:32

[QUOTE=Madpoo;406565]So, pretend I don't know the math very well (okay, enough laughing...)

For the PARI syntax, I followed along with M11 having the factor 23, but what is b= in that example? That's not the residue for M11 (it's 0x48), and that's where I lost ya. :smile:

Let's say we have this as an example: [URL="http://www.mersenne.org/M2000143"]http://www.mersenne.org/M2000143[/URL]:
M2000143 has the factor 6174103888966755151 and a verified residue of 0x7AFB854059F08BEF (well, technically the last 64 bits of it, but you know what I mean).

How would that look if I wanted to use the factor to verify the LL test/residue?[/QUOTE]

b is the decimal value of the final residue mod 2^11-1 see [url]http://oeis.org/A129220[/url]

plug the factor in as f, the 11 gets replaced by the exponent in question, and b gets replaced by the decimal value of the full final residue. PARI/gp can still be quite slow another language might be better to use.

frmky 2015-07-26 21:00

[QUOTE=Madpoo;406565]and a verified residue of 0x7AFB854059F08BEF (well, technically the last 64 bits of it, but you know what I mean).[/QUOTE]
There lies the problem. You only have the residue mod 2^64. Therefore after performing the loop, you know s = (res + k * 2^64) mod f for some unknown k. The problem is that you can always find a value of k that works for any residue. For your M2000143 example, the decimal residue following the loop using f is 1838775420032325409. From this and the verified residue, you can find k = 815412537616906440. For the invalid residue 0x6AFB854059F08BEF, you can find k = 1201294030677328637.

frmky 2015-07-26 21:20

I'll take all of seven on the second list, 34769731 to 45951173 inclusive.

Madpoo 2015-07-26 22:18

[QUOTE=frmky;406573]There lies the problem. You only have the residue mod 2^64. Therefore after performing the loop, you know s = (res + k * 2^64) mod f for some unknown k. The problem is that you can always find a value of k that works for any residue. For your M2000143 example, the decimal residue following the loop using f is 1838775420032325409. From this and the verified residue, you can find k = 815412537616906440. For the invalid residue 0x6AFB854059F08BEF, you can find k = 1201294030677328637.[/QUOTE]

Well, I guess I'm still confused on how exactly I could verify the residue given the factor, since all we have are the last 64 bits of the residue.

I won't worry about it terribly though.

There's around 26,000-27,000 exponents where the residue is verified, but has a factor. Another nearly 18K exponents where a factor was found after only 1 LL test was done.

I thought it'd be cool to figure out, for those 18K tests, whether the residue was good or not. It'd be a pretty impressive and large set of data to figure out the reliability of the LL tester.

Madpoo 2015-07-26 22:44

[QUOTE=Madpoo;406578]I thought it'd be cool to figure out, for those 18K tests, whether the residue was good or not. It'd be a pretty impressive and large set of data to figure out the reliability of the LL tester.[/QUOTE]

At the very least, I've found 104 instances where a verifying LL test exists for an exponent, but there is also a bad/mismatched residue from some yahoo. :smile: That's 104 more "bad" results I can glean out of the data right away.

Plus the tens of thousands of "good" results I can pull out, where those verifying residues exist but it was factored.

Yeah, I think that kind of info is useful, and maybe I can get George to approve adding new "result types" instead of the single "factored" code that exists now. Like:
Factored - Unverified
Factored - Verified
Factored - Bad

I don't know if we'd need a "Factored - Suspect" to indicate if the residue was originally marked "suspect" or not. Don't know if that matters as much.

science_man_88 2015-07-26 22:54

[QUOTE=frmky;406573]There lies the problem. You only have the residue mod 2^64. Therefore after performing the loop, you know s = (res + k * 2^64) mod f for some unknown k. The problem is that you can always find a value of k that works for any residue. For your M2000143 example, the decimal residue following the loop using f is 1838775420032325409. From this and the verified residue, you can find k = 815412537616906440. For the invalid residue 0x6AFB854059F08BEF, you can find k = 1201294030677328637.[/QUOTE]

does this change fail the test ?

[QUOTE](19:52) gp > f=6174103888966755151;s=4;b=[SPOILER]8861823203214920687[/SPOILER];for(x=3,2000143,s=Mod(Mod(s,f)^2-2,1<<64));if(lift(s)==lift(Mod(Mod(b,f),1<<64)),print("factor found:"f))
factor found:6174103888966755151[/QUOTE]

edit:never mind it says it finds the factor either way with the hex to decimal converter I used to convert it to decimal for testing.

frmky 2015-07-26 23:17

Huh?

[CODE](16:16) gp > f=6174103888966755151;s=4;b=8861823203214920687;for(x=3,2000143,s=Mod(Mod(s,f)^2-2,1<<64));if(lift(s)==lift(Mod(Mod(b,f),1<<64)),print("factor found:"f))
(16:16) gp > lift(s)
%8 = 1838775420032325409
(16:16) gp > lift(Mod(Mod(b,f),1<<64))
%9 = 2687719314248165536
[/CODE]

Edit: OK, you found the problem. :-)

chalsall 2015-07-27 00:21

[QUOTE=Madpoo;406503]There are 47 in the list.[/QUOTE]

I just took the last 23 in this list. As in, I now "own" all of the candidates from the list of 47. Better build another list Aaron.... :wink:

BTW, seven from my first batch of 14 from the original list are now completed. None matched. All but the highest two will be completed in about three hours.

BTW2... Aaron, two in the list of 47 were sub-optimally TF'ed. Anything between 40M and 50M should be TF'ed to 72. I'll TF them myself in parallel.

Madpoo 2015-07-27 00:30

[QUOTE=chalsall;406589]I just took the last 23 in this list. As in, I now "own" all of the candidates from the list of 47. Better build another list Aaron.... :wink:

BTW, seven from my first batch of 14 from the original list are now completed. None matched. All but the highest two will be completed in about three hours.

BTW2... Aaron, two in the list of 47 were sub-optimally TF'ed. Anything between 40M and 50M should be TF'ed to 72. I'll TF them myself in parallel.[/QUOTE]

Here's 4 more, using the same criteria as the last shorter list. They recently became available as assignments expired, or extra TF work completed:
[CODE]exponent Bad Good Unk Sus worktodo
37952297 19 3 5 5 DoubleCheck=37952297,71,1
41573579 17 2 6 3 DoubleCheck=41573579,72,1
41626679 11 2 6 10 DoubleCheck=41626679,72,1
41678927 17 2 6 3 DoubleCheck=41678927,72,1[/CODE]

I'll work on a larger list in a bit.

Mark Rose 2015-07-27 01:23

In less than a day, I'll begin finishing the trial factoring for about 90 50M+ exponents per day. Finishing the trial factoring of the 50M+ exponents will take me approximately two weeks. That should be good for daily lists for a while. :)

Madpoo 2015-07-27 01:29

[QUOTE=Madpoo;406590]I'll work on a larger list in a bit.[/QUOTE]

Here are some more:
[CODE]exponent Bad Good Unk Sus worktodo
34865477 9 2 15 8 DoubleCheck=34865477,71,1
34949749 9 2 15 8 DoubleCheck=34949749,71,1
34996139 5 1 3 0 DoubleCheck=34996139,71,1
35143627 5 1 10 6 DoubleCheck=35143627,71,1
35476471 5 1 10 6 DoubleCheck=35476471,71,1
35578973 9 2 15 8 DoubleCheck=35578973,71,1
36131197 9 2 15 8 DoubleCheck=36131197,71,1
36337391 9 2 15 9 DoubleCheck=36337391,71,1
36453931 5 1 10 6 DoubleCheck=36453931,71,1
36529453 5 1 2 2 DoubleCheck=36529453,71,1
36808613 5 1 10 6 DoubleCheck=36808613,71,1
36815837 9 2 15 8 DoubleCheck=36815837,71,1
37260259 5 1 9 2 DoubleCheck=37260259,71,1
38501137 5 0 6 3 DoubleCheck=38501137,72,1
40882241 9 2 15 9 DoubleCheck=40882241,72,1
41471879 9 2 15 8 DoubleCheck=41471879,71,1
41501183 5 0 5 0 DoubleCheck=41501183,72,1
41620973 11 2 1 8 DoubleCheck=41620973,72,1
42019207 5 1 9 2 DoubleCheck=42019207,72,1
42022873 9 2 15 9 DoubleCheck=42022873,72,1
42100753 39 9 13 18 DoubleCheck=42100753,72,1
42174677 9 2 15 8 DoubleCheck=42174677,71,1
42240043 19 4 5 4 DoubleCheck=42240043,72,1
42791519 9 2 15 8 DoubleCheck=42791519,71,1
43263431 9 2 15 8 DoubleCheck=43263431,70,1[/CODE]

Mark Rose 2015-07-27 01:50

[QUOTE=Madpoo;406595]Here are some more:
[CODE]exponent Bad Good Unk Sus worktodo
41471879 9 2 15 8 DoubleCheck=41471879,71,1
42174677 9 2 15 8 DoubleCheck=42174677,71,1
42791519 9 2 15 8 DoubleCheck=42791519,71,1
43263431 9 2 15 8 DoubleCheck=43263431,70,1[/CODE][/QUOTE]

These four need more trial factoring. I'll start immediately.

Edit: should be finished 60 minutes from now.

Prime95 2015-07-27 02:11

I'll do these

DoubleCheck=35578973,71,1
DoubleCheck=36131197,71,1
DoubleCheck=36337391,71,1
DoubleCheck=36529453,71,1
DoubleCheck=36808613,71,1
DoubleCheck=36815837,71,1
DoubleCheck=37260259,71,1

chalsall 2015-07-27 02:39

[QUOTE=chalsall;406589]BTW, seven from my first batch of 14 from the original list are now completed. None matched. All but the highest two will be completed in about three hours.[/QUOTE]

OK, twelve of the fourteen from the original list now complete. Only one match (41350963).

Mark Rose 2015-07-27 03:11

[QUOTE=Mark Rose;406597]These four need more trial factoring. I'll start immediately.

Edit: should be finished 60 minutes from now.[/QUOTE]

Done. No factors found.

frmky 2015-07-27 04:36

I'll grab these 3:

DoubleCheck=34865477,71,1
DoubleCheck=34949749,71,1
DoubleCheck=34996139,71,1

Madpoo 2015-07-27 14:33

Y'all are doing great helping out with these.

I wanted to share a funny quadruple check I just turned in (which matched one of the other 3, fortunately):
[URL="http://www.mersenne.org/M47316187"]M47316187[/URL]

Specifically, the user name on the 2nd failed attempt made me laugh. :smile:

Mark Rose 2015-07-27 15:31

[QUOTE=Madpoo;406634]Y'all are doing great helping out with these.

I wanted to share a funny quadruple check I just turned in (which matched one of the other 3, fortunately):
[URL="http://www.mersenne.org/M47316187"]M47316187[/URL]

Specifically, the user name on the 2nd failed attempt made me laugh. :smile:[/QUOTE]

That's funny :D

frmky 2015-07-27 23:41

I grabbed these 3:
DoubleCheck=35143627,71,1
DoubleCheck=35476471,71,1
DoubleCheck=36453931,71,1

chalsall 2015-07-28 02:51

OK, twelve of the batch of 47 have now completed: 34938793, 35090593, 35379403, 35863103, 36009163, 36086483, 36186791, 36381031, 36435929, 36532913, 36576769 and 36670561.

36009163 and 36186791 matched; the rest didn't.

Madpoo 2015-07-28 14:36

Another quintuple check needed
 
Here's another one I just checked in that will now need a quint check:
[URL="http://www.mersenne.org/M36498731"]M36498731[/URL]

If anyone is game for it, help yourselves, otherwise it can wait 'til the double-checkers get there naturally... I still feel pretty good about the stability of my systems. :smile:

It's still fun finding these mismatches when so many before me have failed. LOL

Madpoo 2015-07-28 15:13

[QUOTE=chalsall;406700]OK, twelve of the batch of 47 have now completed: 34938793, 35090593, 35379403, 35863103, 36009163, 36086483, 36186791, 36381031, 36435929, 36532913, 36576769 and 36670561.

36009163 and 36186791 matched; the rest didn't.[/QUOTE]

Wow, good catches.

So, in general it sounds like these strategic double checks are paying off, finding (probably) bad results months or even years ahead of when they would have been found normally.

Ignoring the possibility that someone will be upset if they think we're singling out their machines for extra scrutiny, I'll keep making lists, which will slowly improve in accuracy over time as we fit more work into the bad/good categories.

I'm not keeping track too much of the exponents themselves beyond grabbing whatever ones are currently unassigned, and that match whatever criteria scoops in enough to keep people busy working on these if they want... let me know if we should try and make this more organized in any way, but it probably doesn't need to be controlled to any great extent.

I'm working on a new list now. I realized there was a little problem in my query... nothing serious. Just that some Primenet v4 entries have a "null" as the legacy computer name, and I was actually keying on that being null or not to see if it was a v4 or v5 machine. I've modified it and as a result I may find additional legacy systems to include.

Unfortunately, if a v4 user had a "null" machine name there's no way to know if a bunch of bad entries were really from the same system or not, but I'll see how it goes and just ignore those if it's obviously giving me useless data.

Mark Rose 2015-07-28 15:51

[QUOTE=Madpoo;406729]Here's another one I just checked in that will now need a quint check:
[URL="http://www.mersenne.org/M36498731"]M36498731[/URL]
[/QUOTE]

Mine now.

My residue for [url="http://www.mersenne.org/M35759939"]M35759939[/url] also matched yours.

Madpoo 2015-07-28 18:01

[QUOTE=Madpoo;406731]I'm working on a new list now. I realized there was a little problem in my query... nothing serious. Just that some Primenet v4 entries have a "null" as the legacy computer name, and I was actually keying on that being null or not to see if it was a v4 or v5 machine. I've modified it and as a result I may find additional legacy systems to include.[/QUOTE]

Treating "null" legacy computer names appropriately didn't really add too much. Most of the entries like that had lots of good ones and didn't really have tons of bad stuff. There were maybe 5 add'l "bad" machines in the list accounting for a handful of exponents (all of which I think are currently assigned anyway).

Here's a list of 50 more though. Exponents where the machine has either (good=0 and bad >= 3) or (bad > good*3):

[CODE]exponent Bad Good Unk Sus worktodo
34680487 19 5 15 7 DoubleCheck=34680487,71,1
34852141 19 5 15 7 DoubleCheck=34852141,71,1
34968539 4 1 7 2 DoubleCheck=34968539,71,1
35068729 8 2 6 1 DoubleCheck=35068729,71,1
35186419 4 1 2 1 DoubleCheck=35186419,71,1
35653117 19 5 15 7 DoubleCheck=35653117,71,1
35792423 19 5 15 7 DoubleCheck=35792423,71,1
35938541 8 2 6 1 DoubleCheck=35938541,71,1
36568277 19 5 15 7 DoubleCheck=36568277,71,1
36771491 4 1 2 2 DoubleCheck=36771491,71,1
37034737 19 5 15 7 DoubleCheck=37034737,71,1
37272679 14 4 5 5 DoubleCheck=37272679,71,1
37478869 10 3 12 1 DoubleCheck=37478869,71,1
37532009 4 1 8 1 DoubleCheck=37532009,71,1
40882241 10 2 15 8 DoubleCheck=40882241,72,1
40904273 4 1 12 0 DoubleCheck=40904273,72,1
40930163 4 1 5 3 DoubleCheck=40930163,72,1
40944077 18 5 2 0 DoubleCheck=40944077,72,1
40981987 19 5 15 7 DoubleCheck=40981987,72,1
41352907 13 3 5 3 DoubleCheck=41352907,72,1
41397359 4 1 12 0 DoubleCheck=41397359,72,1
41501183 5 1 4 0 DoubleCheck=41501183,72,1
41542901 7 2 5 0 DoubleCheck=41542901,72,1
41572033 4 0 28 1 DoubleCheck=41572033,72,1
41573579 17 2 6 3 DoubleCheck=41573579,72,1
41585389 4 1 8 1 DoubleCheck=41585389,72,1
41618251 10 3 12 1 DoubleCheck=41618251,72,1
41620973 11 2 1 8 DoubleCheck=41620973,72,1
41626679 11 2 6 10 DoubleCheck=41626679,72,1
41678927 17 2 6 3 DoubleCheck=41678927,72,1
41713033 4 1 12 0 DoubleCheck=41713033,72,1
42005081 7 2 5 0 DoubleCheck=42005081,72,1
42019207 5 1 9 2 DoubleCheck=42019207,72,1
42022873 10 2 15 8 DoubleCheck=42022873,72,1
42100753 39 9 13 18 DoubleCheck=42100753,72,1
42222599 4 1 8 1 DoubleCheck=42222599,72,1
42237521 14 4 5 5 DoubleCheck=42237521,72,1
42240043 20 4 5 3 DoubleCheck=42240043,72,1
42279269 15 1 3 4 DoubleCheck=42279269,72,1
42324563 14 2 4 8 DoubleCheck=42324563,72,1
42376109 10 3 12 1 DoubleCheck=42376109,72,1
43428943 4 1 12 0 DoubleCheck=43428943,72,1
45923561 4 1 5 3 DoubleCheck=45923561,72,1
55306079 39 9 13 18 DoubleCheck=55306079,73,1
55306243 39 9 13 18 DoubleCheck=55306243,73,1
55392221 3 0 4 3 DoubleCheck=55392221,73,1
55397983 16 2 29 14 DoubleCheck=55397983,73,1
56341511 3 0 4 3 DoubleCheck=56341511,73,1
56761349 16 2 29 14 DoubleCheck=56761349,73,1
56769421 16 2 29 14 DoubleCheck=56769421,73,1[/CODE]

Mark Rose 2015-07-28 19:03

From Madpoo's list in post [url=http://www.mersenneforum.org/showpost.php?p=406364&postcount=17]#17[/url], all the exponents (56M,58M) are now fully trial factored:

DoubleCheck=[url=http://www.mersenne.org/M56066851]56066851[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56114683]56114683[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56123623]56123623[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56140543]56140543[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56152813]56152813[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56153221]56153221[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56153731]56153731[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56155243]56155243[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56174539]56174539[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56191063]56191063[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56198677]56198677[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56233799]56233799[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56234723]56234723[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56238131]56238131[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56251597]56251597[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56270477]56270477[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56291761]56291761[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56301781]56301781[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56302951]56302951[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56303309]56303309[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56316023]56316023[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56318287]56318287[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56330089]56330089[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56341039]56341039[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56341069]56341069[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56348293]56348293[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56348329]56348329[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56377691]56377691[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56382917]56382917[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56406829]56406829[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56412943]56412943[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56420057]56420057[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56420477]56420477[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56447749]56447749[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56468263]56468263[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56507819]56507819[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56523427]56523427[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56524151]56524151[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56539541]56539541[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56554453]56554453[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56555581]56555581[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56556187]56556187[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56556767]56556767[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56558023]56558023[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56561341]56561341[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56583407]56583407[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56617061]56617061[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56621053]56621053[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56621099]56621099[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56689639]56689639[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56703797]56703797[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56704411]56704411[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56714327]56714327[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56715653]56715653[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56752879]56752879[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56753273]56753273[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56800829]56800829[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56838473]56838473[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56852687]56852687[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56868491]56868491[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56870449]56870449[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56871487]56871487[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56926871]56926871[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56934463]56934463[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56953073]56953073[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56974039]56974039[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M56975939]56975939[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56985451]56985451[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56990413]56990413[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M56997691]56997691[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57000329]57000329[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57017123]57017123[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57032629]57032629[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57055759]57055759[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57058711]57058711[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57086539]57086539[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57086573]57086573[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57090923]57090923[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57090931]57090931[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57091109]57091109[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57110177]57110177[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57139189]57139189[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57160193]57160193[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57200413]57200413[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57206161]57206161[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57207373]57207373[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57208741]57208741[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57246751]57246751[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57248761]57248761[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57248981]57248981[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57266147]57266147[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57269221]57269221[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57270149]57270149[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57290173]57290173[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57293443]57293443[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57310541]57310541[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57320843]57320843[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57352369]57352369[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57363203]57363203[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57370591]57370591[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57379519]57379519[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57383791]57383791[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57412753]57412753[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57424993]57424993[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57462527]57462527[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57466907]57466907[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57472673]57472673[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57512177]57512177[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57526771]57526771[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57528319]57528319[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57538057]57538057[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57563417]57563417[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57578579]57578579[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57591283]57591283[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57612227]57612227[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57616457]57616457[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57647329]57647329[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57650933]57650933[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57686581]57686581[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57721933]57721933[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57752603]57752603[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57759731]57759731[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57773579]57773579[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57797717]57797717[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57798791]57798791[/url],75,1
DoubleCheck=[url=http://www.mersenne.org/M57798857]57798857[/url],75,1
DoubleCheck=[url=http://www.mersenne.org/M57815321]57815321[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57829861]57829861[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57846857]57846857[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57860441]57860441[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57865001]57865001[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57866573]57866573[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57869209]57869209[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57878021]57878021[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M57880943]57880943[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57890233]57890233[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57896317]57896317[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57901159]57901159[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57908723]57908723[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57910667]57910667[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57910703]57910703[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57917443]57917443[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57925729]57925729[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57927973]57927973[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57929611]57929611[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57936041]57936041[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57951667]57951667[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57957623]57957623[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57982069]57982069[/url],76,1
DoubleCheck=[url=http://www.mersenne.org/M57982157]57982157[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57989489]57989489[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M57995393]57995393[/url],73,1

There are five additional exponents in subsequent posts that are not in this list. Three of them are in the previous post just now.

I haven't checked to see if there was any additional trial factoring done outside this project, nor if the exponents are already assigned.

I'll post a new list every one to three days, depending on how quickly I finish the TF'ing above 50M.

Madpoo 2015-07-28 21:13

[QUOTE=Mark Rose;406745]From Madpoo's list in post [url=http://www.mersenneforum.org/showpost.php?p=406364&postcount=17]#17[/url], all the exponents (56M,58M) are now fully trial factored:[/QUOTE]

As a reminder (I forgot what these were too)...

These were from a list of exponents where they'd been done twice already and needed triple-checking.

One or the other of the results for these is bad, but unknown which one.

Testing these is tangential to targeting exponents where only one test has been done by a suspect machine, but in these cases with 2 tests so far, there's a good chance one of those is correct.

What I'd like to do is come up with a list of these "double-checked but still unverified" exponents and come up with a list of those where *both* computers that did the test are suspect. However I was having a little hiccup wrapping my head around how best to do that query so I've put it off for now. I'm grabbing all of these that I can and just doing a triple-check on all of them (I've pulled all of them below 35.2M I think)... it'd take me a while to get all of them below 58M done though. :) ~ 2500 of those below 58M. 3092 of them total (including > 58M)

By the way, not that we're keeping score or anything, but so far I now have 81 exponents where my result (presumably correct?) didn't match any of the previous runs. :smile: (some are assigned, and I think 58 are unassigned for a new test)

There are 3 needing quintuple checks, 14 needing quad checks, and then the rest needing a triple check.

These 2 are still unassigned and need a quint check if anyone is feeling bold:
36742943
46013437

That first one might not technically count... it was one of the odd ones where the same user submitted their result (same shift count) multiple times.

Mark Rose 2015-07-28 22:13

[QUOTE=Madpoo;406761]What I'd like to do is come up with a list of these "double-checked but still unverified" exponents and come up with a list of those where *both* computers that did the test are suspect. However I was having a little hiccup wrapping my head around how best to do that query so I've put it off for now.[/quote]

SELECT exponent FROM results WHERE <computer bad> GROUP BY exponent HAVING COUNT(exponent) > 1

Something like that oughta work :)

[quote]
I'm grabbing all of these that I can and just doing a triple-check on all of them (I've pulled all of them below 35.2M I think)... it'd take me a while to get all of them below 58M done though. :) ~ 2500 of those below 58M. 3092 of them total (including > 58M)
[/quote]

Don't you have like a thousand cores or something? Surely it couldn't take too long... ;)

[quote]
These 2 are still unassigned and need a quint check if anyone is feeling bold:
36742943
46013437
[/quote]

Mine now. Should be done in a few days. Stealing cycles from my SoB work lol

frmky 2015-07-28 22:25

The last four from my original reservation of eight are done. 34665871 matched while 34671661, 34686947, and 34691149 did not.

From the subsequent reservations, 4 are done. 34865477 matched while 34769731, 36589351, and 36717713 did not.

Madpoo 2015-07-28 22:28

[QUOTE=Mark Rose;406766]SELECT exponent FROM results WHERE <computer bad> GROUP BY exponent HAVING COUNT(exponent) > 1

Something like that oughta work :)[/QUOTE]

Yeah, my query has a lot of other stuff going on, but I guess if I use the same "bad" criteria that would work for the most part.

Doing that, with the same thresholds for bad machines, I get just one exponent, and it's already assigned. :(
38497531

With slightly looser thresholds I get this one, also assigned:
35414699

It's the same user who did both checks... I already picked up all the unassigned exponents matching those. I figure if a user didn't get it right the first time, how confident am I they'll be right the second time? :smile:

chalsall 2015-07-28 22:45

[QUOTE=Mark Rose;406745]I'll post a new list every one to three days, depending on how quickly I finish the TF'ing above 50M.[/QUOTE]

Gosh... If LaurV doesn't get back in the game soon you're quickly going to pass him in the stats! :wink:

Mark Rose 2015-07-28 23:43

[QUOTE=chalsall;406771]Gosh... If LaurV doesn't get back in the game soon you're quickly going to pass him in the stats! :wink:[/QUOTE]

Shh.... don't wake him...

chalsall 2015-07-29 12:25

OK, the last two from my first batch of 14 have completed. 43207327 matched, 49520501 didn't.

The next twelve from my second batch of 47 have also completed. 35662219, 35963243, 36273859, 36503917 and 36648539 matched. 35006537, 35159191, 36026687, 36138559, 36392009, 36536707 and 36688991 didn't.

Aaron... Should we keep posting the results here, or will your queries just take into account the new data to refine the statistics for the suspect machines?

chalsall 2015-07-29 14:07

[QUOTE=Madpoo;406739]Here's a list of 50 more though. Exponents where the machine has either (good=0 and bad >= 3) or (bad > good*3):[/QUOTE]

OK, I've just taken all of these.

Mark Rose 2015-07-29 14:12

The 55M exponents from post #17 are ready:

DoubleCheck=[url=http://www.mersenne.org/M55075387]55075387[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55091521]55091521[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55100509]55100509[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55110421]55110421[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55112119]55112119[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55131227]55131227[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55134139]55134139[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55137587]55137587[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55197383]55197383[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55203487]55203487[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55213993]55213993[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55235711]55235711[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55245227]55245227[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55249981]55249981[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55250353]55250353[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55277483]55277483[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55304789]55304789[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55380629]55380629[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55390507]55390507[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55404101]55404101[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55410991]55410991[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55467701]55467701[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55474831]55474831[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55477861]55477861[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55481191]55481191[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55510541]55510541[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55526041]55526041[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55531279]55531279[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55547719]55547719[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55551827]55551827[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55570813]55570813[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55584869]55584869[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55610531]55610531[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55625953]55625953[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55627493]55627493[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55638719]55638719[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55643017]55643017[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55659697]55659697[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55660051]55660051[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55675049]55675049[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55693711]55693711[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55696681]55696681[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55700563]55700563[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55741409]55741409[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55764517]55764517[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55780817]55780817[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55791137]55791137[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55799053]55799053[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55802897]55802897[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55815013]55815013[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55817737]55817737[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55817789]55817789[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55817869]55817869[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55818001]55818001[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55864219]55864219[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55875761]55875761[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55876199]55876199[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55889903]55889903[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55891183]55891183[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55895927]55895927[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55898959]55898959[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55900433]55900433[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55914011]55914011[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55922969]55922969[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55926331]55926331[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55975757]55975757[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55981307]55981307[/url],74,1
DoubleCheck=[url=http://www.mersenne.org/M55981963]55981963[/url],73,1
DoubleCheck=[url=http://www.mersenne.org/M55993829]55993829[/url],73,1

If there's no point in me posting these lists, let me know :)

Madpoo 2015-07-29 16:46

[QUOTE=chalsall;406800]OK, the last two from my first batch of 14 have completed. 43207327 matched, 49520501 didn't.

The next twelve from my second batch of 47 have also completed. 35662219, 35963243, 36273859, 36503917 and 36648539 matched. 35006537, 35159191, 36026687, 36138559, 36392009, 36536707 and 36688991 didn't.

Aaron... Should we keep posting the results here, or will your queries just take into account the new data to refine the statistics for the suspect machines?[/QUOTE]

I've employed a little "cheat" in my query to take into account whether the machine that did the other test is of good quality or not... if so I mark the mismatched residue as "bad" in my tally for that machine.

I'll look at your system(s) and see if they fall into my category of "awesome systems" or not. For now I'm including any CPU with >= 80 good results and <= 1 bad results.

--- computing ... ---

Okay, looks like your systems are in the awesome category, at least the one that checked in M36688991. 1190 good, zero bad, and 25 unknowns.

So yup, the alternate residues you checked in will tick the bad count of the other system up a notch. Well done. :smile:

About half of the systems I see on your account would qualify... the rest simply don't have enough good results yet but zero bad ones. Only one of your systems has a single bad result but 779 good ones, so it'd still be in the awesome category.

chalsall 2015-07-29 16:57

[QUOTE=Madpoo;406813]So yup, the alternate residues you checked in will tick the bad count of the other system up a notch.[/QUOTE]

OK, cool.

A quick thought... Are any of the "bad" machines still reserving and submitting work?

If so, might it be worth assigning them (say) DC Cat 2 work? High enough that they won't slow down any milestones, but low enough that any bad results will quickly be triple checked?

Obviously there would need to be (possibly lengthy, possibly heated) discussion as to what parameters would be used to make such a decision.

Thoughts?

Madpoo 2015-07-29 17:45

[QUOTE=chalsall;406815]OK, cool.

A quick thought... Are any of the "bad" machines still reserving and submitting work?

If so, might it be worth assigning them (say) DC Cat 2 work? High enough that they won't slow down any milestones, but low enough that any bad results will quickly be triple checked?

Obviously there would need to be (possibly lengthy, possibly heated) discussion as to what parameters would be used to make such a decision.

Thoughts?[/QUOTE]

Possibly... it could work. I think right now the assignment system is kind of clunky and slow because it does a lot of the same work to figure out stuff about the user/computer on each and every assignment, rather than all at once and then simply getting a list of XX exponents. That's why, for example, the manual assignment page is terribly slow if requesting more than a certain amount of work.

To take into consideration some kind of "this computer has xx bad, yy good results" and make a decision based on that as well, it's not terribly complicated per request, but again I'd like to see some improvements in the code so that calculation is made just once for all XX assignments being asked for, and not once for each assignment. :smile:

But that's a side discussion for George, James and myself, I guess.

Otherwise, assuming that was feasible, it could work similar to the reliability index/confidence level that's already calculated for a machine. I don't remember what all goes into those numbers... it's tucked away somewhere in the code on the site. Maybe we could just as easily setup some kind of nightly task that figures out the latest good/bad ratio and adjusts the reliability/confidence on the fly.

Those things are exposed on the website I think, where you can tick a box saying "I've fixed this CPU" or whatever, so if someone notices their machine has a lot of bad stuff and they correct it, they can check that box. But then again the only way I'd trust a system with a lot of bad results is if they start submitting a bunch of good ones, and that's accomplished by doing double checks that match. Should there be a manual override if so-and-so is sure they've fixed their reliability problems?

To start I'd set the baseline as "more bad than good". If that's the case, well hey, how about doing some double-checks for now? :smile: And... let the argument begin. LOL

frmky 2015-07-29 19:16

But I run for MOAR GHz-DaYs! Who cares if the residue is right?? :buddy:

frmky 2015-07-30 00:14

I just reserved the last remaining numbers on the previous lists, 41471879, 42174677, 42791519, and 43263431.

Mark Rose 2015-07-30 16:04

Everything above 54M in post #17 is done. I'm not going to post a list unless someone wants it.

Madpoo 2015-07-30 16:48

Here's a new list using the same criteria (3 times as many bad as good...) 21 of them.
[CODE]exponent Bad Good Unk Sus worktodo
42287257 10 2 15 8 DoubleCheck=42287257,72,1
42299431 4 1 4 0 DoubleCheck=42299431,72,1
42369917 7 2 5 0 DoubleCheck=42369917,72,1
42469309 4 1 8 1 DoubleCheck=42469309,72,1
42519247 15 3 5 4 DoubleCheck=42519247,72,1
42522107 10 3 12 1 DoubleCheck=42522107,72,1
42574673 7 2 2 4 DoubleCheck=42574673,72,1
42592393 4 0 28 1 DoubleCheck=42592393,72,1
42614993 4 0 28 1 DoubleCheck=42614993,72,1
42635947 4 1 2 1 DoubleCheck=42635947,72,1
42636029 4 1 2 1 DoubleCheck=42636029,72,1
42650897 13 3 5 3 DoubleCheck=42650897,72,1
53357987 17 2 29 14 DoubleCheck=53357987,73,1
54537607 6 0 12 4 DoubleCheck=54537607,73,1
54812297 3 0 4 3 DoubleCheck=54812297,73,1
54880909 17 2 29 14 DoubleCheck=54880909,73,1
54881429 17 2 29 14 DoubleCheck=54881429,73,1
55533703 4 0 6 0 DoubleCheck=55533703,73,1
55612019 6 1 4 3 DoubleCheck=55612019,73,1
55813103 4 0 6 0 DoubleCheck=55813103,73,1
55813129 4 0 6 0 DoubleCheck=55813129,73,1[/CODE]

Madpoo 2015-07-30 16:51

[QUOTE=Mark Rose;406896]Everything above 54M in post #17 is done. I'm not going to post a list unless someone wants it.[/QUOTE]

Thanks for doing those. Did you find any factors in the bunch?

Mark Rose 2015-07-30 17:17

[QUOTE=Madpoo;406903]Thanks for doing those. Did you find any factors in the bunch?[/QUOTE]

Yeah, I've found a few factors recently:

[url=http://www.mersenne.org/report_exponent/?exp_lo=46332389&full=1]46332389[/url]
[url=http://www.mersenne.org/report_exponent/?exp_lo=47549387&full=1]47549387[/url]
[url=http://www.mersenne.org/report_exponent/?exp_lo=50218603&full=1]50218603[/url]
[url=http://www.mersenne.org/report_exponent/?exp_lo=50375903&full=1]50375903[/url]
[url=http://www.mersenne.org/report_exponent/?exp_lo=50556641&full=1]50556641[/url]
[url=http://www.mersenne.org/report_exponent/?exp_lo=52293907&full=1]52293907[/url]
[url=http://www.mersenne.org/report_exponent/?exp_lo=55612153&full=1]55612153[/url]
[url=http://www.mersenne.org/report_exponent/?exp_lo=57909959&full=1]57909959[/url]

Not all of these show up in that post #17 list. I did have some previous assignments, but I don't keep track of my work too closely, and I read there may have been additional lists sent through PM.


All times are UTC. The time now is 06:03.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.