![]() |
|
|
#1 |
|
"Kieren"
Jul 2011
In My Own Galaxy!
2·3·1,693 Posts |
I am puzzled by a recent action by PrimeNet. Attached are the lines from the Assignment page which refer to the first core on my CPU:
As you can see, Core 1 had been working on M52390157 for some days. Yesterday, it was assigned M332220523 (!), and stopped work on the exponent in progress. Is this normal behavior? I have no objection to running the enormous exponent (5 year expected completion), but why did this supersede work in progress? Also, since this assignment the following has been appearing: [Mon Aug 29 08:07:31 2011 - ver 26.6]However, Core 1 is indeed working on M332220523 with iteration times approaching half a second. That seems about right, as it seems to be proportional to the size of the exponent. Any thoughts/comments would be appreciated. Thanks! |
|
|
|
|
|
#2 |
|
Dec 2003
Paisley Park & Neverland
101110012 Posts |
If Prime95 does some trial factoring and maybe P-1 on the new exponent then this is normal - because there may be a factor waiting to be found within some seconds of processing time and so your queue would be empty if you have no other assignments. When the trial factoring is done (and maybe P-1) Prime95 can be sure how long the test will take and will resume work on the first exponent.
|
|
|
|
|
|
#3 |
|
"Kieren"
Jul 2011
In My Own Galaxy!
2·3·1,693 Posts |
Thanks, MrHappy.
|
|
|
|
|
|
#4 |
|
"Kieren"
Jul 2011
In My Own Galaxy!
27AE16 Posts |
Later observation:
However, I must note that M332220523 is doing LL work, as was M52390157. Also, in the screenshot above, the latter shows 22 days to completion. That has now been updated to 1855 days. Again, I don't especially mind, but it does seem wasteful to put the work already done so far on the back burner that it will be reassigned. |
|
|
|
|
|
#5 | |
|
Jun 2003
7·167 Posts |
Quote:
When you say "M332220523 is doing LL work" do you mean that it was assigned as an LL? Or that the client is actually doing the LL test? It's normal for the client to do TF and P-1 factoring on an LL-assigned exponent prior to starting the LL test on that exponent. Additionally, if you have SequentialWorkToDo=0 in your prime.txt file, it will do TF and P-1 on a new exponent before finishing LL testing the current one. I'm also a bit puzzled as to why you've been given a 100M digit exponent to test. These are not for the faint-hearted. (In fact only one such test has ever been completed) You should not get them unless you specifically ask for them. |
|
|
|
|
|
|
#6 |
|
"Kieren"
Jul 2011
In My Own Galaxy!
100111101011102 Posts |
Hi Mr. P-1,
I'll try to answer, at least to the best of my abilities. 1) Is it doing actual LL work? Yes, as I understand things. Please see the screen shot attached (Worker 1). Note that two workers (3&6) recently completed and started new exponents. They are doing P-1 Stage One. Worker 1 changed during Saturday night. When I became aware of the change it appeared as it does now, except for the percent complete.One additional note: In worktodo.txt for Worker 1 it says: [Worker #1]M52390157 is the exponent which it was working on previously. I hope that I have given useful information. Thanks for responding and posing the questions. Last fiddled with by fivemack on 2011-08-30 at 07:43 |
|
|
|
|
|
#7 |
|
Sep 2002
Oeiras, Portugal
26·23 Posts |
M332220523 is currently being LL tested, as per the screenshot posted above.
It has skipped the last bit of TF and the P-1 stage (check the "Exponent status" report). Something is not right, and the test shouldn´t proceed until TF and P-1 are done. Such a long test will very likely turn out to be a waste of time and processing resources, with the currently available hardware, and doing it without previously performing the optimum TF and P-1 trials makes matters even worse. |
|
|
|
|
|
#8 | |
|
Account Deleted
"Tim Sorbera"
Aug 2006
San Antonio, TX USA
10000101010112 Posts |
Quote:
By the way, if you want to do both but do 52390157 first, you can stop and close Prime95, edit the worktodo.txt to put 52390157 before 332220523, and start Prime95 again. P.S., you might be wondering why this number is of any interest when it's so far above the normal first-time LL tests. It's because it has just over 100 million (a.k.a. "100M") digits, (to be exact, 100,008,343) and qualifies for the next EFF prize of $150,000, if it turns up prime. Looking at it from a purely practical standpoint, we probably shouldn't bother doing anything with 100M digit tests with today's hardware, (due to the length of time, cost, electricity, chance of error, etc.) but GIMPS is more geared for user choice than pure practicality. P.P.S. you should keep your assignment IDs private (those are the long hex strings after Test=), to make sure nobody steals your assignment. Last fiddled with by Mini-Geek on 2011-08-29 at 22:57 Reason: edited out assignment key |
|
|
|
|
|
|
#9 |
|
"Richard B. Woods"
Aug 2002
Wisconsin USA
22·3·641 Posts |
Moderators,
Please remove the assignment key shown in post #6. |
|
|
|
|
|
#10 |
|
"Kieren"
Jul 2011
In My Own Galaxy!
2·3·1,693 Posts |
OK. I know how this came about. See this thread:
http://www.mersenneforum.org/showthr...ight=332220523 In an unfocussed moment I entered this exponent (Advanced>Test...) while reading through various threads. Ignorance also played its part. I did not fully appreciate what I was asking the program to do. I vaguely remembered this, but could not find the appropriate thread until the light dawned and I did a search for the number itself. The line in worktodo.txt finally sank in, "AdvancedTest" as well. Sorry for stirring things up to no good end. I believe that I should probably remove that line from Worker #1 in worktodo.txt, no? Many thanks to those who responded. Please forgive the ignorance which made it necessary. |
|
|
|
|
|
#11 | |
|
Account Deleted
"Tim Sorbera"
Aug 2006
San Antonio, TX USA
17·251 Posts |
Quote:
|
|
|
|
|
![]() |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| What does * represent on Recent Results report? | petrw1 | PrimeNet | 5 | 2018-07-09 21:14 |
| recent cleared list | ramgeis | PrimeNet | 8 | 2016-05-14 23:51 |
| Recent Results/Factoring | Primeinator | PrimeNet | 4 | 2014-10-26 00:56 |
| Primenet Recent Results Request | garo | PrimeNet | 7 | 2014-02-03 02:53 |
| List of recent exponents that have been passed out | jasong | PrimeNet | 5 | 2007-05-18 22:56 |