mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > PrimeNet

Reply
 
Thread Tools
Old 2020-04-19, 18:13   #1794
S485122
 
S485122's Avatar
 
Sep 2006
Brussels, Belgium

22·3·131 Posts
Default

I posted this in the triple check Thread.

I checked all the exponents a user tried to assign himself.

Exponents which he could not reserve :

83553139
83554087
83627359
83630863
83633729

Each of them has two expired P-1 assignments but no current reservation...

He was able to reserve the following exponents with only one expired P-1 assignment :

83543501
83562067
83602901

And of course he has able to claim those that have no expired assignment at all : - )

Others have confirmed this observation on other exponents. That unavailability has been discussed earlier...

Indeed something for the managers of PrimeNet.

Jacob

Last fiddled with by S485122 on 2020-04-19 at 18:15 Reason: simplified ?
S485122 is offline   Reply With Quote
Old 2020-04-25, 14:20   #1795
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

931210 Posts
Default Double assignments?

So, something's a bit strange on the Primenet Work Distribution Map. Some of the ranges above 110M have more TF assignments listed than there are candidates.

Drilling down, this report is showing that SRBase has two TF assignments per candidate. I didn't think Primenet supported more than one TF assignment at a time.
chalsall is online now   Reply With Quote
Old 2020-04-25, 17:46   #1796
LaurV
Romulan Interpreter
 
LaurV's Avatar
 
Jun 2011
Thailand

32×971 Posts
Default

Envious?
You can't get two assignments for the same expo huh?

Joking apart, I think it is a mistake, "somebody" opened a door for them to be able to get lots of assignments, and forgot to close it properly. Even if you want to do more bitlevels, it makes no sense to do more assignments for the same expo. Either reserve it "from 72 to 77", or just reserve it "from 72 to 73" but do not return the results until you finish with it. Creating one assignment for every bit would be silly. I think it is an error, which will be fixed soon.

We did both ways in the past, and also the other way around, we were looking for work to low bitlevels, like 73, but the server only assigned us 72 to 76 or so, therefore we did the 72 to 73 and reported it, and we have seen that the assignment is gone, therefore we didn't do the rest of the bits. Doing them would be detrimental, as somebody else could get the assignments and then either we or them, will end with "error 40, results not needed, your work is not wanted, you wasted your ticks in vain".

We raised this flag in the past, but it was not heard. (well, we didn't insist hard, because we liked the idea of picking low hanging fruits too, and yes, we just tried, this is still working!)

Last fiddled with by LaurV on 2020-04-25 at 17:47
LaurV is offline   Reply With Quote
Old 2020-04-25, 18:55   #1797
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

25×3×97 Posts
Default

Quote:
Originally Posted by LaurV View Post
Joking apart, I think it is a mistake, "somebody" opened a door for them to be able to get lots of assignments, and forgot to close it properly. Even if you want to do more bitlevels, it makes no sense to do more assignments for the same expo. Either reserve it "from 72 to 77", or just reserve it "from 72 to 73" but do not return the results until you finish with it. Creating one assignment for every bit would be silly. I think it is an error, which will be fixed soon.
That's why I raised the observation...

Aaron has said before that the Primenet database doesn't have a bit-map for TF bit-levels. And, therefore, it can cause problems if TF'ing is reported out-of-order. As in, as you said, if you're going to go to 77, go 77 in one step (or multiple, but in order).
chalsall is online now   Reply With Quote
Old 2020-04-26, 03:56   #1798
LaurV
Romulan Interpreter
 
LaurV's Avatar
 
Jun 2011
Thailand

32×971 Posts
Default

Yeah.

Another thing that is still happening (and we used to take advantage of it in the past) is the following: you go to manual assignment for GPU page and take few high-exponent assignments, where you select the last bit to be 72 or 73, and the range over 400M (where there still are assignments available for this low bitlevel). You click the button, and get a list of as many expos you requested, but they are not to 72, they are "upgraded" by the server to be 75, or 76. Well, nothing wrong with it, gpu72 also does that, if you let "it" decide, the last bit will be raised to the "optimal" level.

But now, you close the manual assignment page, without adding any work to your cards, and go again to account/my assignments and ask for the assignments you have. The server will say "you have these assignments already:" and give you the list.

But... BUT! In the list of assignment you get now they are all to 72 (or how much you requested) and NOT to the "inflated" values. You can grab and paste THIS list to your cards, work on them, and NOT to the inflated expos, and you just got low hanging fruits.

As the server is set now (with this "bug/feature") it is in fact recommended to do so! Because, if you use Misfit, for example, and automatically report each bitlevel as soon as you finish it, then the assignment is "gone" after your first report. Then, if you continue, as per the inflated list, i.e. continue to TF to the levels PrimeNet gave you initially, then you WILL end up doing the higher bitlevel in the same time somebody else is getting them assigned and working them (you or the other user will get error 40 upon completion of the higher bitlevels). That is because the assignment, as said, is gone when misfit first reports the low bitlevel.

Moral is that after you request assignments from PrimeNet using the manual form, you should every time go to the "my assignments" and check what you really GOT. And work on that. The initial grab list (the inflated one) is not to be trusted. Maybe that's why some users who got manual assignments got "work not needed", and not because of poaching.

Last fiddled with by LaurV on 2020-04-26 at 04:02
LaurV is offline   Reply With Quote
Old 2020-04-26, 15:44   #1799
petrw1
1976 Toyota Corona years forever!
 
petrw1's Avatar
 
"Wayne"
Nov 2006
Saskatchewan, Canada

22×3×5×73 Posts
Default Any chance you can properly direct these results?

For some reason my first batch of results from a new PC got credited to -Anonymous- even though the results.txt has my userid.

Recent results are properly credited to me.

Here are the offenders.

Code:
[Sat Apr 25 19:07:42 2020]
UID: petrw1/Stinky_Pete, M56694191 completed P-1, B1=1500000, B2=30000000, E=12, Wh8: BB5C3657
UID: petrw1/Stinky_Pete, M56694097 completed P-1, B1=1500000, B2=30000000, E=12, Wh8: BB5B365E
UID: petrw1/Stinky_Pete, M56693999 completed P-1, B1=1500000, B2=30000000, E=12, Wh8: BB563650
UID: petrw1/Stinky_Pete, M56694181 completed P-1, B1=1500000, B2=30000000, E=12, Wh8: BB52365E
UID: petrw1/Stinky_Pete, M56693969 completed P-1, B1=1500000, B2=30000000, E=12, Wh8: BB273654
UID: petrw1/Stinky_Pete, M56694887 completed P-1, B1=1500000, B2=30000000, E=12, Wh8: BB57365D
UID: petrw1/Stinky_Pete, M56696713 completed P-1, B1=1500000, B2=30000000, E=12, Wh8: BB5C3636
[Sat Apr 25 19:14:04 2020]
P-1 found a factor in stage #2, B1=1500000, B2=30000000, E=12.
UID: petrw1/Stinky_Pete, M56670233 has a factor: 184602903967341014096340982753 (P-1, B1=1500000, B2=30000000, E=12)
petrw1 is offline   Reply With Quote
Old 2020-04-27, 00:59   #1800
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

2×72×73 Posts
Default

Quote:
Originally Posted by petrw1 View Post
Recent results are properly credited to me.
Fixed.
Prime95 is online now   Reply With Quote
Old 2020-04-27, 02:22   #1801
petrw1
1976 Toyota Corona years forever!
 
petrw1's Avatar
 
"Wayne"
Nov 2006
Saskatchewan, Canada

22·3·5·73 Posts
Default

Quote:
Originally Posted by Prime95 View Post
Fixed.
Thanks but they show up in my results page without a Computer Name
petrw1 is offline   Reply With Quote
Old 2020-04-27, 03:58   #1802
Prime95
P90 years forever!
 
Prime95's Avatar
 
Aug 2002
Yeehaw, FL

157628 Posts
Default

Quote:
Originally Posted by petrw1 View Post
Thanks but they show up in my results page without a Computer Name
You can't have everything :)
Prime95 is online now   Reply With Quote
Old 2020-05-10, 21:09   #1803
S485122
 
S485122's Avatar
 
Sep 2006
Brussels, Belgium

22·3·131 Posts
Default TF assignments accepted for Cat 1 exponents ?

It seems that it is possible to reserve an exponent for trial factoring (with the trial factoring expiry rules) when an exponent is in CAT 1. A query on the active assignments between 50M and 52M wil show it.

IMO once an exponent is in CAT 2, but even more in CAT 1 it should not possible to assign it for manual testing and certainly not for trial factoring.

Jacob
Attached Thumbnails
Click image for larger version

Name:	Capture.JPG
Views:	24
Size:	134.9 KB
ID:	22282  
S485122 is offline   Reply With Quote
Old 2020-05-11, 18:04   #1804
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

246016 Posts
Default

Quote:
Originally Posted by S485122 View Post
IMO once an exponent is in CAT 2, but even more in CAT 1 it should not possible to assign it for manual testing and certainly not for trial factoring.
Seconded.

Another (IMO more serious) issue is the 1002 assignments in the "Cat 1" 91M range.

The assignee, SRS Technology, has some compute behind him. But these need to be worked in the next few days, or else expired. He's about to hoard from the LL'ers.
chalsall is online now   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Official "Faits erronés dans de belles-lettres" thread ewmayer Lounge 39 2015-05-19 01:08
Official "all-Greek-to-me Fiction Literature and Cinema" Thread ewmayer Science & Technology 41 2014-04-16 11:54
Official "Lasciate ogne speranza" whinge-thread cheesehead Soap Box 56 2013-06-29 01:42
Official "Ernst is a deceiving bully and George is a meanie" thread cheesehead Soap Box 61 2013-06-11 04:30
Official "String copy Statement Considered Harmful" thread Dubslow Programming 19 2012-05-31 17:49

All times are UTC. The time now is 01:01.

Thu Sep 24 01:01:51 UTC 2020 up 13 days, 22:12, 0 users, load averages: 1.50, 1.37, 1.52

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.