mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > PrimeNet > GPU to 72

Reply
 
Thread Tools
Old 2015-09-28, 04:47   #23
kladner
 
kladner's Avatar
 
"Kieren, ktony"
Jul 2011

2×72×97 Posts
Default

Quote:
Originally Posted by kladner View Post
South East sky in North Side Chicago clouded up just before Prime Time. About an hour ago, I went out and could see a bright spot in that direction. Not even that, now.
So I saw a crescent out the window. I told a sleeping Dan that he could see something, and I was going out to check on it. He said, "If you go out, take a camera." By the time I got my shoes on, figured out appropriate time-exposure settings on my camera, mounted it on the tripod, and got it outside, the moon was just a bright smudge, again.

At 11:40 PM CDT, I can see a nice, completely full moon out the window.
kladner is offline   Reply With Quote
Old 2015-09-28, 12:01   #24
alpertron
 
alpertron's Avatar
 
Aug 2002
Buenos Aires, Argentina

22×7×47 Posts
Default

In Buenos Aires there were no clouds, and I was able to see the first half of the lunar eclipse until 12 AM local time (3AM UTC) including the red phase. There were some shows in TV in which they showed the Moon through telescopes during about 5 hs so all phases of the Eclipse could be seen.
alpertron is offline   Reply With Quote
Old 2015-09-28, 19:44   #25
Madpoo
Serpentine Vermin Jar
 
Madpoo's Avatar
 
Jul 2014

3·5·7·31 Posts
Default

Quote:
Originally Posted by chalsall View Post
Whoops. Sorry about that!

One of my "stacks" was offline for several weeks because of a dead UPS (power is hell here in Bim!). I had thought the candidates would complete in time and/or Primenet would tell them that the candidates were no longer assigned to them.

Madpoo, please transfer the credit as appropriate if it's not too much trouble.

Now, back to the eclipse (it's spectacular!!!)
If anyone had their assignment "poached" (but not really), holler at me and I'll figure out what goes where. If you check in your work as a triple-check, it'll say it was for an already verified exponent but you'll still get credit. In fact, I *think* that might be true for assigned TF work as well.

Heck, you'll even get credit for turning in an LL result on a factored exponent, as long as you had a valid assignment id, just out of pity for the work put in. LOL
Madpoo is offline   Reply With Quote
Old 2015-10-01, 11:37   #26
guido72
 
guido72's Avatar
 
Aug 2002
Rovereto (Italy)

3·53 Posts
Default

Quote:
Originally Posted by TheJudger View Post
Hi Chris,

can you check those two exponents? I got them for P-1 from GPU to 72, I just wanted to return them but obviously results are not needed.A bit strange are the results anyway (same for both exponents):
User "Guido Lorenzini" reported "no factor from 273 to 274" and "no factor from 274 to 275" on 2015-09-17 and than reported factors within exactly that range on 2015-09-25.

Oliver
Hi Oliver. This is Guido writing.
I've experimented some iussue in the recent past with my GPUs. I immediately thought this was due to the overclock. (I had some PMs with Chris about that). So I've reset the GPUs freqs to the stock level. But this wasn't enough. The Nvidia driver was still crashing randomly while Mfaktc's running. These crashes were even more frequent when I was trying to run CudaLucas [even if the error checks were reasonably low (0,012)] (This is why I've decided to abandon my efforts with Cudalucas). So I've start checking between each GPU. I think I've found the guilty and removed it from the search. The results you've noticed are due to the fact that I've rerunned some tests when I saw something weird but, onestly my checks are absolutely not sistematic. Hope thing are going well now and my results are reasonably reliable. I'm really sorry for any problem I've caused to GIMPS.
Guido
guido72 is offline   Reply With Quote
Old 2015-10-01, 12:08   #27
guido72
 
guido72's Avatar
 
Aug 2002
Rovereto (Italy)

15910 Posts
Default

Hi all!
As suggested to me by Chris I'm pointing out here another weird behaviour I've noticed running Misfit. I was reluctanct in doing this. I was afraid to look like a moron since I was overclocking my hardware and, above all, I'm not able to give more specific infos...
Some time ago I switched between GPU72 and Primenet as source for my assignements (I was "frustraded" because I was finding so few factors at these relative high fact bit levels...). Misfit was set to balance it as soon as the difference in the workload between my 5 GPUs was more than 6 exps (this is because 4 GPUs are quite fast and one is much slower: 7/800 Ghz vs 150 GHz/day). Well, I don't know why but, after that swiching I found in my misfit_worktodo file quite weird assignements: some were duplicates, some were already factored, finally some were mismatching with GPU72 ones. As suggested by Chris, I've completely erased all of my worktodo files, re-downloaded a new worktodo file from GPU72 and restarted everything... By now I do not have yet exactely understand what happened...
guido72 is offline   Reply With Quote
Old 2015-10-01, 20:50   #28
VictordeHolland
 
VictordeHolland's Avatar
 
"Victor de Hollander"
Aug 2011
the Netherlands

23·3·72 Posts
Default

Would it be possible to find out all the assignments that specific GPU did so we (the community) can recheck them?
VictordeHolland is offline   Reply With Quote
Old 2015-10-02, 07:57   #29
guido72
 
guido72's Avatar
 
Aug 2002
Rovereto (Italy)

3×53 Posts
Default

Unfortunately no. I set the saving period for log files to 60 days in Misfit just after the troubles: I may say wich GPU has done what with specific timestamp right now... but no info for the recent past. Anyway: just 2000 results or something has been credited to me in the last two months...
I'm still getting this from "Al"
File:GIOM_SENT\RESULTS_EXPORTED_130881874957746989_79a1a560.txt @@ 01/10/2015 17:38:15
PROCESSING RESULT: NO FACTOR FOR M73400729 FROM 2^73 TO 2^74 [MFAKTC 0.20 BARRETT76_MUL32_GS]
ERROR CODE: 40, ERROR TEXT: TF RESULT FOR M73400729 WAS NOT NEEDED
PROCESSING RESULT: NO FACTOR FOR M73400729 FROM 2^74 TO 2^75 [MFAKTC 0.20 BARRETT76_MUL32_GS]
ERROR CODE: 40, ERROR TEXT: TF RESULT FOR M73400729 WAS NOT NEEDED
PROCESSING RESULT: NO FACTOR FOR M73431269 FROM 2^73 TO 2^74 [MFAKTC 0.21 BARRETT76_MUL32_GS]
ERROR CODE: 40, ERROR TEXT: TF RESULT FOR M73431269 WAS NOT NEEDED
PROCESSING RESULT: NO FACTOR FOR M73431269 FROM 2^74 TO 2^75 [MFAKTC 0.21 BARRETT76_MUL32_GS]
ERROR CODE: 40, ERROR TEXT: TF RESULT FOR M73431269 WAS NOT NEEDED
PROCESSING RESULT: NO FACTOR FOR M73400969 FROM 2^73 TO 2^74 [MFAKTC 0.21 BARRETT76_MUL32_GS]
ERROR CODE: 40, ERROR TEXT: TF RESULT FOR M73400969 WAS NOT NEEDED
PROCESSING RESULT: NO FACTOR FOR M73400969 FROM 2^74 TO 2^75 [MFAKTC 0.21 BARRETT76_MUL32_GS]
ERROR CODE: 40, ERROR TEXT: TF RESULT FOR M73400969 WAS NOT NEEDED
This is just an excerpt. I've got a couple of dozen of these messages recentely...

Everything seem to work properly here (I mean my hw...) right now...

Last fiddled with by guido72 on 2015-10-02 at 08:05 Reason: integration...
guido72 is offline   Reply With Quote
Old 2015-10-03, 01:20   #30
TheJudger
 
TheJudger's Avatar
 
"Oliver"
Mar 2005
Germany

1,103 Posts
Default

Hi Guido,

good to know the reason. Thank you for clearing up the situation.

Oliver
TheJudger is offline   Reply With Quote
Old 2015-10-03, 18:00   #31
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

227816 Posts
Default

Quote:
Originally Posted by guido72 View Post
I was reluctanct in doing this. I was afraid to look like a moron since I was overclocking my hardware and, above all, I'm not able to give more specific infos...
In my opinion, the most important thing a technical / scientific person can do is to share what they observe, and be willing to admit they might have done things sub-optimally based on counter evidence and/or argument (but they should fight like heck if they think they're correct).

The best thing that will happen is they're proven correct. The worst is they're proven wrong and/or they made a mistake; and thus they learn.

This why I enjoy the Sciences so much; it often takes thick skin, but everyone benefits at the end of the day.
chalsall is offline   Reply With Quote
Old 2015-10-04, 02:26   #32
LaurV
Romulan Interpreter
 
LaurV's Avatar
 
Jun 2011
Thailand

22·5·419 Posts
Default

Quote:
Originally Posted by chalsall View Post
the most important thing a technical / scientific person can do is to share what they observe, and be willing to admit they might have done things sub-optimally
+1
LaurV is offline   Reply With Quote
Old 2015-10-08, 21:54   #33
swl551
 
swl551's Avatar
 
Aug 2012
New Hampshire

2×401 Posts
Default

Quote:
Originally Posted by VictordeHolland View Post
Would it be possible to find out all the assignments that specific GPU did so we (the community) can recheck them?
Actually Yes... In the MISFITBackups directory there is a file for every results.txt exported and the filename is the path to the GPU instance

this file
BACKUP_RESULTS_EXPORTED_D-os_shared_root-GIMPS-TF-MFAKTC-Results.txt_130855415262115913_fa30fa1e

came from
D:\OS_SHARED_ROOT\GIMPS\TF\MFAKTC

Thanks

Scott
swl551 is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
tf results not needed MatWur-S530113 GPU to 72 10 2017-12-09 15:36
Results not Needed RMAC9.5 PrimeNet 3 2013-06-26 13:16
not needed zeit PrimeNet 3 2008-04-25 08:03
Help needed AntonVrba Math 3 2007-03-06 10:55
V24.12 QA help needed Prime95 Software 5 2005-06-17 15:54

All times are UTC. The time now is 01:45.

Tue Mar 31 01:45:27 UTC 2020 up 5 days, 23:18, 0 users, load averages: 1.81, 1.67, 1.53

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.