![]() |
![]() |
#1 |
Feb 2004
California
910 Posts |
![]()
One of my computers is enclosed in the cabinet. Before I installed Prime95 I had Power Mgmt power down monitor after 15 min and hard drive after 20min to keep things cool. Since Prime95 I set Hard drive not to power down at all because I thought that was necessary for Prime95 to run 24/7. Now CPU temps go well over 60C with the door closed. No errors yet but I live in the desert and summer is coming. Can I power down the hard drive and increase the time in between writes to the disk safely? Would 12 hours be too long? Would this impact the time that Prime is running or its performance? Please, advise and if necessary expand the answer to any other issues with this if I didn't foresee them. Thank you all very much.
|
![]() |
![]() |
#2 |
Sep 2002
29616 Posts |
![]()
If you are only trial factoring you could use a floppy to store all the working files for Prime95 :
prime.ini, worktodo.ini, local.ini, prime.spl, results.txt, .log, and save file p#######. The program and any dlls would be on the hard disk and are only read when starting the program and maybe when communicating with the server. There is an option to use a different directory/drive for the working files. I have it write to the floppy every 10 minutes. I don't know what the storage requirements are for P-1 and LL/DC so I don't know if the save file(s) would fit. If you have a zip drive maybe that would work OK or if on a network have it write to a network drive (not in the cabinet). You can have the writes to disk delayed (I am not sure how long), the issue will be loss of hours of work if it crashes. |
![]() |
![]() |
#3 |
Aug 2002
London, UK
3·31 Posts |
![]()
In a nutshell: most of the heat comes from Prime95 using the CPU (processor) to its maximum all of the time. Most PCs are hardly using any of their potential CPU power, but Prime95 is so well written and carefully optimised that it pushes the CPU to its limit, and as this uses much more electrical power than usual, it generates much more heat. This is behind the temperature increase you have noticed.
Note also that by default Prime95 writes checkpoint files (a temporary note of the result so far) to the hard disk every 30 minutes. This gives it something to go back to in case the PC crashes or the application is not shut down in an orderly way. Because of this, if you set the HD to power down every 20 minutes, it will only need to spin up 10 minutes later. This will increase the mechanical and thermal strain on the HD enormously, so I suggest you don't do this. (You can increase the time between checkpoints, but then you risk losing more work if Prime95 has to go back to the latest one: the tradeoff is yours to make.) I would look very carefully indeed at your cooling - it isn't a good idea to run Prime95 on a PC inside a closed case. The heat may eventually damage some part of the PC. I would say you need to leave the cabinet doors open 24/7, and even then the airflow may be noticably impeded by being inside the cabinet. Last fiddled with by Reboot It on 2004-02-17 at 19:09 |
![]() |
![]() |
#4 | |
Aug 2002
London, UK
10111012 Posts |
![]() Quote:
|
|
![]() |
![]() |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Fast Breeding (guru management) | VictordeHolland | NFS@Home | 2466 | 2020-09-20 06:51 |
Improving the queue management. | debrouxl | NFS@Home | 10 | 2018-05-06 21:05 |
Script-based Primenet assignment management | ewmayer | Software | 3 | 2017-05-25 04:02 |
Skylake's power management under Linux is dreadful and you shouldn't buy one until it's fixed | Dubslow | Hardware | 8 | 2016-04-15 09:14 |
Nvidia Cuda 6.0 unified memory management | Manpowre | GPU Computing | 4 | 2013-11-28 18:53 |