Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
Hi there,
There is an issue with role permissions that is being worked on at the moment.
If you are having trouble with access or permissions on regional forums please post here to get access: https://www.boards.ie/discussion/2058365403/you-do-not-have-permission-for-that#latest

Best cooling system?

Comments

  • Moderators, Science, Health & Environment Moderators Posts: 10,088 Mod ✭✭✭✭marco_polo


    What sort of a case and a cooler are you currently using? Should be able to get more out of it than 5% on air.


  • Closed Accounts Posts: 64 ✭✭wcp1200


    Not sure. Some big ass alienware one. How would i find out? Its possibly the area 51. Just using stock fans. When I clock it to 10% plus its ok while doing nothing but the minute i go on a stream or play a game the fans go crazie and the screen freezes. cores hit high thirties and then it freezes. But what do you think of the cooler?


  • Moderators, Technology & Internet Moderators Posts: 18,381 Mod ✭✭✭✭Solitaire


    Its not overheating. Could be one of a whole buttload of problems that's causing the crashing. Most obvious one is memory: its own stability could have already been marginal, and once you bump up the FSB to OC the CPU the memory will get OC'd too if you don't drop the memory divider! :o

    Or if it was a custom Alienware rig the CPU may have come pre-undervolted and could need a voltage boost. I do recall that the first binning of the 45nm quads were dire; out of those if the name didn't end in "50" it was guaranteed naff. They really hated certain mobos too :o But I wouldn't give up just yet; given <40 degrees is still well within the optimum temp range for a C2D I don't think heat has anything to do with it :p


  • Closed Accounts Posts: 64 ✭✭wcp1200


    Solitaire wrote: »
    Its not overheating. Could be one of a whole buttload of problems that's causing the crashing. Most obvious one is memory: its own stability could have already been marginal, and once you bump up the FSB to OC the CPU the memory will get OC'd too if you don't drop the memory divider! :o

    Or if it was a custom Alienware rig the CPU may have come pre-undervolted and could need a voltage boost. I do recall that the first binning of the 45nm quads were dire; out of those if the name didn't end in "50" it was guaranteed naff. They really hated certain mobos too :o But I wouldn't give up just yet; given <40 degrees is still well within the optimum temp range for a C2D I don't think heat has anything to do with it :p
    Ok thanks. The cpu voltage is at 1.50000 and it can go up to 1.90000. Should I put it up? Memory votage is at 1.850 and it can go up to 3.425. Should I put this up? How do I drop the memory divider? Im a OC noob so any help on this would be appreciated.

    Thanks


  • Moderators, Technology & Internet Moderators Posts: 18,381 Mod ✭✭✭✭Solitaire


    Whoa! Anything over 1.36V voids the warranty on Core2 CPUs; anything over 1.5V makes them melt!! :eek: And what kind of memory do you have in there? Most DDR2 is safe up to 1.9V; past that only modules with heatspreaders should be overvolted.

    And OCing a CPU is always an experimental process; you should drop the memory divider to slow down your RAM below even stock speed so they don't interfere with stability. OC them back to a good speed after you've found a stable CPU setting.


  • Advertisement
  • Closed Accounts Posts: 64 ✭✭wcp1200


    In the asus ai suite it says the cpu volt is at 1.28000.



    Operating System
    MS Windows Vista Home Premium 64-bit SP1
    CPU
    Intel Core 2 Quad Q9400 @ 2.66GHz 34 °C
    Yorkfield 45nm Technology
    RAM
    4.0GB Dual-Channel DDR2 @ 420MHz (6-7-7-31)
    Motherboard
    ASUSTeK Computer INC. P5N-D (Socket 775)
    Graphics
    @ 1680x1050
    NVIDIA GeForce GTX 260 (nVidia)
    NVIDIA GeForce GTX 260 (nVidia)
    Hard Drives
    488GB NVIDIA STRIPE 465.77G (SCSI)
    Optical Drives
    HL-DT-ST DVD-RAM GH22NP20 ATA Device
    Audio
    High Definition Audio Device


  • Moderators, Technology & Internet Moderators Posts: 18,381 Mod ✭✭✭✭Solitaire


    Oh crap :o

    P5N-D is a nightmare for quads and AI Suite makes it ten times worse :mad:

    First thing first: Bin AI Suite if you can, its worse than useless - not only are the readouts misleading there's been talk that it really doesn't do the P5N-D any good at all and can kill it or the CPU :eek: It can also cause a buttload of glitches :(

    The P5N-D is also a pain for bugs, sometimes when OCing it will tell your OS you've physically changed your CPU for some reason :confused: Vdroop is also horrid on that model. Revert all AI Suite settings to default then try to uninstall it and do all future OCing straight from the BIOS. Try HWMonitor, just take any software measurements (particularly of power consumption!) with a big pinch of salt... :p

    You should probably bin your current OC - wipe CMOS, then update the BIOS on the P5N-D to the latest version and clear CMOS one more time before trying anything. Also find out what RAM is in there! PC4200, PC5300, PC6400 etc. latencies and what maximum voltage they'll tolerate - it should be on a wee label physically on the RAM modules! Not that the memory will be running fast at all until we figure out what we're doing with the CPU... ;)


  • Closed Accounts Posts: 13,874 ✭✭✭✭PogMoThoin


    The only way to overclock a cpu is through the bios, all software is crap


  • Registered Users, Registered Users 2 Posts: 86,729 ✭✭✭✭Overheal


    /uninstalls AI suite..


  • Closed Accounts Posts: 64 ✭✭wcp1200


    Ok so i reset and unistalled everything. Than went to bios and set cpu volt to 1.27500. Set fsb to 1500. Now its running at 3.0ghz while on this forum. Any program to stress test it?


  • Advertisement
  • Closed Accounts Posts: 13,874 ✭✭✭✭PogMoThoin


    Occt, keep an eye on temps and report back. Generally lockups or windows reboots are usually ram problems (it should be on a divider running slow to take it out of the equasion for now, you can push that later), failing in occt usually means the chip needs more volts


  • Closed Accounts Posts: 64 ✭✭wcp1200


    Ok so its been runing perfect at 3ghz. What can this cpu handle? 3.6?


  • Registered Users, Registered Users 2 Posts: 2,928 ✭✭✭VenomIreland


    Each CPU is different, see how high you can take yours safely, use something like Core Temp, Speedfan or HWMonitor to monitor the temps when using OCCT.


  • Closed Accounts Posts: 64 ✭✭wcp1200


    At what temps should i get worried at?


  • Closed Accounts Posts: 64 ✭✭wcp1200


    Ok so problem. When i put the cpu fsb above 1500 it just gives me a system boot failure. I noticed that the cpu and ddr2 are linked so when i put the cpu up to 1500 the ddr went up to 900. Should i try unlinked and just do the cpu?


  • Closed Accounts Posts: 13,874 ✭✭✭✭PogMoThoin


    Each CPU is different, see how high you can take yours safely, use something like Core Temp, Speedfan or HWMonitor to monitor the temps when using OCCT.

    OCCT has coretemp built in
    wcp1200 wrote: »
    Ok so problem. When i put the cpu fsb above 1500 it just gives me a system boot failure. I noticed that the cpu and ddr2 are linked so when i put the cpu up to 1500 the ddr went up to 900. Should i try unlinked and just do the cpu?

    Yes, You need to run Your ram slower. What ram have You got?


  • Moderators, Technology & Internet Moderators Posts: 18,381 Mod ✭✭✭✭Solitaire


    Drop the memory divider to the minimum then mess with it once you have the desired CPU settings dialed in ;)


  • Closed Accounts Posts: 64 ✭✭wcp1200


    Oh god this thing is complicated. Ok heres what i did. I changed voltages from auto to this. vcore-1.35 dram-auto NB-1.36 SB-1.5 HT-1.38. Then i raised the cpu to 1600(3.2ghz) and left the ram at 800. It got passed the usual failed boot but than after the windows loading bar where your login comes up the screen just went black. So after that i went back to the bios and dropped the ram to 400(minimum) and so far its working fine. So can i keep the ram at the lowest setting or will this affect performance or maybe i need to put the voltages up more? Thanks for the help so far guys.

    I posted ram in a earlier post. Its elpida dual ddr2 4GB.


    Ok so i tried to put it to 3.4ghz but when i save and exit it goes to restart but the screen just stays black so should i up the voltages again?


  • Moderators, Technology & Internet Moderators Posts: 18,381 Mod ✭✭✭✭Solitaire


    wcp1200 wrote: »
    I posted ram in a earlier post. Its elpida dual ddr2 4GB.

    That doesn't tell us what speed it is supposed to run at, or at what latencies. 420MHz isn't a standard speed either! :o
    Ok so i tried to put it to 3.4ghz but when i save and exit it goes to restart but the screen just stays black so should i up the voltages again?

    I think its time to move in small steps. Drop the RAM to minimum speed then try setting the FSB to 413MHz to get a 3.3GHz CPU clock and see if it holds up. Try increasing chipset voltages a tiny bit before overvolting the CPU again. Also check what voltage the CPU is "seeing" with CPU-z to see what kind of Vdroop that mobo's saddled you with - if the CPU voltage still seems very low due to Vdroop you can be a little more aggressive with overvolting it but try to avoid getting a reading in CPU-z higher than 1.35V! :eek:


  • Closed Accounts Posts: 64 ✭✭wcp1200


    Ok so things were going well with 3.2 but i ran the occt thing and got an eroor in core 2 just after 8 mins. But i dont know what the error was. Temps look fine. I have ram set to lowest. 1.4 vcore and its 1.35 in cpu z. Ill find out the ram after work. What details do ye want exactly?


  • Advertisement
Advertisement