1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Good Old P4 EE

Discussion in 'Overclocking & Cooling' started by Outback Bronze, Apr 29, 2012.

  1. Outback Bronze

    Outback Bronze

    Joined:
    Aug 3, 2011
    Messages:
    493 (0.44/day)
    Thanks Received:
    104
    Hi guys. Got a small problem here that sum of yous might b able to shed sum light on.

    Im doing an old school retro build. Here are the system specs:

    P4 Extreme Edition 3.4ghz
    Abit IC7-G MAX3
    2x1 Gigs Corsair ddr His x1950 pro
    Casecom 650w
    Veloci Raptor 300g
    Corsair 650D

    Ok, so i havent completly finished the build yet coz im waitn on sum water cooling parts. Anyway i went about doing a little overclocking on this beast (with the stock intel HSF) and had no joy at all. I put everything stock and was increasing the vcore only and checking temps. So what happens is i can get 1.65v to boot into widows but 1.675 and higher wont. Basically what im asking is what is failing on me? Is it heat or the shitty power supply ive got. I was doing a little testing and the 12v rail was dropping to 11.55 under load.
    Again no FSB OC only upping the volts. This is very strange! :confused:
  2. dark2099

    dark2099

    Joined:
    Feb 20, 2008
    Messages:
    4,398 (1.85/day)
    Thanks Received:
    755
    Location:
    where everyone wants to be
    If you are unable to change the multiplier or FSB in bios, leave the volts alone. Since you can't actually OC the chip, upping the volts would be useless.
  3. Outback Bronze

    Outback Bronze

    Joined:
    Aug 3, 2011
    Messages:
    493 (0.44/day)
    Thanks Received:
    104
    I dont think im able to change the multiplyer but the FSB definetly. I had it at 3.65ghz already and wanted to up the volts but windows crashes. That why i was testing vcore only. 1.65 all good 1.675 and up-bios good but widows no good.
  4. hat

    hat Maximum Overclocker

    Joined:
    Nov 20, 2006
    Messages:
    16,912 (5.97/day)
    Thanks Received:
    2,063
    Location:
    Ohio
    There's no point in just raising the voltage. If you leave everything else alone and just raise voltage it won't give any performance increase at all, just make your cpu draw more power and produce more heat. Maybe your cpu is hitting thermal limits and shutting down or maybe it just can't take that much voltage and craps out. You only raise voltage when you push your cpu clock too far and it goes unstable, raising voltage on its own with no other changes won't help any.
    Crunching for Team TPU
  5. Outback Bronze

    Outback Bronze

    Joined:
    Aug 3, 2011
    Messages:
    493 (0.44/day)
    Thanks Received:
    104
    Yer im wondering about the internal thermals coz i can get in2 window with 1.675v with no overclocking on the cpu but when im starting to overclock (3.65gz) it will only allow me to enter @ 1.625v. Im starting to believe that its governed by temperature. Is this normal 4 a P4 EE? Its getting into widows with the above clock speed and volts and is only @ 60 degrees.
  6. hat

    hat Maximum Overclocker

    Joined:
    Nov 20, 2006
    Messages:
    16,912 (5.97/day)
    Thanks Received:
    2,063
    Location:
    Ohio
    Is that figure idle or load? If idle, something somewhere along the line could be loading it to the breaking point.
    Crunching for Team TPU
  7. Outback Bronze

    Outback Bronze

    Joined:
    Aug 3, 2011
    Messages:
    493 (0.44/day)
    Thanks Received:
    104
    Its actually under 50% load(doesnt crash) but if i increase the voltage just one notch, windows will not boot. Im gona set up a water cooler on it soon to see if its just internal thermals. Surely i can put 1.75 volts on it, with no cpu overclock and at least be able to boot windows?
  8. Dippyskoodlez New Member

    Joined:
    Jul 1, 2005
    Messages:
    5,061 (1.52/day)
    Thanks Received:
    230
    Location:
    Ohio
    With the stock cooler, that is not only a lot of voltage for a P4, but thermal output is going to start skyrocketting. The standard HS is not going to like you at all. Those traditionally had a TDP of 115-130w already at stock IIRC. If its the dual core Prescott (I think those were the only EE), you are going to be in a world of hurt without an upgrade.

    If you cannot boot with a simple voltage boost, your motherboard may not be able to handle it, or you may have too small of a PSU (Multimeter to check your rails; POST takes a lot of peak current, and may overwhelm a borderline PSU). It may be old, but do not underestimate the power consumption.

    Keep in mind, also, you may not be reliably reading the thermal probe via software. Older CPU's (P4/Athlon XP series) often times had motherboard placed probes that were not reliable for an accurate core temp. If you are seeing 60 C on boot, you are already much too high, even with a current chip.
    Outback Bronze says thanks.
  9. Outback Bronze

    Outback Bronze

    Joined:
    Aug 3, 2011
    Messages:
    493 (0.44/day)
    Thanks Received:
    104
    Cheers bud. Well i hooked up a 800w with 65 amps on the 12 volt and it still crashes with a simple overvolt. So does this mean the motherboard cant handle this cpus voltage? Also can i cancel out the internal thermal observation? Thanks. Btw the P4EE s478 is a northwood.
  10. Sasqui

    Sasqui

    Joined:
    Dec 6, 2005
    Messages:
    7,544 (2.37/day)
    Thanks Received:
    1,332
    Location:
    Manchester, NH
    How many connector pins are you using for the CPU power on the board? The EE processors spcifically called for all 8, where you can use only 4 on most others. I saw that fail first hand on a buddy's rig.

    IC7-G... I had one, great board.
  11. Outback Bronze

    Outback Bronze

    Joined:
    Aug 3, 2011
    Messages:
    493 (0.44/day)
    Thanks Received:
    104
    Its only 4 pins. I dont think youll find a s478 mobo with 8 pin. Ive been tryn to read a few web pages on this cpu and board and sum people are having no problems getting 1.8v and up @ 4ghz+. I cant even boot windows with 1.7v only, no cpu overclock. I did put it at 1.8 and bios wouldnt even load? Its got me.

    Quote from a few web pages i found....

    Overclocking the P4EE:

    I've had some interesting experiences with the P4EE lately, and with overclocking it. It is certainly not the best overclocker (at least my chip). But one thing is for sure...it likes its voltage.

    Managed to run 3.6 GHz stably at default voltage with the Mach 2. Stable is being able to run Prime95 on one virtual CPU and CPUBurn on the other (100% usage) for 24 hours with no problems. Moving upward from 3.6 GHz requires, like clockwork, 0.05V per 5 MHz FSB (80 MHz). I decided to call it quits at 4080 MHz (255 FSB)@1.85V. I don't want to have the voltage that high 24/7 so I'll have to run it a little slower to keep the voltage down.

    Noticed a funny thing with the Winbond system monitor that came with the IC7-MAX3. It reports the CPU temp as 5.5°C under load, while the evaporator temp shows -28°C. End of quote.

    Another quote....

    Your right about the EE loving the voltage. I've found that some mobos with weaker core voltage regulation won't cut it as well. Although the Asus claims Prescott compatability, the core fluctuates badly at higher voltages with the EE in there. The P4C800 couldn't run as high as my IC7-G does. I'm able to run 24/7 at 4.210ghz with 1.875v on the IC7-G, so I'm very happy. End of quote.

    So why the hell i cant put a simple over volt(1.7 and uup) just to check temps is beyond me. I wonder if i change the memory that will do sumthin? Grrr bloody trouble shooting....
  12. Dippyskoodlez New Member

    Joined:
    Jul 1, 2005
    Messages:
    5,061 (1.52/day)
    Thanks Received:
    230
    Location:
    Ohio
    The mach 2 is a entry level phase change system.

    I'm willing to bet temperatures, TBH. 3.4-3.6 i also the top end for Northwoods. They really didn't have much headroom; letalone on stock cooler.
  13. eidairaman1

    eidairaman1

    Joined:
    Jul 2, 2007
    Messages:
    12,214 (4.68/day)
    Thanks Received:
    1,405
    Gallatin (Northwood Extreme) or Prescott (Hot Chip)

    Ill take Gallatin
  14. Dippyskoodlez New Member

    Joined:
    Jul 1, 2005
    Messages:
    5,061 (1.52/day)
    Thanks Received:
    230
    Location:
    Ohio
    Most definitely, but if you had phase, Prescott would outperform long term.
  15. eidairaman1

    eidairaman1

    Joined:
    Jul 2, 2007
    Messages:
    12,214 (4.68/day)
    Thanks Received:
    1,405
    He isnt running Phase, He is running a Radiator and Waterpump
    Last edited: May 2, 2012
  16. Outback Bronze

    Outback Bronze

    Joined:
    Aug 3, 2011
    Messages:
    493 (0.44/day)
    Thanks Received:
    104
    So its not the board its the internal temps? Coz i hope so. Dont get me wrong guys its not my every day system. Its just for a bit of fun/headache. Anyways im gona donk a water cooler on this puppy and see what she does. Id just assumed i could run a high voltage on it without it crashing on me with the stock HSF. Ive been given a P4 3.4 s478 prescott and im gona see what that does one day too. I just really want the EE @ 4ghz for this build. Its just a nice addition.
  17. Dippyskoodlez New Member

    Joined:
    Jul 1, 2005
    Messages:
    5,061 (1.52/day)
    Thanks Received:
    230
    Location:
    Ohio
    I only mentioned it because his reference quote was using a Mach 2 phase change doing the same thing he is attempting to do on air (Minus the stability). Temperature is very important.

    [​IMG]

    You can have the same clock and Vcore at a different temperature and see different power consumption. Having a 40% lower power consumption can mean the difference between chip life and death. (Yes I know this is ivybridge, but the physics is just more apparent with the 3d)
  18. repman244

    repman244

    Joined:
    Apr 7, 2011
    Messages:
    1,099 (0.89/day)
    Thanks Received:
    450
    My guess here is that temperature of the CPU is limiting you.
    I still have my old P4 Prescott which I managed to clock around 4.1GHz on air (with a Delta fan :D), it was a complete suicide run but the thing I noticed while OCing is that it gets really unstable at higher temps (at least my chip did), without the Delta fan strapped to the cooler I couldn't even reach 4GHz because of too low voltage, if I increased the voltage system would just crash.
    I was OCing on ABIT AA8-DuraMax, I can't remember the voltage I was feeding the CPU but it was somewhere close to maximum of the board, I was hitting around 58-60C on idle with that.
  19. Outback Bronze

    Outback Bronze

    Joined:
    Aug 3, 2011
    Messages:
    493 (0.44/day)
    Thanks Received:
    104
    Once i slap the water cooler on we'll all find out whats really going on. Like i said i hope its temp. Years ago i used to run a P4P800-E with a 3.0c and i could clock that to 3.75ghz stable with a water cooler and push voltage up2 1.8 with no crash. Once i install the water cooler on this P4EE im going to raise the voltage only and see if i get the same issue. Btw yes it is a Gallatin. Its just deriven for the Northwood core.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page