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

Theory for finding OC potential without OC?

Discussion in 'Overclocking & Cooling' started by xvi, Aug 26, 2010.

  1. xvi

    xvi

    Joined:
    Nov 10, 2006
    Messages:
    1,959 (0.69/day)
    Thanks Received:
    1,092
    Location:
    Washington, US
    I have a ton of processors in my For Sale thread. I'd like to find some way to test the overclocking potential of each processor, but I don't want to have to go through tons of good thermal paste to clock them up high. Instead, I figure I'll let them run at stock clock speeds, but slowly reduce the voltage until they become unstable. Unless I'm mistaken, the lower the minimum stable voltage to run stock clocks, the higher their OC potential is. Additionally, I won't have to worry about excessive heat/volts.

    Is this a safe and reasonable theory?
    Crunching for Team TPU
  2. 3volvedcombat

    3volvedcombat New Member

    Joined:
    May 10, 2009
    Messages:
    1,514 (0.78/day)
    Thanks Received:
    261
    Location:
    South California, The desert.
    Your still going to have to use thermal paste, your good stuff and a shit load of it. It doesn't matter how much thermal paste you put on your processor- if your putting the processor in the socket and running it, your going to have to use thermal paste- period.

    Makes no sense.

    And the chips in your for sale arnt worth overclocking now, because they will still be slow at the task of playing some high end games.
  3. xvi

    xvi

    Joined:
    Nov 10, 2006
    Messages:
    1,959 (0.69/day)
    Thanks Received:
    1,092
    Location:
    Washington, US
    Of course I'm going to use thermal paste. I never said any of the processors would break hwbot's SuperPI record, outperform your neighbors i7 or score you an extra 50 FPS on Crysis. They're just there for people who have a use for them. Speed up a NAS that uses compression. Open source router boxes. HWBot junkies looking for extra points. If no one here takes them, they'll be torn down for the metal in them.

    Not everyone obsesses over "high end games".
    Crunching for Team TPU
  4. WhiteLotus

    WhiteLotus

    Joined:
    Jul 30, 2007
    Messages:
    6,532 (2.52/day)
    Thanks Received:
    847
    use a program, can't for the life of me remember what it is, and it will tell you the VID.

    edit, coretemp
  5. xvi

    xvi

    Joined:
    Nov 10, 2006
    Messages:
    1,959 (0.69/day)
    Thanks Received:
    1,092
    Location:
    Washington, US
    VID is a new one on me. Lower the better? Higher the better?

    Also, these are P3s, P4s and Athlon XPs.
    Crunching for Team TPU
  6. Paulieg

    Paulieg The Mad Moderator Staff Member

    Joined:
    Feb 19, 2007
    Messages:
    11,914 (4.34/day)
    Thanks Received:
    2,979
    Location:
    Wherever I can find the iron.
    Lower the VID the better. It's basically the stock vcore. It could indicate a better clocking chip, but does not guarantee it. Same with undervolting.
    xvi says thanks.
  7. cadaveca

    cadaveca My name is Dave

    Joined:
    Apr 10, 2006
    Messages:
    13,864 (4.53/day)
    Thanks Received:
    6,934
    Location:
    Edmonton, Alberta
    Not exactly.

    Bioses are programmed with TDP maximums. A chip with a lower VID actually is a chip that gets more current from the board(basic stuff...VoltsxAmps=Watts), and THAT'S what makes 'em clock better...there's more current available.


    Also, you cannot make this assumption without first knowing how those chips are binned to receive that lower VID.

    Think back to Core2 "B" stepping chips when they first came out...great clockers, lower VIDs, but ran HOT...they ran hot, because they were programmed to receive more current. They clocked better...more current.
    xvi says thanks.
  8. xvi

    xvi

    Joined:
    Nov 10, 2006
    Messages:
    1,959 (0.69/day)
    Thanks Received:
    1,092
    Location:
    Washington, US
    I might be starting to catch on here. A "good" processor rated at, say, 100 Watts may be tested by the manufactuer and recieve a VID of, say, 1.3v.
    A similar "bad" processor rated at 100 Watts may also be tested and recieve a VID of 1.5v.

    Watts = Volts times Amps(W = V x A), so if we know watts and volts, Amps = Watts divided by Volts.
    Do a bit of math and the "good" processor will run at ~77 amps while the "bad" processor will run at ~66.7 amps.

    ..and since high amperage = stable voltage, high amperage = stability at high power draws? ..which, long story short, equals high overclockability?

    Late Edit: ..but once the processors are on, don't they drop down to a standard voltage? And if the amperage varries from processor to processor, wouldn't two processors at the same voltage have different wattages?
    Last edited: Aug 27, 2010
    Crunching for Team TPU
  9. overclocking101

    overclocking101

    Joined:
    Apr 8, 2009
    Messages:
    2,886 (1.47/day)
    Thanks Received:
    405
    Location:
    vermont
    older chips dont have a vid much like i7's. Also i have seen countless times where someone will have a lower tock stable voltage and not oc as high as some with higher. example. i had 5 i7-860's here for a while for builds etc. and the one that overclocked the highest had a 1.18v stock voltage and there were 3 other cpu's that ran same stock speed but with 1.12v or so. you get my point. there really is no way to tell without doing the deed and even if you hit 5ghz with some theres a chance someone else wont be able to hit that same speed.
    xvi says thanks.

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

Share This Page