Results 1 to 6 of 6
  1. #1
    New Member
    Join Date
    Jan 2008

    OC'd IP35 Pro slower than stock???

    I built the system in my sig about a month ago. I was able to get to a little over 3GHz without any trouble. In order to see if it was worth it to go further, I started running the trial version of PCMark05 with different OC settings (set through uGuru) to see how much I was gaining by pushing further (FSB changed, but nothing else). Here is what I found:

    Run 1: FSB at 322 (2.9GHz), PCMark score 8189
    Run 2: FSB at 302 (2.7GHz), PCMark score 8235
    Run 3: FSB at 280 (2.5GHz), PCMark score 8413

    Higher scores are better, so what's up with this??? If I believe the results, I am better off with stock settings. What am I missing here?

  2. #2
    disk11's Avatar
    Join Date
    Dec 2003
    Heatware Profile
    What did you set the memory timings and ratio to for each of the OCs? I dunno how much memory bandwidth affects PCMark, but that could be one thing.
    My Heatware
    OCforums Metal Club member # 665
    OCforums Musicians Club member # 665
    Main/Gaming: i7 920, Noctua U12P-SE, Asrock x58 Extreme, 6GB pc16000 Gskill Pi, Powercolor 7870, 256GB Crucial M500, 160GB Intel 320, Lite-On DVD burner, 600W OCZ modX
    File server: AMD e-350, Asus E35M1-M Pro, 4GB Crucial, OCZ Vertex 3 (OS), 3x Toshiba 1TB RAID 5, 2x 2TB WD Green in RAID 1, Antec 900, Corsair 520HX

  3. #3
    New Member
    Join Date
    Jan 2008
    Memory was running 620 at 1:1. Remember it was the same for all runs as was everything except FSB.

  4. #4
    Overclockix Snake Charming Senior
    Alt OS Content Editor
    Stratus_ss's Avatar
    10 Year Badge
    Join Date
    Jan 2006
    Ontario Canada
    Author Profile Folding Profile Heatware Profile
    are the timings on the ram set to auto?
    if they are that would explain it. As the fsb increases the board will automatically relax the timings. Relaxed timings mean slower throughput for benchmarking.

    In practical use the stick about ocing a c2q states that there is little difference for having tighter timings but that it really shows up in benchmarks
    Those wishing to contribute to the AltOS section front page please PM me.

    Leviathan41: "Playing with a controller because the mouse and keyboard is too easy is like racing with a bicycle because a car is too fast."

    Chris Fisher "Windows is like Administration with training wheels"

  5. #5
    Quote Originally Posted by ManicRC View Post
    Memory was running 620 at 1:1. Remember it was the same for all runs as was everything except FSB.
    On intel boards fsb is linked with memory speed. If memory was running 620 in 1;1 ratio then you fsb should have been 310. You decrease/increase fsb and it decreases/increases memory speed. Make sure you have memory multiplier set to 1:1 not running auto. Also set you timings manually.

  6. #6
    Senior Member jason4207's Avatar
    10 Year Badge
    Join Date
    Nov 2005
    Concord, NC
    Also, reboot the PC b/n runs.
    MAIN RIG: 3570K @ 4.5GHz 1.272v H100i Extreme4-Z77 8GB-G.Skill @ 2200MHz 9-11-11-28 1.60v Fractal Design Arc
    GTX670_DCuII_TOP @ 1358/7244 1.212v Xplosion-DTS 480GB-SSD 1TB-Black Seasonic_X-750
    Panny 42S2 1080P Plasma || PLV-Z4_720P_Projector_95" Paradigm_6.1_DTS-ES
    G9_Mouse Logitech_Illuminated_KB Logitech_G27_Wheel+Microsim_Racing_Pod ***HEAT***

    MAIN OFFICE: G3258 4.6GHz 1.271v H70 Z97-Pro4 8GB 2200MHz 9-10-10-29 1T 1.575v
    GTX660 1241/6772 256GB-SSD Antec300-430W Custom Triple Screen 4998x1080

    BEDROOM: Q9550 4GHz TRUE Max2Form 4GB 1150MHz 5-5-5-15 2T 1.8v
    280GTX 240GB-SSD Thermaltake Case-750W Hanns-G 28" 1200p

    unRAID FILE SERVER: E5200-M0 @ 2GHz 0.856v MSIP43Neo3-F 4GB-Corsair @ 800MHz CM590 Antec550W
    20TB-Parity_Protected_Storage / 1TB-Cache / 2TB-Parity / 10x2TB-Storage

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts