![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
LuminaryJanitor
2[H]4U
- Joined
- Oct 26, 2005
- Messages
- 2,340
Made my first overclocking attempt a couple of days ago, and it went pretty smoothly, except that now my system will only post about 60% of the time. Hitting the reset a couple of times always gets it running again, and once it's up, it's stable for 8 hours under Prime95. Not sure what information is relevant, so I'll just give you everything I can think of...
The system:
CPU is a 3000+ Winchester @260x9 = 2340MHz, up from 200x9 = 1800MHz. 33C idle, 44C load.
Vcore set to 1.575, though it's reported as 1.49-1.54.
2x1GB DDR @216MHz, 2.5-3-2-5 2T, Vdimm = 2.6V. It's generic crap, but it survived a Memtest86+ pass and an hour under Prime95 at 225MHz (running 1:1) with these timings.
Motherboard is an MSI K8N Neo4 Platinum (nForce4). Chipset voltage at 1.6V, HTT at 4x.
PSU is a CoolerMaster eXtreme Duo 650W (active PFC version).
The problem:
According to the motherboard's LEDs, it hangs on "Processor Initialisation" a bit under half of the time. Posts fine after at most two reboots.
Still happens with the multiplier as low as 5.
Still happens with the memory at 187MHz, 3-5-5-15.
Still happens with HTT at 3x.
Still happens with Vcore set to 1.601, Vdimm at 2.7.
Still happens with chipset voltage at 1.65. I'm reluctant to push this further, since I'm not sure what's safe or what the temp is (is it the one labelled "Motherboard" in Everest? 36C at the moment), plus my x1950pro is sitting about 5mm in front of the fan. In any case, dropping it to 1.50 doesn't make things any worse (though this causes a Prime95 failure within a minute), so I don't think this is behind it.
While I've got your attention... What might be holding back my OC? I seem to have a bit of room to move with a 44C load temp, but I think I've tried everything I listed above and I can't get it from 260 to 264. Is 0.05V too much Vcore fluctuation? Which of the CPU / GPU (x1950pro, 648/1485) is better off sharing a 12V rail (18A each) with the motherboard?
The system:
CPU is a 3000+ Winchester @260x9 = 2340MHz, up from 200x9 = 1800MHz. 33C idle, 44C load.
Vcore set to 1.575, though it's reported as 1.49-1.54.
2x1GB DDR @216MHz, 2.5-3-2-5 2T, Vdimm = 2.6V. It's generic crap, but it survived a Memtest86+ pass and an hour under Prime95 at 225MHz (running 1:1) with these timings.
Motherboard is an MSI K8N Neo4 Platinum (nForce4). Chipset voltage at 1.6V, HTT at 4x.
PSU is a CoolerMaster eXtreme Duo 650W (active PFC version).
The problem:
According to the motherboard's LEDs, it hangs on "Processor Initialisation" a bit under half of the time. Posts fine after at most two reboots.
Still happens with the multiplier as low as 5.
Still happens with the memory at 187MHz, 3-5-5-15.
Still happens with HTT at 3x.
Still happens with Vcore set to 1.601, Vdimm at 2.7.
Still happens with chipset voltage at 1.65. I'm reluctant to push this further, since I'm not sure what's safe or what the temp is (is it the one labelled "Motherboard" in Everest? 36C at the moment), plus my x1950pro is sitting about 5mm in front of the fan. In any case, dropping it to 1.50 doesn't make things any worse (though this causes a Prime95 failure within a minute), so I don't think this is behind it.
While I've got your attention... What might be holding back my OC? I seem to have a bit of room to move with a 44C load temp, but I think I've tried everything I listed above and I can't get it from 260 to 264. Is 0.05V too much Vcore fluctuation? Which of the CPU / GPU (x1950pro, 648/1485) is better off sharing a 12V rail (18A each) with the motherboard?