The google for yourself. This is exactly why I never provide sources, People ask for them, then refuse to believe them when they have been proven wrong, its pointless trying to prove your right. People hate so much being wrong, they say any source that is against them is bullshit and nit pick. The fact is, if the system can boot, the system will run, There is no otherway around it, If it boots, it runs
The fact is, if the system can boot, the system will run, There is no otherway around it, If it boots, it runs
/jjz dies a little inside. Okay then, clock your system up as high as possible in 10 mhz increments until it bluescreens. Then back off another 10 mhz so you can get it to boot. Then play a new game that you like for 4 hours straight without any crashes while running a prime 95 on one of the cores (I obviously suggest a single-core game so that it is playable). Enjoy crashing in 1-5 minutes.
Yeah, I have pushed mine past 4.2ghz and gotten some benchmarks in, but it's unstable, crashing Prime in a few minutes or seconds. That seems pointless to me. I can live without 2 days of running Prime type of stability, but I don't want erratic crashes either. Rock solid at 4.2.
Speaking of instability, when you've hit a wall (like I may have) how can you tell if it could be your motherboard holding you back? Or your ram? Here's what I plan to try... Tell me if this process of elimination for determining the bottleneck makes any sense:
Test motherboard: Lowering the CPU multiplier and raising the FSB in small increments. Bumping voltage once unstable. Once voltage is increased and instability reached, back it off; max value found. Test ram: Lowering the CPU multiplier and adjusting the FSB and ram divider to increase the ram speed only. Bumping voltage once unstable. Once voltage is increased and instability reached, back it off; max value found.
Can I do something like that to determine the max speeds for those parts? With new ram coming, I'm mainly curious about my Gigabyte motherboard's limit, that I haven't reached nor approached it yet.
That would be a reasonable way to test, yes. It would not completely eliminate whether y our mobo / ram is the cause, or your cpu; however, it would certainly give insight. That is the same method I use to test for walls if I think the OC I am getting is below what the standard is. It's not perfect; however, it's much better then just giving up and saying "oh well, i hit a wall somewhere and I guess I'm happy enough"
Sorry for double-post here (again :() so i'll make it plain & simple : 2813MHz, 1.2250v, 3199RPM, GPU score 12k+ 3DMarks, CPU - 9070 3DMarks. High 3DMark Vantage preset. OT (well almost, since Pi is benchie too) : 27.xxx sec with 1M Pi calculations, with 2.8GHz overclock. Going for 2.9 & hopefully 3.0GHz clock. Stay tuned. P.S. How is this thread will be concluded ? And if not - how long will it last ? Take care tech dudes & overclockers. :rock:
Your score is actually pretty low. I get around 7700 in vantage with my setup (when I had the CPU speed @ 3.4ghz). Your talk of overclocking is tempting me to go for 4+ghz on this chip.
1st of all : wassup, dude ? Long time no see. :rock: Glad to see you too. :) 2nd in line : low for what ? 3.4GHz ? I have 2.8GHz & going to raise it to 2.9 or maybe 3GHz if lucky. BTW : no Xigmatek for Israel & no TRUE for me. Back to stock AMD heatsink, but that one is off topic. Gonna see if i can pull out with stock AMD cooling, 3.3 or 3.4GHz. P.S. Do you mean low 9070 3DMark score for CPU ? I ticked the "Disable PPU" option in Vantage. :cheers: Stay in touch buddy.
No, I mean 12k for two 280 is low.
If anybody's interested, you can see my 3DMark scores here (01-05 right now):
Not bad doc. Im close to you. Its that damn 500+mhz frequency boost you've got on me with the CPU. I need to get a 45nm dually and take you to town =p