Need some advice OC'ing 3200+ .90nm

[bigpgo]

Sorry for the noob question, but I couldn't find this on the forum: in BIOS, I have

VID [1.500v]
CPU Voltage [By CPU VID]

To overclock the cpu, (after raising the FSB) should I raise "VID" and keep "CPU Voltage" the same, or change "CPU Voltage" to something like +3.3%, keeping "VID" the same? What's the difference between the two? Thanks.

 

[Duvie]

I believe it is one or the other.....At least in my case that is how it acts as ican set it to 3.3% and get 1.47v no matter if I have 1.45,15, or 1.55v set at same time in the other seting...

 

[bigpgo]

So I upped voltages and such, setting the FSB to 220 - prime95 ran for 7 hours no errors.
I then increased the voltage some more, setting FSB to 240. At this speed, Prime95 is failing within 2 hours.

Here are my settings:

High Performance Mode [Manual]
Aggressive Timing [Disabled]
Dynamic OC [Disabled]
Spread Spectrum [Disabled]
HT Freq [4x]
Cool&Quiet [Disabled]
CPU Ratio [x10]
FSB [240]
AGP Freq [67]
CPU VID [1.550]
CPU voltage [over VID 5.0%]
Mem Voltage [2.75 V]
AGP Voltage [1.55V]

For memory, I have 2T, 3-3-3-10, MAX Memclock: 166 (just playing it safe for now)

Any help appreciated.

 

[Markfw]

change the HT freq to 3x, and the Memory speed to 166 (not the fsb, leave that at 240)

Edit: I see you max memclock. so just try the ht 3x. I also you realized 240x4=960, so that should be good also. Try new bios.

 

[bigpgo]

Thanks, I'll try that. Btw, are my voltages okay (still not crystal clear on what the 2 settings need to be set to)

 

[bigpgo]

Ok, just changed these settings:

HT [x3]
CPU VID[1.525]
CPU Voltage[over VID 8%]

everything else same as above, just started running Prime95.

BTW, I checked my "vcore" under HW Health section of the bios - it fluctuates between 1.53 and 1.55 volts. What gives? I thought vcore = 1.525v + 8% = 1.647v ? Are my core readings wrong? Thanks

 

[RealityTime]

Originally posted by: bigpgo
Ok, just changed these settings:

HT [x3]
CPU VID[1.525]
CPU Voltage[over VID 8%]

everything else same as above, just started running Prime95.

BTW, I checked my "vcore" under HW Health section of the bios - it fluctuates between 1.53 and 1.55 volts. What gives? I thought vcore = 1.525v + 8% = 1.647v ? Are my core readings wrong? Thanks

dude on the .90nm winchester just keep voltage at 1.55 in bios forget the over % stuff for now. see how far you can push it with that. You're using the same air cooling i am with the alcu zalman. I have the 130nm newcastle 3500+ it ocs to 2.6 with the zalman, but is a bit too hot for me, so I keep it at 2.5. Try these settings...

fsb 240
cpu multiplier: 10
htt: 4x
cpu voltage: 1.55
cpu overdimm: by cpu
in ddr config set ddr to 166, timings to 1t, and whatever your mem will take.. I really think your ram might be able to do 1:1 though (as in mem at 200, with relaxed timings, try mem 200, 2.5-3-3-10-1t) otherwise with the 166mem speed in dram config set it to your 2-2-2-10-1t or 2-2-2-8-1t, as your ram will be at 204mhz with the 240fsb speed which is close to your rated timings by 4mhz over. If system is stable with those settings try ramping fsb to 250 and again check for stability.

 

[vmaddogv]

Running same processor and mobo. Try the htt at 3x, fsb to 240, cpu mult at 10x, cpu vid at standard, vid at 5.0% mem at 2.7, stock agp, pci/agp to 67. To leave ram out of picture, run at default timmings at 133. After thats stable increase to 166 then to 200 loosening timings.

 

[bigpgo]

So it looks like I didn't get a very good CPU, and I prolly won't be able to RMA it because it doesn't overclock as well as the others. I tried to lower voltages as you guys have mentioned and it's not Prime95 stable.
Currently, I'm at:

HTT [3x]
FSB [240]
CPU VID [1.525]
CPU Voltage [+8%] (I guess that's 1.647v effective)
(memory out of the picture at 133)

Prime95 has been running for 10 hrs and I think it will continue no problem.
My temps are: 34 / 52 Celsius.

Should I keep what I have, or scale back the FSB a bit to be able to lower the voltage, or keep trying to raise the FSB to the limit at current voltage? Thanks for all the help so far.