-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Northwood 3.0GHz MAX?
- Thread starter slix_88
- Start date
- Joined
- Feb 20, 2001
Hi, welcome to the forums.
The maximum stable frequency for a CPU is not strictly defined for each CPU - it depends on many things including:
For a P4 on air or water, I think 3.4 to 3.6 shouldnt be too hard, up to 3.9 to 4.0GHz with good watercooling and a good chip. Perhaps over that with phase change.
The maximum stable frequency for a CPU is not strictly defined for each CPU - it depends on many things including:
- Week and Stepping - some batches of CPUs are far better for overclocking than others.
- Cooling. Water or phases change cooling can help increase the overclock.
- Voltage. More voltage often means more stability, however too much can cause damage or overheating.
For a P4 on air or water, I think 3.4 to 3.6 shouldnt be too hard, up to 3.9 to 4.0GHz with good watercooling and a good chip. Perhaps over that with phase change.
- Joined
- Jan 12, 2001
- Location
- Kansas, USA
Most 3.0C should do 3.4 gig or more on default 1.525v. It's the luck of the draw, each one is slightly different. Generally you'll need to change the voltage in the BIOS. What cooling do you have that runs below room temp?
Last edited:
- Thread Starter
- #5
I am purposely using CPUCool, as my bios does not permit me to change any FSB or Voltage settings. It is for that reason why I am using CPUCOOL.
I have also encountered another problem which involves CPUCool. I had previously fine tuned the FSB to 215 (3.15GHz) and the reading in both WCPUID and CPU-Z were 3.15GHz. However, performance wise, I didn't see any performance leap. Not a secon in SUPERPI nor a comparison difference in aquamark and Sisoft Sandra. I certainly know my ICS, and have the matching one in CPUCOOL (that isn't an issue). What is it?
I have also encountered another problem which involves CPUCool. I had previously fine tuned the FSB to 215 (3.15GHz) and the reading in both WCPUID and CPU-Z were 3.15GHz. However, performance wise, I didn't see any performance leap. Not a secon in SUPERPI nor a comparison difference in aquamark and Sisoft Sandra. I certainly know my ICS, and have the matching one in CPUCOOL (that isn't an issue). What is it?
- Joined
- Oct 26, 2004
- Location
- Bay Area, CA
Well, I have a Northwood 3.06 P4 and I have been able to get stable OC's up to 3633MHz on air which is pretty impressive in my opinion. I have found that adequate voltage is key, and of course, the cooling to accompany it.
- Joined
- Apr 9, 2001
Foxie3a said:15FSB isn't enough to start noticing changes, atleast not drastic ones.
If you pick out your hardware carefully, you can do nice things, even on stock cooling...
On my box it says my voltage is 1.55 max. Although it runs fine at 1.47v.
Roger,
The northwoods have multiple VID's. The max VID for both a 3.0C, 3.2C and 3.4C are 1.55v. That's why it says "1.55 MAX" on the box.
The reason yours runs fine at 1.47v is because 1.47v is one of the VIDs supported by the chip. So you are not undervolting the chip at all--you are within the lower VID spec for your CPU. 1.475v is the minimum VID, 1.55v is the max VID.
Oddly enough, for a P4 3.4 EE, the vid range is 1.525-1.600v. Which is pretty strange (the vid voltages are bumped up a range of .075v).
Yet, for a 3.2EE, the vid range is the same as on the 3.0C, 3.2C and 3.4C (!), -- 1.475v-1.55v (!).
I wonder why the vid range difference between 3.2EE and 3.4EE (1.525-1.600v)...
At first my 3.0c was stable @3.6 on stock voltage, after a few crashes (5) I got her stable with 1.55 volts. My XP-90 keeps temps under 55c in a case with only 2 80mm fans 
Can someone exsplain more about that voltage thing? Like my bios says 1.42 V-core (yet it's set to 1.55) and MM5 shows 1.47 to 1.51 at idle and 1.46 to 1.50 under full load. Funny how the V-core never shows 1.55 or even 1.52 for that matter. I'm thinking if I set the V-core to 1.59 it will never actually see more than 1.55 (the max) and get her to 3.8 
I had her at 3.9 at 1.59 but crashed loading Doom3 
Can someone exsplain more about that voltage thing? Like my bios says 1.42 V-core (yet it's set to 1.55) and MM5 shows 1.47 to 1.51 at idle and 1.46 to 1.50 under full load. Funny how the V-core never shows 1.55 or even 1.52 for that matter. I'm thinking if I set the V-core to 1.59 it will never actually see more than 1.55 (the max) and get her to 3.8 I had her at 3.9 at 1.59 but crashed loading Doom3 
- Joined
- Aug 19, 2004
- Location
- Florida
slix_88 said:
Is that room+8 at idle and room+20 at load? If so then it makes sense (about 32 C and 44 C.)
My 3.0c gets to 3.45 GHz (230 fsb) with air cooling at stock voltage (1.525 v) and is prime95 and Far Cry stable. As was said, each chip is different.
