• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Pushing my P4 3.06

J

jfb9301

[H]ard DCOTM x4
2FA
Joined
Jan 11, 2005
Messages
1,734
I've got a P4 3.06 SL6S5 Malay. That is a C1 stepping, 130 Micron, HT Northwood. 533 FSB.

Running with stock cooling on a ASUS P4G8X Delux. 1 GIG of Kingston HyperX PC3000 running Dual Channel. So far I've managed 3.357 GHz @ 1.600v (as set in bios) This is powered by a thermaltake silent pure power 480W. I'm using the P4 12volt connection vice the 4 pin molex for extra juice to the board (does this make any difference?). How far can you push this chip (CPU Database @ overclockers.com is down). How far can I push the voltage on a northwood?
So far it is stable at this setup.
Case is cooled by 2 120mm Sunon fans @ 7v. Case temp = room temp no matter what the load is.
Motherboard monitor shows the voltage on the strong side of 1.600... like 1.680
Idle 38C, loaded 57C (Max rating from intel 68C)
Running Prime95 torture test for max load and heat, combined with ATITool pushing my 9600 Pro (lite) which is also overclocked 476/253. No instablilities.

Only real complaints are I cannot seem to reach 3.4GHz stable and no matter what the BIOS settings (rev 1006) no manual adjustment of CAS timings has any effect. CAS timings seem to stay the same @ those reported by SPD even tho the RAM is waaay over the PC2100 (I think it is) required by 533
 
Both connections are 12v and therefore it actually doesn't help. I'd use the dedicated cable designed for it.

Theres a few things to try, settings wise, but you may have simply reached the limit of your CPU. Sometimes that's all you can get. BTW I wouldn't push your voltage to anything higher than 1.65. If you go to 1.7 you'll likely experience SNDS. Sudden Northwood Death Syndrome. Northies don't like 1.7v.

Can you adjust your ram ratios? I'm not familiar with that board but that could help. Sometimes memory voltage increases will help. I could hit 3.4GHz on my 3.06 that I had back in the day. My memory held me back from getting any futher. My specific ram hated being set to DDR333 speeds and hated doing it at any CAS latency other than 2.0. It was Samsung ram in my case.
 
thanks for the heads up on SNDS..... Yikes... this is a borrowed CPU and board.... at least till the tax man gives me enough loot to get a LGA775 on a 925X or 925XE.

Since I was peaking at 1.680 (Yikes, I really do not wish to burn this processor up) unloaded I got worried (voltage would drop to 1.600 under load) so I dropped the BIOS voltage to 1.575 and am giving that another go on stability (I might heaven forbid have skipped over that setting) Idle it hits 1.648 and loaded it droops to 1.552 not exactly stable voltages, but they seem to be getting the job done. If 1.575 is stable I'll raise my Vddr by 0.1v and give 3.4 anther go.

Ratios? I can adjust AGP/PCI ratio if that is what you mean... no adjustment for FSB/RAM

3.357 is a no go at 1.575v only on my 1.600 setting
Raising Vddr did not improve stability here
currently checking 3.335 for stabilty @ 1.575v.....#$^#$^#*$#
currently checking 3.311 for stability @ 1.575v
 
So after a little stability testing and another voltage adjustment, It looks like my choices are:

3.357GHz @ 1.600v Peaking at 1.692v (revised from 1.680)
or
3.311Ghz @ 1.550v Peaking at 1.600v

A little reaserch at intels sight netted me the documentation about the northwood which lists 1.750v as the max voltage that can be applied to this core. So maybe SNDS was from people who set greater than 1.7 and it peaked greater than the max for the core, and just maybe my 1.600 peaking at 1.692 unloaded (full load dropped right down to 1.600) is really OK. I could use some input from you guys on this. I did a forum search on SNDS and it looks like I might OK at the higher voltage.clock I might even be able to push for 1.625 but not much more without overvolting to 1.750.

reguardless 3.311 @ 1.550 is not to shabby and I could just run with that.
 
Ram ratio's are decided by memory speed configuration. DDR400 is 1:1.
 
no ram ratio adjustment available.

finally got a crash on 3357 @ 1.600v
and I crashed on 3311 @ 1.550v

currently testing 3311 @ 1.575v and it looks like everything is good. Guess I've hit the wall for this setup. higher freqs crash no matter what the voltage (shying away from captain-insane-o voltages tho). I'm all about overclocking but not if my machine crashes more than windows98. Only thing left is to chase down why my CAS settings default to SPD. I'm sure I can get them much more aggressive than SPD as they are way over the requirement for 533. I'm headed over to ASUS's site to do some reasearch.
 
You must log in or register to reply here.
Back
Top