having a hard time at 245fsb...

[phuang318]

I have a p4c800-e and a 2.6c, 2 sticks of ocz pc4000 gold but using 1 for finding the max fsb.

I am trying to test how far my cpu will go and I'm stuck at 245fsb. I am using Prime95 to test.
Here is what I have done so far:

I have set my fsb on 240, memory settings are at default (didn't really mess with it, but the timings are 2.5-4-4-8, performance mode on standard and memory acceleration at auto). The ratio is set at 4:5 with 320 as the selection. I have ran prime 95 for 18 hours and everything seemed stable. The computer was also laying flat. After seeing the good results, I pushed the fsb to 245, there were no errors for about 3 hours and this was with the a/c on, temps were looking good. I then decided to put the computer right side up with the panels hooked up and it ran for an additional 21 minutes before prime came up with errors. So then I ran the test again and after 20 minutes it failed. I was never able to get it running pass that 3 hours or so.. At first I thought it might be overheating, I then put a smartfan2 on top of the nb and used as3 on the chip. Didn't ehlp.

Next day, I got another 2.6c and it had the same problem as the first. Couldn't run it at 245fsb for more than 30 minutes. I didn't do the 18 hour "burn in" don't know if that actually mattered or not. The other strange thing is that I have noticed that the Vcore would fluctuate from 1.584 to 1.664 but most time it would be on 1.664 (using Asus probe.) Is the 1.584 with an Antec 430 normal when I have adjusted the vcore to 1.6? I am running out of ideas to push this thing past 245.. Someone please help!! I would really appreciate it..

Edit: Memtest86 has been running for 45 minutes with no problems at 245fsb..

 

[infinite012]

You never said whether VDimm is at stock? Try pushing it up to 2.8volts. Also, run 2 instances of Prime95 to see if it is actually stable or not (hyper-threading).

 

[batmang]

make your game acceleration settings these, from top to bottom: auto/normal/auto/disable/disable
should help if their not to set ti that already.

 

[phuang318]

Originally posted by: batmang
make your game acceleration settings these, from top to bottom: auto/normal/auto/disable/disable
should help if their not to set ti that already.

The vdimm is at 2.8, this isn't a abit board, does not have game accelerator. Although, all other settings are set on standard, no turbo enabled..

 

[Budman]

I am trying to test how far my cpu will go and I'm stuck at 245fsb.

If you're trying to find out the max CPU speed then set your ram to DDR266 in the bios,this will take make sure the ram is not limiting you.

After you find out your max cpu fsb you can start putting your ram back up to speed.

 

[stevejst]

I don't think it is the ram, Prime95 fails on CPU not on ram. Given that he says that flat is OK and upright is not my best guess is - heatsink is not installed properly. That fluctuation on the vcore seems a bit excessive but not impossible if that is some junky PSU.

 

[mechBgon]

The Vcore of a Pentium4 varies by design. As Icore goes up, Vcore goes down. Refer to Figure 2.4 on page 26 of this PDF for a graph showing what to expect.

 

[stevejst]

Which makes sense, and it looks linear which make sense too.
His problem, if there is any, is somewhat larger than usual variation of vcore which I find unrelated to the power properties of the CPU since I do not see that variation on my two boards. What I see is a common variation of 0.03V, like in 1.52-155V.
He described a significantly larger variation 0.07-0.08V. So my guess is -- instability of the power supply or/and the board itself.
Besides I see no way to monitor icore or the power output in the BIOS so we can only speculate anyway.

 

[cow123]

i also have an antec truepower 430 - vcore seems to fluctuate a bit for me also. running a barton 2500 @ 2ghz on 1.6v
8rda+
1 stick of 512mb ram @ 366mhz
40gb maxtor 7200rpm
no floppy drive
cd-rw
2 low speed exhaust fans
1 side intake
radeon 9800np

this hardly seems like a power draining setup, yet my vcore fluctuates very excessively:MBM5 readout

edit: vcore is Core 1

 

[bacillus]

for what it's worth, I think that you just have a poor chip for o/clocking.
one of my 2.6's just won't remain stable above 242fsb!

 

[Perplx]

I think its the northbrige it gets quite hot, I have the same board just tape on a 40mm fan on it like me, my ram still stucks though.

 

[stevejst]

I was reading around a bit about this. The board is the problem for two reasons:
1. vcore instability is a problem of p4c800 boards, these variations of vcore are not unusual with these boards and that has an impact on overclocking
2. board lacks active nb cooling which is always a problem for overclocking.

Shortly dump Asus and buy Abit though overclocking to 240 is not that shabby if you are happy with it. Though there are no warranties, as a matter of fact Asus might have better track record on quality than Abit.