Once you've lived at 3.2 there ain't no going back!

Hulk

Diamond Member
Oct 9, 1999
4,998
3,428
136
A few months ago I built my new system (see sig) on the advice and posts from everybody on this board. Thanks to everybody.

A few days ago my system gave the 1 long 3 short beeps and would not post.

I had to clear the CMOS and pull a stick of memory to get it going.

Then I simply could not get back to the stable overclock I had. Once I was in Windows it was solid but it would fail on post 4 out of 5 times. I tried a million different things but just couldn't get it happening.

For a few days I was running stock settings because I got tired of fiddling around. And let me tell you there is a difference from 2.13 to 3.2! It was like somebody put the brakes on my system. Just little things like opening applications, switching apps, opening new web tabs, all the little things were just a bit slower. Just a bit, but noticeable.

So I took a chance and updated the BIOS from 614 to 1004, the latest BIOS for my board.

Well I have to say this made a HUGE difference. At this point every setting I've tried has worked. I'm where I was at before but with the chipset and memory voltages lower.

It's just my experience but the 1004 BIOS is a winner in my book.

------------------------------UPDATE-------------------------------------
Since I upgraded the BIOS on 1/30 I have not had one single problem with my system.
I know some guys were concerned about electron mitigation so I just wanted to update this thread and let everyone know everything is still fine over 2 months later. Vcore is still set at 1.35 in BIOS, actual is 1.288 idle and a little lower under load.
 

SparkyJJO

Lifer
May 16, 2002
13,357
7
81
Yeah the difference between 2.8GHz and the stock 1.8GHz of my opteron 165 is really noticeable as well
 
Oct 4, 2004
10,515
6
81
Glad to see you are back in business but any idea why an OC just magically stopped working overnight? It might be useful to know (for the future) 'why' something happened, even if you have solved the problem.
 

Cheex

Diamond Member
Jul 18, 2006
3,123
0
0
Originally posted by: theprodigalrebel
Glad to see you are back in business but any idea why an OC just magically stopped working overnight? It might be useful to know (for the future) 'why' something happened, even if you have solved the problem.

Good point.

@ OP: It would be good for the readers (like myself) who have or plan to get a C2D (like myself) or a mobo like yours...to have your problem and solution as a decent real world example to go by.
 

Cogman

Lifer
Sep 19, 2000
10,283
134
106
I think I read in an article (tomshardware oc article) that OCed processors degrade over time, so if you push it to the limit it will become unstable faster then it does running at stock speeds. If that is the case, I guess you just have to bump up the voltage.
 

Imyourzero

Diamond Member
Jan 21, 2002
3,701
0
76
Originally posted by: Cogman
I think I read in an article (tomshardware oc article) that OCed processors degrade over time, so if you push it to the limit it will become unstable faster then it does running at stock speeds. If that is the case, I guess you just have to bump up the voltage.

I don't think I've ever heard of that being a problem here on AT, and needless to say there are a LOT of overclocked CPUs here.

I'm not saying it couldn't happen, but most people probably don't keep their CPUs long enough that it would become an issue. I wonder if this has been discussed on OCForums?
 

StopSign

Senior member
Dec 15, 2006
986
0
0
Originally posted by: Cogman
I think I read in an article (tomshardware oc article) that OCed processors degrade over time, so if you push it to the limit it will become unstable faster then it does running at stock speeds. If that is the case, I guess you just have to bump up the voltage.
All processors degrade over time. Overclock just degrades them faster. However, the difference is negligible because not a lot of people here are keeping the same processor for over 5 years.
 

Hulk

Diamond Member
Oct 9, 1999
4,998
3,428
136
Originally posted by: theprodigalrebel
Glad to see you are back in business but any idea why an OC just magically stopped working overnight? It might be useful to know (for the future) 'why' something happened, even if you have solved the problem.


That is a good question and I honestly don't know.

All of a sudden I turned it on and it gave me the 1 long 3 short beeps and didn't post. Next try it posted so I upped the voltage to 1.375 thinking I was on the line at 1.35.

Then I restarted to test and it would not post. That's when I had to clear CMOS and pull a stick to get it to post. At that point it was very finicky about posting at any settings other than stock. I mean I had it going at 6x400=2400 okay finally.

So I decided to stop messing around with it and upgraded the BIOS.

As I said it's been two days now of heavy stress testing and no a single problem.

Now in the interest of full disclosure with my old overclock I had NB,SB, and FSB Termination voltages on auto, now I have them set manually. That could be causing some difference. But my overclock was working with BIOS 1004 out of the box. Since it was so solid I decided to fine tune things a bit. As I said with the earlier BIOS it was so finicky that I was reluctant to change anything once I got it going because of the hassle to restart.

I hate removing the video card to clear the CMOS (remove battery) and pulling the RAM.

I *liked* the P5B with the 614 BIOS. I mean it did work perfectly at stock but was a little picky with the oc settings

Now with the 1004 BIOS I'm loving it. They seem to have worked out the things that were causing my issues.

I'm running my memory at 4-4-4-12 at 2.0 Volts now. I ran 8+ hours of memtest at those settings with no errors.


 

Boyo

Golden Member
Feb 23, 2006
1,406
0
0
:thumbsup: Good work on getting your PC back up and running full speed. I can imagine the difference in speed.
 

Thor86

Diamond Member
May 3, 2001
7,886
7
81
Increasing voltages degrade your processor over time faster, not overclocking.
 

Dolorous Dave

Senior member
Feb 23, 2004
317
0
0
Originally posted by: Thor86
Increasing voltages degrade your processor over time faster, not overclocking.

Apparently, throwing your processor against the wall also degrades it's performance over time. Go figure.
 

LintMan

Senior member
Apr 19, 2001
474
0
71
Originally posted by: Hulk
I hate removing the video card to clear the CMOS (remove battery) and pulling the RAM.

You might not have to do that. The P5B manual mentions a sort of OC BIOS reset feature to help you out if you render the system un-POSTable... Basically, if your settings prevent the system from POSTing, turn off the power supply switch (meaning the switch on the PS itself, not the case front switch), and then turn it back on, and it will come up with default settings. (I'm not sure if you need to leave it off for a particular length, I waited 10 secs or so).

Anyway, this worked for me last night: I installed the 1004 flash, and upon reboot it complained about messed up settings and prompted me to hit F1 to reset them. But for some stupid reason, I suddenly thought I was supposed to power cycle after flashing the BIOS (like on my old MB), so I just turned off the power (at the case switch). After that, the system wouldn't POST anymore. So I tried the power switch cycling trick, and it worked great - the system POSTed and gave me the reset BIOS prompt again. (I don't think it clears the BIOS, it just allows you to boot despite bad BIOS settings)

 

Hulk

Diamond Member
Oct 9, 1999
4,998
3,428
136
Originally posted by: LintMan
Originally posted by: Hulk
I hate removing the video card to clear the CMOS (remove battery) and pulling the RAM.

You might not have to do that. The P5B manual mentions a sort of OC BIOS reset feature to help you out if you render the system un-POSTable... Basically, if your settings prevent the system from POSTing, turn off the power supply switch (meaning the switch on the PS itself, not the case front switch), and then turn it back on, and it will come up with default settings. (I'm not sure if you need to leave it off for a particular length, I waited 10 secs or so).

Anyway, this worked for me last night: I installed the 1004 flash, and upon reboot it complained about messed up settings and prompted me to hit F1 to reset them. But for some stupid reason, I suddenly thought I was supposed to power cycle after flashing the BIOS (like on my old MB), so I just turned off the power (at the case switch). After that, the system wouldn't POST anymore. So I tried the power switch cycling trick, and it worked great - the system POSTed and gave me the reset BIOS prompt again. (I don't think it clears the BIOS, it just allows you to boot despite bad BIOS settings)

Ah! Nice tip. Thanks.

My SATA CD/DVD R/RW arrived today and I just installed it. Now I'm IDE free. Are there any drivers I can remove that I don't need anymore? I'm not running RAID.

 

idiotekniQues

Platinum Member
Jan 4, 2007
2,572
0
71
i notice significant differences in speed on my 6400 on my over 1ghz 24/7 oc, with such a minimal voltage increase that it should barely affect the chip.