E6750 Overclock, need a second opinion

[phantom93]

Im running a E6750 on a MSI P35 Neo mobo, DDR2 800 Wintec AmpX memory. I read the article http://www.tomshardware.com/2007/09/21/sbm_overclocking_the_competitors/ and i have the same case and CPU that they used in that article. So anyways I decided to OC my CPU. 3.2ghz was easy just like in the article so i tried 3.4 which also works however I didnt try any games with it but I ran the windwos assesment test while having speedfan open to moniter my temps. seemed good so i tried 3.5ghz, BSOD!!! yay... then i remembered that in the article they raised voltages so I raised my Mem to 2.1v and NB to 1.45v (currently at that, i raised them by like .05v at a time to be safe), the article recommended using a 1.475v for CPU but I used 1.400 Windows booted up! so i tried crysis and got a BSOD. Now my Mobo has grey, white and red colored numbers for volts. Grey = Safe, White = Performance, and Red = dangerous. upon entering 1.4 I get the white and above using 1.425 it turns red. So obviously 1.475 is red... Now I really want 3.5Ghz but I really don't want to go into the red, and the red could be safe just my BIOS says its not. I am NOT going for the 3.6 and 3.7ghz they achieved in the article as I have stock cooling and i dont have that mobo which i know is 2x betetr then mine or more. Basicly what I'm trying to ask is, Has anyone gone over 1.425v with a E6750 and still have a working CPU? Currently my CPU under load on 3.5ghz gets (it can run the vista assesment but crysis crashes my comp) 47C last i checked using 3.5ghz 1.4 NB 2.1 Mem and 1.4 CPU. Obviously it will be hotter using a 1.45 NB and 1.425 CPU but that gives u guys an idea of my current under load temp. So if any of you have reached 1.45 or higher with a E6750 on stock cooling (if that makes a diffrence) please let me know, let me know ASAP if u did on My mobo the MSI P35 Neo. Thanks in advance

 

[cnumartyr]

It might not be advisable for a 24/7 overclock but Dell is shipping OC'd systems at 1.6 vCore.

I use 1.416 vCore for my Q6600, but I've had it up to 1.6 vCore without issues. It's not going to hurt for short periods, but it may shorten the life of the CPU. As long as temps are kept in check it should still be fine.

 

[phantom93]

It might not be advisable for a 24/7 overclock but Dell is shipping OC'd systems at 1.6 vCore.

I use 1.416 vCore for my Q6600, but I've had it up to 1.6 vCore without issues. It's not going to hurt for short periods, but it may shorten the life of the CPU. As long as temps are kept in check it should still be fine.

Yeah i just wanted to know if it fryd me cpu or not, thanx man

and if that doesnt help would raising the VTT FSB Voltage help? defualt is 1.2 so 1.3?1.4? or would it not do anything

 

[phantom93]

YES i got it, thanx dude, raising it to 1.475v is like totaly making it work.

 

[cnumartyr]

YES i got it, thanx dude, raising it to 1.475v is like totaly making it work.

After you get it working and stable, start reducing the voltage until you notice issues. Anytime you can minimize the voltage for an overclock it's a good thing. 1.475 might of been needed for them, it might be overkill for you and you can "prolong" the life of your CPU by decreasing it until it is unstable and then bumping it up a notch.

 

[phantom93]

yeah actually it just crashed, and i opened BIOS and i realized i had the voltage at 1.4625 on accident, so now its 1.475 and I will see how this goes. So yeah if this works then 1.475 is the min for a working 3.5Ghz OC for me. However at 1.4625 i did 2 loops of crysis benchmark which b4 i coudnt even load crysis and it lasted for about 45min before the crash. so this should work, if it doesnt ill lower to 3.4Ghz

 

[phantom93]

so that didnt work, 3.4Ghz at a 1.425 volts works perfectly, ran both crysis CPU tests and no issues.

 

[cnumartyr]

so that didnt work, 3.4Ghz at a 1.425 volts works perfectly, ran both crysis CPU tests and no issues.

That's not a stress test.

My best advice to you is to LEARN to overclock and not follow some guide.

Read the guide at the top.

 

[phantom93]

I KNOW how to overclock, i am STATING that it's doing sumthing it couldn't earlyer. I read the guide when I built my computer. I posted this thread for a second opinion if I should go above 1.425volts for my Vcore. I like how you say not to follow a guide then recommend I follow a guide...

 

[cnumartyr]

I KNOW how to overclock, i am STATING that it's doing sumthing it couldn't earlyer. I read the guide when I built my computer. I posted this thread for a second opinion if I should go above 1.425volts for my Vcore. I like how you say not to follow a guide then recommend I follow a guide...

If you know how to overclock you'd know about voltages. Or you'd know why it wasn't stable. You raised random stuff because the article did. You had no idea what was causing the instability. Knowing how to go into the BIOS and change stuff isn't knowing how to overclock.

The "guide" at the top is more of a learning tool. I gave you the opinion you wanted, now stop being so damn condescending when I'm trying to help you.

I'm done helping you, enjoy "your" OC!

 

[Mondoman]

phantom, cnu is correct (although not super tactful ). You really should read up on what each of the voltages affects before changing them semi-randomly. For example, why did you boost the DIMM voltage? What is the voltage spec for your specific RAM? Does it have onboard voltage regulation (which essentially makes boosting DIMM voltage harmful to the DIMM)? Obviously, these questions don't matter as much if you don't mind burning out RAM, CPU, MB chips, etc, but it's not clear to me from your posts whether or not that's an issue for you.

Regarding the CPU, it's a good idea to know at least 3 key voltages for your specific CPU:

1) Its "VID" voltage; this is the voltage programmed into the chip on the assembly line after the chip has been tested and indicates what Intel says is the maximum voltage for proper operation of that specific chip. You can find this, for example, on the main display page of CoreTemp near the bottom.
2) The range of VID voltages for that CPU model. To find this, use Intel's SSpec Finder (http://processorfinder.intel.com/Default.aspx). For the e6750 G0 stepping (http://processorfinder.intel.com/details.aspx?sSpec=SLA9V), the range given is 0.85V-1.5V. In theory, this indicates the range of VID voltages for different individual e6750 CPUs, and gives some idea of the designed voltage range for the e6750.
3) Intel's "absolute maximum" voltage for the CPU. This is found in Intel's "datasheet" for the CPU (http://www.intel.com/design/core2duo/documentation.htm). For the e6750, it's 1.55V. The "absolute maximum" voltage is the voltage above which Intel says the processor will almost certainly be damaged.
Voltage (1) is 100% safe.
Voltage (2) is likely to be pretty safe, but less safe the higher you go above your VID value. For e2xxx CPUs, I like to start about 0.075V above the VID value; obviously, OCing to FSBs above 1333MHz is a whole different ballpark, since neither the CPUs nor the MBs were designed for that.
Voltage (3) is for Evel Knievel.

As a final note, it's a good idea to use a number of the free "standard" programs used by OCers; if nothing else, they make comparisons between systems easier.
memtest86+: thoroughly exercises RAM, and doesn't even require an OS installed (bootable floppy and CD images are available).
CPU-Z: reports on CPU, RAM, chipset details and current settings. No temp info.
CoreTemp: reports Tj temps of Intel Core and later designs; also reports some other info, including VID value.
Orthos: This and related programs stress the CPU and/or RAM by calculating Mersenne primes. The most common parameter is the the "blend" test, which stresses both CPU and RAM, but you can also choose small FFTs to stress mainly the CPU. When reporting temps with this, note which parameter was chosen.

 

[phantom93]

I raised the DIMM voltages because if i go over 800mhz on them they get really picky and my system won't boot, anyawys i ran Sisoftware burn in test for my CPU and its stable according to that and after about 1hr, 1hr 30min about now no problems. 3.4Ghz with a 1.45vcore. Im currently lowering the things I rasied .05v at a time. I didn't just jump up to what the article said. I kept all the volts in Grey and tried my mem and NB first to go into the white. I know how to overclok but I am new with CPUs as my old case was a POS, raising my vid cards GPU by 5mhz would imitate a voclano goign off in my comp lol. I can get 3.2ghz w\o changing any voltages. This was mainly just an experiment to see how far i could go and 3.45Ghz is my limit. Due to mobo limitations and cooling. Im probably going to just back down to 3.2 and leave it there. I raised the NB voltage bcause yes the article said to but i did because it seemed to make sense. That a higher OC on a CPU would cause stress on the nb chipset so logicly in my mind rasing that would allow for a higher OC. The mem i knew coudln't handle over 800mhz and I had raised that voltage quite some time ago. I knew from experience that raising the CPU voltage from many articles and forums i have seen allows ppl to get a higher clock. I was reluctent to raise it which is why I made this thread in the 1st place to get other ppl opinions. Ty for the help mond and cnu.

EDIT: I ran othos 5min ago and it crashed so i reverted my voltages and went to 3.2ghz