can someone guide me step by step to OC my cpu?

Raptor_Jesus

Newbie
Joined
Jan 3, 2008
Messages
210
Reaction score
0
hey guys can someone give me stet by step advice on how to do this?
is there a desktop tool I could use?

"The factory-set multiplier is 13x"

so I just select the next step in the bios which would be 14x? and see how it runs?
 
jesus christ It just jumped from 2612Mhz to 3015.6Mhz



D:

will this make a big differnce with games?
 
hey guys can someone give me stet by step advice on how to do this?
is there a desktop tool I could use?

"The factory-set multiplier is 13x"

so I just select the next step in the bios which would be 14x? and see how it runs?

I strongly advise you to read the following:
http://www.overclockers.com/topiclist/index20.asp
http://www.pcstats.com/articleview.cfm?articleID=1804
http://www.xbitlabs.com/articles/cpu/display/oc-guide.html
http://www.overclock.net/overclock.php?file=articles/overclocking-guide.htm

Those are just a few of the thousands of sites that deal with "How-To" guides on overclocking. If you understand the basics, then overclocking becomes much less scary. Also, not having your PC in front of me to play with, makes it hard for me to be able to tell you how to overclock your hardware & what to do in the even that you push a setting too far & thus making your PC not boot.

I would advise that you read the owners manual for the motherboard thats in your PC, to learn what performance options it might have. Also lean how to reset your Bios, also known as clearing your CMOS. This step will need to be done in the event that you make a change that your hardware does not approve of & decides to crash.

Please list out exactly the type of RAM your have installed, how much, along with the make & model of your CPU as well as the type of motherboard.

It's 1:12am, I just got home from work & so i' beat. In the mean time, look over those sites I posted to & read up a little on what all is involved. I'll post more once I wake up.

-MRG
 
Congratulations for overclocking your cpu, I wouldn't have the courage to do it myself and I never did actually. But calm down, my friend, people will answer to your questions eventually.

By the way, how are you running exactly, the settings (high, very high or whatever it is), the FPS etc. I want to know because I will eventually have a computer almost exactly like yours, except with Core 2 duo E6850 @ 3ghz and some motherboard and HD differences, I want to know how it runs for obvious reasons :thumbs:
 
everything on high except for shadows in Crysis 1400x900res
everything on high CoD4 max AA and AF also 1400x900res :)
 
everything on high except for shadows in Crysis 1400x900res
everything on high CoD4 max AA and AF also 1400x900res :)

Did you Prime-test it?

Your system must be prime stable for minimum of 8hrs in order for your OC results to be valid.

For Dual-core systems; Both cores must be tortured at the same time, simultaneously.

Post back here with Prime95 torture results and Pics plus 3Dmark06 benchmark results.

Otherwise, you are OC-ed but not stable. ;)

Good Luck.
 
Your system must be prime stable for minimum of 8hrs in order for your OC results to be valid.

Post back here with Prime95 torture results and Pics plus 3Dmark06 benchmark results.

Otherwise, you are OC-ed but not stable. ;)

Good Luck.

I'm sorry Mr.Fife, but I very much disagree with that last comment. One program alone will not determine if his system is stable or not. Now if he overclocks his system & he has problems, thats one thing, but if he overclocks it & has no problems, it really does not matter what a program will say otherwise. Stress-Testing for 8hrs straight? Whoever here plays for 8hrs straight any game that fully pushes their hardware to the limit, is a person that needs to get off their a$$ & go spend some time in the sun. :) It does not take 8hrs to detect if there is a performance issue with any PC after you have tweaked a few settings.

My old Test Setup:
P4 1.60GHz OC to 2.40GHz w/air Cooling
Epox 4BEA Mobo
512MB Generic PC-3200 DDR

For years I have heard nothing but good things about Prime95. About how if you fancy yourself a Tech guy, that you used P95. Then years ago when I first started getting into Overclocking, I did what everyone advised me to do & that was to "Stress-Test" my system with Prime95. So I did & within minutes, I got error messages stating that my CPU Overclock was not stable or something to that effect. So I tried it again & same error message. I thought that was odd, as at the time, I had zero problems playing any of the games that were installed on my system. I never had any hardware issues & I overclocked the crap out of that old CPU. We're talking an OC of 800MHz too. I have long since installed that Epox mobo into a PC I made for a friend years ago (Which is still up & running I might add) I still have the old 1.60GHz PC CPU as well that I have used as a Test CPU when I work on older systems for my clients.

Now that the subject of P95 has come up, i'm curious if my results would be the same now, as they were then if I was to test my settings that I currently have my E6400 OC to???

Ok, to put some weight behind my above statements, I just finished downloading Prime95 & i'm gonna run it. Now you have got me curious...

-MRG
 
OK. Your point is well made.

There are two types of Oc-ers out there.

A) The carefree OC-ers who OC and don?t care about validating it.
B) And, the conservative OC-ers, who like to validate their work and will not settle down till they get their satisfaction of their accomplishment, also who like to reach the ceiling; of how high will any given chip will OC.

Sure you can raise FSB here, multipliers there, volts here, volts there, and you may not experience instability while playing a game, to the point of BSOD, as long as you can boot system but, that doesn't necessarily mean you're stable, instability has many different symptoms, which the system may experience; from unstable user program, to driver issue, to BSOD. So, you're either stable or, prime-stable.

When I first ran torture-test on the system in my Sig; P95 did give me an error after few minutes of running. Few tweaks here and there did take care of the problem, and p95 did do the test for 13hrs straight without an error.

Also P95 is good way to determing whether you're overheating. When under stress, P95 will halt operations if CPU temp is out of range.

2a.jpg
 
LMAO!!

While it's been years since I last ran P95 & so I can't remember exactly word for word the error message, Looks like I got it again. This is the results of the test. I have had this PC OC'ed to 2.41GHz for close to 2 years now, or ever since the Core 2 Duo E6400 was released. I have had zero performance issues due to hardware instability.

[Wed Jan 09 00:42:59 2008]
Self-test 1024K passed!
[Wed Jan 09 00:58:25 2008]
Self-test 8K passed!
[Wed Jan 09 01:13:34 2008]
Self-test 10K passed!
[Wed Jan 09 01:29:12 2008]
Self-test 896K passed!
[Wed Jan 09 01:44:39 2008]
Self-test 768K passed!
[Wed Jan 09 01:59:55 2008]
Self-test 12K passed!
[Wed Jan 09 02:15:02 2008]
Self-test 14K passed!
[Wed Jan 09 02:30:22 2008]
Self-test 640K passed!
[Wed Jan 09 02:42:40 2008]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.

My CPU-Z
http://img205.imageshack.us/img205/5200/mrgocsw4.th.jpg
http://img48.imageshack.us/img48/7775/megoc1fk5.th.jpg
http://img165.imageshack.us/img165/5017/mrgoc2uw1.th.jpg

-MRG
 
I can't really see your pictures, they're too small.

I've experienced that error a few times, but my hardware was really not stable and I was getting a lot of BSODs.

It really is a good tool to determine stability. Though, I'm not sure what is going on in your situation.

Right now my Q6600 is P95 stable @ 3.2 GHz for 8+ hours. I had to cut off the program, but some say it's meaningless unless you test for a full 24 hours.
 
I can't really see your pictures, they're too small.

I've experienced that error a few times, but my hardware was really not stable and I was getting a lot of BSODs.

It really is a good tool to determine stability. Though, I'm not sure what is going on in your situation.

Right now my Q6600 is P95 stable @ 3.2 GHz for 8+ hours. I had to cut off the program, but some say it's meaningless unless you test for a full 24 hours.

8hrs is enough for most people. And if nothing happens within the first 8hrs (P95 halting operation) you can almost bet, nothing will happen for 24hrs.

If your OC setting is accurate, voltage and bus frequency is set, VGA and memory bus is locked, multiplier is on the acceptable range, and you made all the checks and balances accordingly; some of the other things (Outside of your OC settings), which could halt P95 torture, could be:
A) An overheating component.
B) A poor PSU; either under-volting or over-volting.
C) Background running programs

When executing P95, you must allocate certain amount of memory for it to occupy, while in operations. If you allocate the entire system RAM to P95 you're background running programs and OS will start to become unstable and force P95 to halt.

Most of the time, a poor PSU will halt P95.
Once you rule out the entire would-be suspects; Check the PSU efficiency.

A poor PSU will not cause the system to crash with BSOD, because, your CPU and memory, most likely won't need that much power from the PSU, for everyday use, even when you're playing games, and your CPU won't reach the extreme and intense area with other programs like it will with P95, for 8hrs, where the PSU becomes crucial to supply clean, stable and sufficient current to your system for 8hrs.

Check your cooling system and CPU/MB temperature while P95 is running. P95 will halt if CPU Temp is out of accepted range.

Good Luck.
 
I totally didn't even think to factor in the quality of my PSU. I have no doubt that it is a cheap POS. I pulled it out of a clients PC over 2 years ago. It's a 450w generic brand. The sticker on the side is all worn & I can just make out the 450w writing.

Sucks to be poor. Wish I could get a better one. Now that I mention it, 2 of the power connectors that are coming out the back are dead. I can't believe I made such a newbie mistake...

Sigh... "God, please allow me to win the lottery, so that I may buy a new PSU." :P

-MRG
 
PSU is absolutely vital part of your system, The Heart (If you like to call it) of your system, when CPU, RAM and video card are the brain and the organs, and system bus are the arteries, vessel and veins of your system.
So, yeah, most systems fail P95 because of poor PSU. The other thing I forgot to mention is the type pf RAM people like to OC with. You definitely can't use value-RAMs if you intent to OC, because, most likely Value RAMs will fail when torturing or you would like to benchmark the system.

I once had an HP Pavilion ran P95 on stock setting, guess what, It failed. lol.
 
Is anyone here familiar with AMD64 overclocking?

I've been thinking about Ocing my system a little bit to see how big of a gain I can get. I'm not planning on changing my CPU's stock cooler (it's a San Diego 3700+) so I'm only looking at trying to OC a bit.

My stock speed is 2.2ghz and I was thinking of trying to OC it to ~2.5 if possible. What worries me is all the stuff I've read while researching overclocking, and about how the overclock affects your entire mobo along with any PCI/PCI-e/Ram you have on your system.

Anyone have any idea or tips to give me while overclocking my system?

Here are the specs:

amd64 3700+
ASUS A8N5X
Ram: 1 stick of DDR400 PNY, 1 stick of 512 DDR266, 1 stick of 256 DDR400 and another 256 of DDR333 (And here's another question, do you guys think I'd get better performance just throwing out that DDR266 chip out or would the higher amount of memory counteract the fact that my PC3200 stick is running slower because of it?)
Vidcard: nvidia 7600GT (Core: 610mhz, Mem: 770 mhz)
HD: WD 320 SATA/300 7200 rpm
 
My main concern with the ram and overclocking is that wouldn't slower memory make the overclock less effective? I mean, as far as I gather, it's better to have 1:1 ratio between HT and my RAM speed, right?

If I used the 2GB of ram at 2100 speeds like I have right now, that means the RAM behaves like DD 266, so I'd have to use a 4:6 ratio for my RAM, correct? How does doing that affect my computer?

I'm basically looking to overclock some, but at the same time I definitely want to do it the smart way.

Also, how does overclocking affect the rest of the components on my mobo? I've heard that OCing my cpu can cause my PCI cards to be overclocked as well, is that true, or is it just when you increase the voltage?
 
Slower memory shouldn't make the overclock less effective any more than it is making your stock speeds less effective. Right?

Yes, 4:6 is the right ratio (or 2:3 if you simplify it). You seem to understand a bit since you got the ram ratio right. So that is for stock speeds. Now when you go to OC and lets say you want your memory to remain close to 133MHz (DDR266) then you would lower the ratio to 3:5. That would make the RAM run at ~122MHz with a stock CPU speed. But as you OC the CPU, the ram will also OC. And if you hit 217MHz for the base CPU frequency then your ram with a 3:5 ratio is back at 133MHz (DDR266). And if you can go higher still with your CPU then lower the RAM ratio again to say 1:2 (100MHz RAM speed). Would take OCing the CPU speed up to 266MHz for your RAM to be back at 133Mhz with a 1:2 ratio. Make any sense or is that too complicated?

When you up the base CPU frequency then memory goes up. PCI speed goes up. PCI-Express goes up. SATA bus goes up. IDE for uses the PCI bus (SATA sometimes does too). You don't want those changing much at all or you will get corrupt data. If you can, LOCK em. Or you are not getting far with your OC at all.
PCI = 33MHz
PCI-Express = 100MHz
SATA = 150MHz (I'm not sure if SATA300 uses 300MHz or simply 150MHz and the SATA300 jumper selected on the motherboard)

Also might want to drop the HyperTransport (HTT) multiplier down one too. A lower HTT speed doesn't impact performance really. And you don't the HTT link's speed maxed out on your board in case it starts to limit your OCing.

Another little guide
http://wiki.extremeoverclocking.com/wiki/Athlon_64_Overclocking_Guide
 
Also, you need to monitor the voltage drop of your motherboard. It could be more than just your Power Supply.

There should be tables out there somewhere that shows the voltage setting and the corresponding actual voltage under load for your board.



Though this is probably significantly less of a voltage drop than your power supply would incur, it can have a substantial effect on your overclocking.
 
How do I lock the PCI frequencies and the others? I'm assuming through the bios, correct?
 
How do I lock the PCI frequencies and the others? I'm assuming through the bios, correct?
Yeah, in the bios. Usually those kind of settings are set to AUTO but the other options they can be set to are in MHz. If your board can lock those speeds then setting it to a specific MHz would lock it.
 
Ok so when I get home I'm gonna mess around with this a little bit. I'll keep you guys updated on how it goes
 
Ok some of the settings in my mobo are completely different from anything I've read, but I managed to get my PC to the following so far:

CPU Multiplier: 10x
CPU Freq: 240mhz

2.4 ghz OC on the CPU.
My RAM is now running at 240mhz.

I'm gonna run Prime95 and test this, then gonna change the CPU freq to 266 and see if it holds. Seeing as I have my RAM at PC2100 speeds, and the bios won't let me change the RAM clock speeds lower than 200 mhz, does that mean I'll be limited by my RAM if I try to OC past 2.66 ghz?
 
Back
Top