• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Phenom II Overclocking.

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

DRaZZ

Member
Joined
Jul 2, 2001
Location
wolverhampton, england
Hi all, I have a Phenom II x2 555BE on a Asus Crosshair IV Formula. The cpu is unlocked to a quad core. These pics are as far as i have been able to push this chip so far. I have seen these chips hit 4.5Ghz, basically im just after any ideas you guys have with regards to pushing this chip further. I have it water cooled and the load temps never exceed 50c. The whole system is in my sig
Cheers guys
DRaZZ
 

Attachments

  • overclock.JPG
    overclock.JPG
    65.2 KB · Views: 73
  • overclock2.JPG
    overclock2.JPG
    50.9 KB · Views: 69
The stickies in this section have something called the "Dolk Guide".
Read it about 5 times and do what it says to. You'll get all you can from there :thup:
 
Already read it matey. Was hoping some people with 4ghz + quad cores could let me in on thier secrets :)

I think on the PII lineup, anything over 4GHz was a crapshoot. Just takes a good chip that doesn't pull a lot of voltage.
 
You did not read them thoroughly enough, or you would have known to keep ht as close as possible to 2000Mhz ;)

Joke apart, you could feed it a little more vCore (up to 1.55v on load) and try to reach 4.1/4.2Ghz.
Very very very few of them will go beyond that under ambient cooling.
 
Drazz,if you're getting into the 50's you still have some headroom temperature wise, though it may only be enough for another 100mhz or so. As said above, drop your ht to 2000 and hope your chip is capable of doing more. It's just a silicon lottery you don't know what you have till you really start pushing. I have a really good 955be that I was able to get "stable" 2 hours prime blend @ 4.3 with 1.536 on the Cpu V, this same chip has been as high as 4650 for benching. I ran super pi32m at those speeds, on 2 cores mind you, 1.6 Cpu v. I believe 4.5 was as high as I've pushed it on all 4.
 
i'll pull the HT back n try it. Iv had the chip at 250x16.5 - 4125, stable till about 30 mins of torture, then it crashes. I can push the chip higher, my max load temp is 47C after hours of prime or gaming, i test with gaming as its real world test and has the added heat from the gcard. I know u have to stay below 55 so iv got a bit to play with. Its the Vcore i havent upped too far, 1.48 is highest i took it, not due to temps but because i heard 1.5 is absolute max (even tho i saw earlier someone at 1.6v)
Im guna go drop the HT (even tho i dont understand why that bein lower is better lol) crank it to 1.48 and try boot in at 4.2, we shall see what happens.
 
Well i booted to windows with 4100Mhz, 1.48V. Lasted for about 20 seconds at the desktop then crash. Its not a temp thing, had temp open in those 20 seconds n temps were same as normal, nothing higher than 41C. I think maybe iv just hit this chips maximum, 4.1 it just doesnt wana do!
 
I guess if i dropped back to 2 cores i may get another 100 or 2 out of it, not worth losing quad core for a few hundred mgz tho! 800Mhz over rated speed isnt too bad though i guess, just means its time to get an 8 core and bump it up over 5ghz!
All the years i have been overclocking theres one fact i have learned time and time again... If you hit a speed barrier you cant seem to pass, throw money at it untill you do !!!
 
Do you think its worth trying to up the Vcore higher than 1.48? Its perfectly stable at 4000 with 1.46v but crashed after 20 secs desktop (no activity) at 4100 with 1.48v......
If you think more vcore will kick its ***, ill give it a whirl! What v do u suggest matey?
 
Do you think its worth trying to up the Vcore higher than 1.48? Its perfectly stable at 4000 with 1.46v but crashed after 20 secs desktop (no activity) at 4100 with 1.48v......
If you think more vcore will kick its ***, ill give it a whirl! What v do u suggest matey?
There is a point where the amount of Cpu V you need to give a chip to get to X Mhz just isn't worth it. You need to figure that out on your own, we cannot tell you. Getting it stable at 4.1 may take more then you're willing to give.
 
Yea, the guy suggested more vcore, i asked him what he thought it would take to stabalize 4.1 also asking if more WOULD stabalize it. As lots of people have pushed past 4ghz, they know things i dont, for instance they may have experience there n may know that it takes 1.5v to get the chips over 4ghz...... This is why i ask questions, so that people who have experience in areas i dont can advise!
 
Yea, the guy suggested more vcore, i asked him what he thought it would take to stabalize 4.1 also asking if more WOULD stabalize it. As lots of people have pushed past 4ghz, they know things i dont, for instance they may have experience there n may know that it takes 1.5v to get the chips over 4ghz...... This is why i ask questions, so that people who have experience in areas i dont can advise!

mandrake4565 said:
There is a point where the amount of Cpu V you need to give a chip to get to X Mhz just isn't worth it. You need to figure that out on your own, we cannot tell you. Getting it stable at 4.1 may take more then you're willing to give.

I agree with what Mandrake says about it not being worth the extra Vcore and heat at some point in the overclocking process. Especially as fits into a 24/7 usage.

Do we know more than you? Nope and certainly n0t about your piece of silicon which is a dice-roll for overclocking potential and outcome. 1.46Vcore for a stable 4.0Ghz is not a lot since many do run 1.525Vcore regular if the temps remain fully within limits.

Now would 1.525Vcore for an overclock from 4.0Ghz to 4.1Ghz be worth it to me? No it would not but it might be worth it in some other person's mind.
RGone...
 
Drazz to expand on what RGone and I said, we both know what our chips are capable of because we have pushed them to their limits. Unfortunately what works for my chip doesn't necessarily work for someone else with the same chip. For example We both have Fx 8350's as well as another forum member. Rgone 8350 needs slightly less Cpu V to get to X Mhz as mine does, from what I've seen usually about .00625-.0125. Where as our other friend that has one usually will need more then both of us to get to X Mhz. The determining factors are also affected by, the motherboard and BIOS used, as well as the cooling etc.... I know my chip will need less voltage when I'm really stepping on it if I can get my water down below 18c.

What I'm trying to get at is we can help, but when it gets down to brass tacks you are the one who needs to determine how much is too much voltage and what your cpu needs to get to X Mhz. For you to do 4.1 ghz you may need 1.5 or it may need 1.6 the only way to see is by doing it, that's if you're comfortable doing so.

From what I see on your screen shots you may want to do the following:
As said above lower the Ht to 2000
You may also want to try setting the Ram to 1333 just to see if you can find stability at 4.1 then try raising it back to 1600. The Imc on the Denubs aren't the strongest and taking that out of the equation may help. If you do find stability at 4.1 you can then put it back and if it becomes unstable you know what is causing it.
 
CPU multi limit....I would drop the ram devider down to 1333 and put HT at 1800,then put the CPU NB at 2400, add 5 to the reference clock at x20 CPU multi then give it a go,you may get the extra 1-200mhz that way but you will be sacrificing other things in the process (ram speed/CPU NB) you may get them to run at the higher speed you're running after adding to the reference clock, but its wise to drop them down before doing so.
 
mandrake4565 said:
Where as our other friend that has one usually will need more then both of us to get to X Mhz. The determining factors are also affected by, the motherboard and BIOS used, as well as the cooling etc...

As said another forum member had an FX-8350 that took more voltage to run X speed and it was suspicioned that the real owners mobo and ram or cooling or something was the problem with it taking so much Vcore to run a big number over 4.9Ghz. So he sent his FX-8350 to me and I put it in my board with my same bios I always use and my good cooling. STILL took BIG Vcore.

So it was the cpu at odds with High Cpu Speed and not the ram, mobo or cooling in this case. But to prove it out we actually had to mail a cpu across the country and test it in a known situation.

You will have to make your situation known. I would imagine you get the idea by now that it is at the user's end where the rubber meets the road. I only typed this post to show what was the issue for one user and it was dead on the cpu and we answered his long-standing question. He had no FEAR of Vcore but still wanted to know for sure it was not something other than the cpu. We proved it out. Wonky Vcore needing cpu.
RGone...ster.
 
Back