cancel
Showing results for 
Search instead for 
Did you mean: 

Serious about OC...but a noobie.

pigeonfood11
Level 7
Hi, I hope you guys can help me get over this milestone of being able to overclock my cpu. I've been wanting to do it for a long time just never had the time to do but now i do. I've read quite a bit about it but always get mixed information whats safe and stable, But I know every cpu can be different, so then i get i skeptical on my ability to do it by myself.
I hope with you guys knowledge i can achieve a safe stable and fast computer.

My specs:
CPU: Intel i-5 4690K 3.4ghz
Mobo: Asus Maximus VI Hero
Ram: Crucial Ballistix Sport 8GB (2 x 4GB) & Avexir Core series 8GB (2 x 4GB)
CPU Cooler: Corsair H100i
Case: Corsair 750D
Case Fans: 10....

Programs im using to stress test...CPUZ, Prime94, Aida64 & CoreTemps
What i have done so far was to update my bios to most up-to-date version, I set all the bios settings to "Load Optimized Defaults".
I then started the computer and ran a "Stability System Test" for 2hrs....and the highest temp was between 55-65.

That is as far as i have gone with overclocking...I guess my next question now is what do I do now with the info I have gotten from using Aida64.
289 Views
21 REPLIES 21

gravy1958 wrote:
if your goal is to set a small increase then it achieves that result completely safely using software... people with my oc problem have posted solutions by increasing core voltage so the software option in this case is better than the manual oc... however you wouldn't get the experience of trial and error and the sense of achievement when it doesn't go pop 😉 jk
have fun be safe.


xD get a load of this guy.

the mis-information is real.

dude, if you're using software to overclock, and then states that all overclocks are inherently unstable, i can't take you seriously.

also, if you bought a pre-build PC, then you just shot yourself in the foot from the beginning.

essentially all DDR3 kits over 1866 is factory overclocked, not to mention almost all custom PCB GPU's, also CPU turbo boost.

Anyway, @Pidgeon Arne can probably tell you all you need yo know, otherwise feel free to tag me 🙂

Watercooling FTW :cool:

"essentially one of the most hardcore overclocking motherboards ever made."
http://www.hardocp.com/article/2014/03/24/asus_rampage_iv_black_edition_motherboard_review/7

Ok so late last night I tried a few things that you suggested Arne. To begin I started by doing probably what most overclockers do and started out high and worked my way down until it booted to windows. I did disable cpu spread spectrum but didnt find any vid support settings to disable. ( any idea what im looking for?)
I started at 4.6 @ 1.3v until it booted, it finally went to windows when i changed the settings to 4.3 @ 1.25v.
I ran RealBench with 16g ram and within about 20 sec it was hitting low 80's.
I stopped it but im kinda confused if even my fan config is even working. When it got to high 70's the fans would kick in faster but not on like "omg im about to fry my cpu mode" didnt even make a dent in bring temps down....i have 10 fans going w/watercooling! I guess this is my new struggle before I can overclock 😞
Below is a pic of my "fan config"....its not the best pic but i did it at work, may someone can shed some light on what works best.
Thanks for all the help fellas.

M11C wrote:
xD get a load of this guy.

the mis-information is real.

dude, if you're using software to overclock, and then states that all overclocks are inherently unstable, i can't take you seriously.

also, if you bought a pre-build PC, then you just shot yourself in the foot from the beginning.

essentially all DDR3 kits over 1866 is factory overclocked, not to mention almost all custom PCB GPU's, also CPU turbo boost.

Anyway, @Pidgeon Arne can probably tell you all you need yo know, otherwise feel free to tag me 🙂


get a load of this guy XD

Actually I would never oc ... if I want faster I buy faster and as I said... people talk in percentages but in reality the difference is virtually nothing. No overclock is going to make earth shattering difference... if you believe it does... then the mis-information is real 😉 Each to their own 🙂

gravy1958 wrote:
get a load of this guy XD

Actually I would never oc ... if I want faster I buy faster and as I said... people talk in percentages but in reality the difference is virtually nothing. No overclock is going to make earth shattering difference... if you believe it does... then the mis-information is real 😉 Each to their own 🙂


xD this guy is hilarious!

keep telling yourself that.

you obviously have no idea what you're talking about.

most ridiculous noob to ever sign up for this forum.

and your "spec's" are a joke xD

Watercooling FTW :cool:

"essentially one of the most hardcore overclocking motherboards ever made."
http://www.hardocp.com/article/2014/03/24/asus_rampage_iv_black_edition_motherboard_review/7

Arne Saknussemm wrote:
Basically you want to set something like a medium value for LLC, disable CPU spread spectrum and any VID support and then set a manual voltage of 1.3volts and then simply up the multiplier one at a time from stock value and see if you can boot into windows. When you fail to boot into windows drop the multiplier a couple of points...boot into windows and see if you can run a stress test like AIDA or better RealBench. Keep an eye on your temps while you do this...anything over 80 I would stop the test.

Run RealBench stress test, selecting your full RAM quantity...for 15 minutes. If you pass try upping multiplier one..if you fail try dropping multiplier one.

One click solutions are not going to give you optimal voltage or a great OC experience frankly.

Great guide to BIOS on that series of motherboards here https://rog.asus.com/forum/showthread.php?33488-Maximus-VI-Series-UEFI-Guide-for-Overclocking

More stuff here https://rog.asus.com/forum/showthread.php?35027-UEFI-BIOS-Tweak-Guides-for-Maximus-VI-Multiple-Incom...



1.3volts is quite a bit to start with isnt it? I mean im reading that link about my motherboard, and im only questioning becuase of what i read below so far.

"There is of course a side-effect. Having power circuitry on-die adds heat. Haswell processors run hot when voltage levels are increased. A very good air cooler is required for voltage levels above 1.15V.

1.20V-1.23V requires use of closed loop water coolers.

At 1.24V-1.275V dual or triple radiator water cooling solutions are advised."

Arne_Saknussemm
Level 40
Excellent....you're doing your reading...:D

I think for booting into windows test 1.3v on a H100i is OK but by all means 1.25 is safer. As I said for stress testing you want a hawk eye on temperatures and your ambient temp at home will have a big effect here. above 80 stop immediately and rethink voltage...

I think those voltages are kind of conservative and rightly so given peoples propensity for cooking their CPUs with hours of prime 95...I would advise against that as said...

In the end do what you feel comfortable with despite what I or anyone else may tell you 😉

Arne Saknussemm wrote:
Excellent....you're doing your reading...:D

I think for booting into windows test 1.3v on a H100i is OK but by all means 1.25 is safer. As I said for stress testing you want a hawk eye on temperatures and your ambient temp at home will have a big effect here. above 80 stop immediately and rethink voltage...

I think those voltages are kind of conservative and rightly so given peoples propensity for cooking their CPUs with hours of prime 95...I would advise against that as said...

In the end do what you feel comfortable with despite what I or anyone else may tell you 😉


Is running a stress test for 15minutes after every multiplier point i increase kinda too anal or should work my way down but using the boot screen as my limitation?

Also before i get to ahead of myself, when does DRAM come into overclocking?

Arne_Saknussemm
Level 40
I would not bother with that...the idea is to get an idea of the CPU booting windows. When you fail to boot windows at say 4.6GHz then back down to 4.4 and then do the stress test there....numbers just examples...

For me get RAM up and stable first....though neither exists in isolation...if you get XMP working at stock CPU and then OC CPU you might find you have to tweak RAM a bit since stress overall is increased....vice versa, if you stabilize CPU OC...adding XMP ram settings might mean you have to tweak OC.

I'm a memory first guy....

Arne Saknussemm wrote:
I would not bother with that...the idea is to get an idea of the CPU booting windows. When you fail to boot windows at say 4.6GHz then back down to 4.4 and then do the stress test there....numbers just examples...

For me get RAM up and stable first....though neither exists in isolation...if you get XMP working at stock CPU and then OC CPU you might find you have to tweak RAM a bit since stress overall is increased....vice versa, if you stabilize CPU OC...adding XMP ram settings might mean you have to tweak OC.

I'm a memory first guy....


RAM tweaking sounds complicated, since there is timing involved etc. Im totally clueless on RAM so where do i start, to be honest I always like to start in the unknown before i do something im more familiar with, meaning ive read quite a bit about cpu oc, but not ram.

Arne_Saknussemm
Level 40
Ignore the vid stuff..maybe not in that BIOS...if it's not obvious then it's not there.

Temps..OK most people would saty low 80s stress testing is not a big deal...your pic looks pretty standard and looks like you should have relatively good airflow...you have to remember that stress testing is just that...maximum heat and volts for a bit just to see if the system falls down in the worst case secenario....anything you do after and temps will be 20 degrees less

One thing you say 80s in 20 seconds...literally? First part of that test should be image manipulation and not give you high temps....the second and later parts of the test are pure CPU bashing and then you should see temps rise....