Intel Haswell 4670k + 4770K Overclocking guide

[ProKoN]

 

One thing tho, If i'm not able to get 1:1 ratio at 4.4Ghz with 1.32V, should I be happy with 4000-4200Mhz on the ring? Or should I try increased the uncore voltage and see

if i can manage the temps to maybe get a stable 1:1 ratio? What do you think ProKon?

 

Yes 4000-4200 ring bus is really good. dont obsess over a 1:1 ring ratio. very few chips can do this past 4.1GHz.

 

200mhz of ring bus performance is not even noticeable .

 

 

you can go for broke with the uncore if you want. crank the uncore voltage up to 1.45V and see if you can get your 1:1 ratio. Uncore voltage will not significantly increase over all temps.

 

I think your gonna get great performance at 4.4ghz with the uncore at 4.0-4.2ghz

 

If you want more uncore frequency...keep tweaking

[ProKoN]

Yes temps are hot.

 

So how you make better cpu throttling?

 

 

my challenge is to get stable 4.5 and up, while maintain the max temps below 90.

 

 

1- Get a better cooler. Go to a AIO 240mm+ water cooler or custom loop . this costs money but it is well worth it. Haswell is a beast to cool.

 

2- Delid. I dont recommend this because of the inherent risks associated. but it is a cheap solution to bring temps down. I do it myself

 

NH-D14 is a great cooler ( my favorite air cooler) but it can Not cope to well with Haswell heat output past 1.35V

[ProKoN]

I'll need to use this soon, once I get my NH-D15.

 

Also, @ProKoN , I live in Sherwood Park

 

 

Howdy nieghbor

 

I have a barebones NH-D14 for sale. 25.00, no fans. includes intel mount only.

[Jerakl]

Howdy nieghbor

 

I have a barebones NH-D14 for sale. 25.00, no fans. includes intel mount only.

mmmm, might have to look at that. NH-d15 is 100$. Don't have that kind of money ATM. And I have a 4770K

[ProKoN]

mmmm, might have to look at that. NH-d15 is 100$. Don't have that kind of money ATM. And I have a 4770K

my best suggestion is save your pennies until you can afford a good AIO water cooler. well worth the investment. 4770K is a beast to tame.

 

100 for an air cooler is too much

 

this will perform better than a d15. I have seen mem express retail the waater 2.0 for 69.99 so keep your eyes open

 

http://www.memoryexpress.com/Products/MX39192

[Jerakl]

my best suggestion is save your pennies until you can afford a good AIO water cooler. well worth the investment. 4770K is a beast to tame.

 

100 for an air cooler is too much

 

this will perform better than a d15. I have seen mem express retail the waater 2.0 for 69.99 so keep your eyes open

 

http://www.memoryexpress.com/Products/MX39192

I'm not looking at water cooling ATM.

I'm thinking of picking up a decent Air cooler until I can afford a loop.

[ProKoN]

I'm not looking at water cooling ATM.

I'm thinking of picking up a decent Air cooler until I can afford a loop.

 

you may have to get creative to fit a 240mm rad into your case... i managed to stuff one into an antec 900 so anything is possible

 

Water cooling is where its at for overclocking Haswell.

 

I recommend picking up a swiftech product so you can upgrade the loop

[Jerakl]

you may have to get creative to fit a 240mm rad into your case... i managed to stuff one into an antec 900 so anything is possible

 

Water cooling is where its at for overclocking Haswell.

 

I recommend picking up a swiftech product so you can upgrade the loop

So far all I know is EKWB and Swiftech pump

 

I have a Storm Trooper, as you probably know. I think it will be easy to find rad mount points. I think there might be a 240 on the top actually.

 

Also, in your sig it says HDD Kingston HyperX-3k 240Gb

Isn't that drive and SSD ?

[ProKoN]

So far all I know is EKWB and Swiftech pump

 

I have a Storm Trooper, as you probably know. I think it will be easy to find rad mount points. I think there might be a 240 on the top actually.

 

Also, in your sig it says HDD Kingston HyperX-3k 240Gb

Isn't that drive and SSD ?

 lol yes its an ssd

 

im old school

[Jerakl]

 lol yes its an ssd

 

im old school

 

SSD = HDD

 

[Mjolkspam]

Page 25, so a new post inc.

 

Won't be much now tho.   I tried ring bus at 4100Mhz, crashed after about 5-6 hours. So i'll probably stick with the lastest stable which is,

 

CPU 4.4Ghz, vcore 1.32V

"The Ring", 4000Mhz, uncore 1.32V

vccin @ 2.2V

Memory, 1333Mhz  1.5V

 

My next steps is trying to lower the voltages one or two steps just because I have time and it's fun to do this.

So i'm gonna start with vcore, lower it to 1.31V.

 

After all that, memory is going to run XMP. Shall be fun to see how that goes.

**EDIT**  One

I tried to lower the voltages, no luck (didn't think I would get any lower tho).

But  I ran into a problem today, with my last stable OC..  The one in this post. I did manage that 12+ hours stress test, but today when I played

some BF4 it crashed after 5-10min, got X124 BSOD. CPU was at 50% load or something, 60% at peak. It works this low as I had the FPS locked to 65. So I increased uncore voltage to 1.33V, played for atleast 1 hour

without any crash, this time I had the FPS free, so CPU worked at 80% load.

 

Either BF4 does not like overclocked cpus OR I have an unstable overclock..      I know BF4 vs. a stress test is not the same. Load on the cpu is different.  80% load from BF4 could be

more "punishing" than 100% load from the stress test. 

 

Any ideas?  Should i run a even longer stress test with 1.32V on both vcore and uncore again, or should I use the 1.33V on uncore and do a long stress test?

I'm getting out of ideas.  I'm just pretty lost in my mind as i did an 12+ hour stable stress and then it crash after 5-10min in BF4.

**EDIT** Two

 

Okey.. I'm starting to get pissed on my cpu now

 

I increased the uncore voltage to 1.33V, I didn't have the chance to do a stress test tho.  I'm watching NHL (Pittsburgh vs. NY Rangers), got X124 BSOD!

So for now, I left ring bus ratio at Auto (3800Mhz). uncore at 1.32V tho.   

 

As it looks now, i'm having a hard time getting it stable with:

CPU 4.4Ghz, vcore 1.32V

"The Ring", 4000Mhz, uncore 1.32V (tested 1.33V)

vccin @ 2.2V

Memory, 1333Mhz  1.5V

 

In my eyes, this is just bad..  I'm starting to think I have a bad chip more and more =/        

You got any ideas for me ProKon?

 

Mini Edit!      I know i asked this on page 23, but i need to ask again.  Now when i'm finding my OC not stable anymore for some reason, should i change CPU Ratio Mode  from 'Dynamic' to 'Fixed'? (Noticed you used 'fixed' in your video).

Same goes for voltages, while finding an stable OC, should i use Adaptive or Override?

[morbid_bean]

I would say your fairly stable overall. Chances of you crashing while gaming, editing or doing normal tasks is fairly slim.

 

I would just run it as is. if you crash at all during a normal application give your chip a slight bump in vcore first,  .05 - .1V

 

if it crashes a second time do the same for the ring voltage.

 

vccin should be fine.

 

unless you run all tests at 24 hours and isolate the variables individually its difficult to say what exactly needs to be tweaked

 

I like to give er 18-24 hours my self also If it goes 18 hours I know its fairly solid. Xtu is not the most hardcore stress test but its fairly representative of the type of stress modern applications apply.

 

Nice LoVo overclock. most stock turbo voltage is 1.150V (on the i7, i5 is a bit lower).  you got a lower than stock voltage and all cores at 40x. well done. are you on stock cooler?

 

Excellent thanks for the Input!!    I will give this a try if I happen to crash.   That is good to hear that my Voltage is quite low for a 4.0Ghz.  I was getting into some pretty high / uncomfortable voltages when I was trying for 4.4Ghz, so thats awesome.    I am currently cooling with a Hyper EVO 212 with one fan.   Do you think it would help at all putting another fan on for a Push Pull Configuration?

[ProKoN]

Excellent thanks for the Input!!    I will give this a try if I happen to crash.   That is good to hear that my Voltage is quite low for a 4.0Ghz.  I was getting into some pretty high / uncomfortable voltages when I was trying for 4.4Ghz, so thats awesome.    I am currently cooling with a Hyper EVO 212 with one fan.   Do you think it would help at all putting another fan on for a Push Pull Configuration?

 

 

Push pull usually helps a bit, it also looks beastly

 

I would say you would drop 1-2C by adding another fan

[Harlin]

Okay so to update.

 

Managed to get to 4.4Mhz, but not stable in IETU, but stable @ Desktop. I'm not sure where to go now. Temps looks great after installing a Be Quiet Dark Rock Pro 3 (omg that was a mission..). Before crashing on the below stats, temps are around 70c load, so I seem to have more headroom, but do not want to just keep chucking more into the vcore. The crashes have been both the same, windows freezes completely, forcing me to perform a hard reset.

 

EDIT: Dropped Uncores down to 35, wasn't taking the seeing in BIOS for some reason.

 

Current settings:

CPU VRIN Override LLC: Extreme 

CPU VRIN Override Voltage: 2.0

VCore: 1.30v

Ring Voltage: 1.15

CPU Multiplier: 44

BCLK: Auto

Turbo: Auto

C1E, C3, C6/C7, and EIST: Enabled

 

 

[ProKoN]

Okay so to update.

 

Managed to get to 4.4Mhz, but not stable in IETU, but stable @ Desktop. I'm not sure where to go now. Temps looks great after installing a Be Quiet Dark Rock Pro 3 (omg that was a mission..). Before crashing on the below stats, temps are around 70c load, so I seem to have more headroom, but do not want to just keep chucking more into the vcore. The crashes have been both the same, windows freezes completely, forcing me to perform a hard reset.

 

EDIT: Dropped Uncores down to 35, wasn't taking the seeing in BIOS for some reason.

 

Current settings:

CPU VRIN Override LLC: Extreme 

CPU VRIN Override Voltage: 2.0

VCore: 1.30v

Ring Voltage: 1.15

CPU Multiplier: 44

BCLK: Auto

Turbo: Auto

C1E, C3, C6/C7, and EIST: Enabled

 

 

Thanks for the update

 

not sure why your uncore is hitting 4.0GHz if you only set it for 3.5ghz

 

if you have any overclocking gigabyte software installed i would suggest uninstalling it.

 

im not familiar with the gigabyte tweaker and am unsure if that might be the source of the issue. i know the gigabyte app center oc program  causes issues

 

one suggestion is set the uncore for 35x in the xtu program, restart the computer and see if that makes a difference.

 

if you not having any luck try uninstalling the tweaker program and see if that was the root.

[Harlin]

Thanks for the reply.

 

I've uninstalled EasyTune and have set uncores to 35. If heat is not an issue (yet), where can I look to increase stability. I'm finding the numerous options in the UEFI Gigabyte bios overwhelming, and I'm starting to fiddle around with multiple settings and end up back to 4.2mhz starting over /doh

[ProKoN]

Thanks for the reply.

 

I've uninstalled EasyTune and have set uncores to 35.

 

 

did that help? or does uncore still turbo to 40?

 

 

stability will most likely only be found in these parameters.  all the other settings can honestly be left at auto. there is no magic switch that increases stability.

 

Even  CPU VRIN Override LLC: Extreme........................................................ will not likely yield any better results in terms of stability, however it is the suggested setting.

 

core multiplier

 

uncore multiplier

 

memory frequency

 

memory timings

 

Processor Input voltage ( also known as VCCIN \ VRIN)

 

core voltage

 

uncore voltage

 

memory voltage

[morbid_bean]

I would say your fairly stable overall. Chances of you crashing while gaming, editing or doing normal tasks is fairly slim.

 

I would just run it as is. if you crash at all during a normal application give your chip a slight bump in vcore first,  .05 - .1V

 

if it crashes a second time do the same for the ring voltage.

 

vccin should be fine.

 

unless you run all tests at 24 hours and isolate the variables individually its difficult to say what exactly needs to be tweaked

 

I like to give er 18-24 hours my self also If it goes 18 hours I know its fairly solid. Xtu is not the most hardcore stress test but its fairly representative of the type of stress modern applications apply.

 

Nice LoVo overclock. most stock turbo voltage is 1.150V (on the i7, i5 is a bit lower).  you got a lower than stock voltage and all cores at 40x. well done. are you on stock cooler?

 

 

I thought I would follow up with this post.   Played about 6 Hours worth of games last night, and today I managed to get in 2 or so hours of Guild Wars before I got a Blue Screen.   So looks like I will start bumping up the Vcore a tiny bit and see how it goes.  Kind of a bummer, but oh well its part of the OC Game!   It is a rather HOT day here in Denver and a bit warm in my room.  Sadly my temps are 40c while typing this, while gaming it was reaching up near 55-57 while gaming,  Kind of sucks slightly high for my tastes but, ehhh I can live with it.

 

Is there a way to determine from the BSOd information what would actually need tuning on the overclock?  After restart, this is what I got:

  Problem Event Name:	BlueScreen  OS Version:	6.1.7601.2.1.0.256.48  Locale ID:	1033Additional information about the problem:  BCCode:	124  BCP1:	0000000000000000  BCP2:	FFFFFA8007268028  BCP3:	00000000BF800000  BCP4:	0000000000000124  OS Version:	6_1_7601  Service Pack:	1_0  Product:	256_1

[ProKoN]

 

I thought I would follow up with this post.   Played about 6 Hours worth of games last night, and today I managed to get in 2 or so hours of Guild Wars before I got a Blue Screen.   So looks like I will start bumping up the Vcore a tiny bit and see how it goes.  Kind of a bummer, but oh well its part of the OC Game!   It is a rather HOT day here in Denver and a bit warm in my room.  Sadly my temps are 40c while typing this, while gaming it was reaching up near 55-57 while gaming,  Kind of sucks slightly high for my tastes but, ehhh I can live with it.

 

Is there a way to determine from the BSOd information what would actually need tuning on the overclock?  After restart, this is what I got:

  Problem Event Name:	BlueScreen  OS Version:	6.1.7601.2.1.0.256.48  Locale ID:	1033Additional information about the problem:  BCCode:	124  BCP1:	0000000000000000  BCP2:	FFFFFA8007268028  BCP3:	00000000BF800000  BCP4:	0000000000000124  OS Version:	6_1_7601  Service Pack:	1  Product:	256_1

 

 

looks like a x124, typically related to uncore instability

 

 

00000000000000124

 

 

increase uncore voltage or decrease uncore multiplier.

 

a bump up in VCCIN \ Vrin may help

 

I need 1.45V to get my uncore stable at 4.4GHz on my i7.

 

           1.40V to get 4.4GHz on my i5.

[Mjolkspam]

@ProKoN, please check my post above this one as well. Need answers for both.

 

But this new matter is more important tho..   Yesterday and today when i shut down my PC to sleep, I booted it up again because i forgot to check some stuff.

The problem is tho, i didn't boot into windows. My bios had done a reset for some reason, so I had the option to press either  F1 for Load default and go into bios  OR  F2 to load default but boot into windows.

Here is another deal with the problem, can't do anything, w/e I do.. Change usb-port restart the PC nothing..    I have to clear CMOS by myself back on the motherboard panel, start it up to get a similar

screen as before I cleared the CMOS, but with more info about the PC etc..  BUT I still can't do anything, so another reboot is needed, then I don't even get the screen at all.. It just boots like normal with default settings

 

All this started yesterday, I have no idea why! I have done nothing inside the BIOS.  Just weird.

And sorry if my explanation sucks for this.. I'm no good at that  So please ask if you need more info.  I realy need help with this =/

**EDIT**  We can scratch pretty much the things inside the spoiler..

 

I know why I couldn't press F1 to begin with, I had my CM Storm Ultimate keyboard on "N" mode, so I can press more than 6 keys at the same time, that will do so it doesn't work before the UEFI had loaded

So that is fixed.  But the other problem was still there.

 

I did a test to see if it happened again, so i turned the PC off. Waited about the same time as before, started it up and the same thing happened.  It says the previous OC failed or something like that (A fail safe I guess).

The weird thing is tho, I ran that OC the whole day yesterday and today, but when started up again after a shutdown, all of a sudden it fails?  

And if I load that profile with those OC settings I have no problem what so ever booting into windows?  All this seems weird to me!

 

My thinking off this is that my PSU is failing me?  As I have performance issues in games even on stock (mentioned this before).  What if my PSU can't deliver anymore?

What if that's the reason i'm having such a hard time getting stable OC's even tho it was stable for 12+ hours before.   Like it can't deliver whats needed when it's really needed or something.

 

Please say that i'm wrong.. That it has to be somethings else.   Any ideas?

 

Well off to sleep for now, soon 6 in the morning here in Swe

[ProKoN]

@ProKoN, please check my post above this one aswell. Need aswers for both.

 

But this new matter is more important tho..   Yesterday and today when i shut down my PC to sleep, I booted it up again because i forgot to check some stuff.

The problem is tho, i didn't boot into windows. My bios had done a reset for some reason, so I had the option to press either  F1 for Load default and go into bios  OR  F2 to load default but boot into windows.

Here is another deal with the problem, can't do anything, w/e I do.. Change usb-port restart the PC nothing..    I have to clear CMOS by myself back on the motherboard panel, start it up to get a simular

screen as before I cleared the CMOS, but with more info about the PC etc..  BUT I still can't do anything, so another reboot is needed, then I don't even get the screen at all.. It just boots like normal with default settings

 

All this started yesterday, I have no idea why! I have done nothing inside the BIOS.  Just weird.

And sorry if my explaination sucks for this.. I'm no good at that  So please ask if you need more info.  I realy need help with this =/

 

are you on the latest bios?

 

if you are maybe try flashing it again

 

or even try to roll the bios back to a previous version

 

perhaps the bios got corrupted by your fiddling.  ( it has happened to me)

 

your obviously not game stable if bf4 crashes after 5-10minutes.

 

x124 is ring bus.

[Mjolkspam]

are you on the latest bios?

 

if you are maybe try flashing it again

 

or even try to roll the bios back to a previous version

 

perhaps the bios got corrupted by your fiddling.  ( it has happened to me)

 

your obviously not game stable if bf4 crashes after 5-10minutes.

 

x124 is ring bus.

I'm on the latest bios.

 

I have never reflashed or rolled back a bios before, not on UEFI atleast.  Reflashing the same, just download and install again i guess?

And same goes for if i roll back?   Just install it.

[ProKoN]

I'm on the latest bios.

 

I have never reflashed or rolled back a bios before, not on UEFI atleast.  Reflashing the same, just download and install again i guess?

And same goes for if i roll back?   Just install it.

yes and yes

[Harlin]

did that help? or does uncore still turbo to 40?

 

 

stability will most likely only be found in these parameters.  all the other settings can honestly be left at auto. there is no magic switch that increases stability.

 

Even  CPU VRIN Override LLC: Extreme........................................................ will not likely yield any better results in terms of stability, however it is the suggested setting.

 

core multiplier

 

uncore multiplier

 

memory frequency

 

memory timings

 

Processor Input voltage ( also known as VCCIN \ VRIN)

 

core voltage

 

uncore voltage

 

memory voltage

 

Right.

 

So I'm stable at the below. 2 hours IETU and several hours BF4. Not quite 8 hours of Stress testing, but I'll confirm tomorrow. Temps were fine, max temp was 78. I tried a couple of runs on the IBT but stopped as that was around 90c maximum. I feel this CPU has x44 in it somewhere. Will try with the same settings later on x44 and see what happens...

 

Core x 43

Uncores 36

vCore 1.32

VCCIN 2.0

Vring 1.05

LLC Extreme

Dram 1.5v @ 1333mhz

 

EDIT: So after trying x44, I had a BSOD within 5m of starting IETU, so after the reboot, I upped the VRing from 1.05 to 1.09, and had an immediate BSOD. I do seem to have a CPU which can take voltage, Cinebench/IETu all have me down for 77c max. I'm at a loss as to what is causing the instability. They are 124 errors, but I read that's generally hardware.

 

My memory is Samsung lowprofile 12800 1.35v rated, but I'm running it at 1.5v @ 1333 9-9-9-24.

 

Cheers

Edited May 5, 2014 by Harlin

[ProKoN]

Right.

 

So I'm stable at the below. 2 hours IETU and several hours BF4. Not quite 8 hours of Stress testing, but I'll confirm tomorrow. Temps were fine, max temp was 78. I tried a couple of runs on the IBT but stopped as that was around 90c maximum. I feel this CPU has x44 in it somewhere. Will try with the same settings later on x44 and see what happens...

 

Core x 43

Uncores 36

vCore 1.32

VCCIN 2.0

Vring 1.05

LLC Extreme

Dram 1.5v @ 1333mhz

 

EDIT: So after trying x44, I had a BSOD within 5m of starting IETU, so after the reboot, I upped the VRing from 1.05 to 1.09, and had an immediate BSOD. I do seem to have a CPU which can take voltage, Cinebench/IETu all have me down for 77c max. I'm at a loss as to what is causing the instability. They are 124 errors, but I read that's generally hardware.

 

My memory is Samsung lowprofile 12800 1.35v rated, but I'm running it at 1.5v @ 1333 9-9-9-24.

 

Cheers

 

x124 is typically caused by unstable ring frequency

 

your ring voltage is fairly low even for x36.

 

i would bump up ring voltage to 1.30-1.40V.

 

and try again

 

keep your ram where its at for now.