Overclock.net › Forums › Intel › Intel CPUs › Ocing 3770k causing the BIOS/UEFI to freeze
New Posts  All Forums:Forum Nav:

Ocing 3770k causing the BIOS/UEFI to freeze

post #1 of 5
Thread Starter 
Hello everyone, long time reader, first time posting.

I recently came across a rather annoying issue in which I was hoping you could help me with.

After assembling my new PC and doing some proper cable management, I've decided to overclock my 3770k after reading the following guides:

http://www.overclock.net/t/1247413/ivy-bridge-overclocking-guide-with-ln2-guide-at-the-end
http://www.overclock.net/t/1291703/ivy-bridge-overclocking-guide-asus-motherboards
http://www.overclock.net/t/1392279/info-intel-3570k-3770k-overclocking-tips

And keeping the following images at hand:

http://img708.imageshack.us/img708/9168/volts.png
http://img269.imageshack.us/img269/4421/pvts.png
http://cdn.overclock.net/d/d8/1000x2000px-LL-d8676059_OVERCLOCK.png


As the way I'm overclocking may be flawed and caused this issue, let me start by giving you my understanding on it in no more than one paragraph.

It is from my understanding that I should set the LLC to Turbo to keep the VDROOP to roughly 75% instead of Extreme as that option would tighten the VDROOP even more which would be more of a priority for LN2 cooling than Air/H20 cooling. I've also increase the turbo limits to 200A and 300W, set my voltage to Manual 1.10v to start with and then increased my multiplier one by one, while running prime95 for 10 minutes each time to see if it was stable to go to the next phase. When the BSODs start, I would ever so slightly increase the vcore temper with the CPU PLL as well to reduce temperatures.

I'm also leaving the frequency at the 35x stock multiplier and only increasing it on the Turbo option so it goes up to that overclocked frequency when needed instead of being permanently on a high frequency.

The problem occurred at the very beginning of this process, when I hit 42x multiplier. I got the expected BSOD but when I went back to the UEFI/BIOS, it froze after 1-5 seconds, I would never be fast enough to increase the vcore or decrease the multiplier, forcing me to completely shut down my PC and hit the CMOS reset button on my MoBo to be able to restart. This means that for every trial and error attempt I will need to set up everything in the BIOS again.

Is this normal? Is there anything I'm doing wrong?

As you may need this information to help me, I'll leave you here my PC specs:

CPU: Intel Core i7 3770k 3.5ghz - 4.1ghz
MoBo: Gigabyte Z77 UD5H
Memory: 2x 8GB 1600MHZ CL9 Corsair Vengeance
GFX: Gigabyte Geforce 770 non-reference Windforce 3x OC Edition
PSU: Corsair AX760 80PLUS Plat.
W/C Solution: Corsair H110
Case: Corsair Vengeance C70

I'm using CPU-Z, RealTemp and prime95 to aid me with my overclocking.

Apologies if I didn't go straight to the point but I wanted to make sure I would cover everything so you could see if there was any flaw in my process which would cause the issue.

Thank you in advance for the time you took reading this.
Edited by Furialz - 7/27/13 at 11:56am
post #2 of 5
When overclocking sometimes you will have to.cmos or reflash bios. What your experiencing is normal, it was just really unstable
post #3 of 5
Thread Starter 
I see, that's reassuring!

I also forgot to mention something.

When I ran prime95 at 41 ghz I received a "the application had to close due to an unexpected error" type of message. Is this a prime95 fail test or if prime95 fails, you get a bluescreen?

Furthermore, did you see any flaw in my overclocking method as well or is it proper?

As a last question, regarding prime95 would you recommend running it with -1.5 GB of your memory, 90%?, 85%? I've read so many different things. I have 16 GB, so perhaps 14500?
Edited by Furialz - 7/27/13 at 12:12pm
post #4 of 5
Quote:
I got the expected BSOD but when I went back to the UEFI/BIOS, it froze after 1-5 seconds, I would never be fast enough to increase the vcore or decrease the multiplier, forcing me to completely shut down my PC and hit the CMOS reset button on my MoBo to be able to restart. This means that for every trial and error attempt I will need to set up everything in the BIOS again.


this^ happens to me with my z77 GB board as well when i have a bad OC. I believe this is normal for GB boards... maybe you could try the other way round like set the voltage to around 1.35V and raise the multi until it crashes then give it some more volts.. probably max i would go with that cooler is 1.45V but depends on temps.
Going to be chaep
(16 items)
 
Wife's Rig
(12 items)
 
 
CPUMotherboardGraphicsRAM
i7-3770k Gigabyte GA-Z77X-UP5 TH MSI GTX460 G.Skill TridentX 2400 2 x 4GB 
Hard DriveHard DriveOptical DriveCooling
Plextor M5p 128GB 2 x WD 500GB Blues LG Prolimatech Megahalems Rev C 
CoolingOSMonitorPower
Arctic F12 CO PWM x 2 Push/Pull Windows 7 64 bit Ultimate LG IPS235V Corsair AX850 
Case
Coolermaster HAF 912 advanced (Asia version) 
CPUMotherboardGraphicsRAM
I5-3570K Gigabyte Z68X-UD3H-B3 Inno3D Ichill GTX660 8GB G.Skill RipjawsX 1600 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 128GB WD black 500GB Liteon Coolermaster Hyper 212+ EVO 
OSMonitorPowerCase
Win 7 ultimate Samsung Syncmaster 932gwe+ OCZ ZT series 550W Coolermaster 410 
CPUMotherboardRAMHard Drive
1.67Ghz Atom stock 1GB DDR2 2 x 500GB Western Digital Cavier Blacks [RAID 1] 
CoolingOSPowerCase
stock DSM3.2 [built on linux] stock stock 
OtherOther
Synology DX510 Expansion unit 5 x 1TB WD Cavier Blacks in a RAID 6 
  hide details  
Reply
Going to be chaep
(16 items)
 
Wife's Rig
(12 items)
 
 
CPUMotherboardGraphicsRAM
i7-3770k Gigabyte GA-Z77X-UP5 TH MSI GTX460 G.Skill TridentX 2400 2 x 4GB 
Hard DriveHard DriveOptical DriveCooling
Plextor M5p 128GB 2 x WD 500GB Blues LG Prolimatech Megahalems Rev C 
CoolingOSMonitorPower
Arctic F12 CO PWM x 2 Push/Pull Windows 7 64 bit Ultimate LG IPS235V Corsair AX850 
Case
Coolermaster HAF 912 advanced (Asia version) 
CPUMotherboardGraphicsRAM
I5-3570K Gigabyte Z68X-UD3H-B3 Inno3D Ichill GTX660 8GB G.Skill RipjawsX 1600 
Hard DriveHard DriveOptical DriveCooling
Samsung 830 128GB WD black 500GB Liteon Coolermaster Hyper 212+ EVO 
OSMonitorPowerCase
Win 7 ultimate Samsung Syncmaster 932gwe+ OCZ ZT series 550W Coolermaster 410 
CPUMotherboardRAMHard Drive
1.67Ghz Atom stock 1GB DDR2 2 x 500GB Western Digital Cavier Blacks [RAID 1] 
CoolingOSPowerCase
stock DSM3.2 [built on linux] stock stock 
OtherOther
Synology DX510 Expansion unit 5 x 1TB WD Cavier Blacks in a RAID 6 
  hide details  
Reply
post #5 of 5
Thread Starter 
I started having different issues and I'm a bit unsure if I should just a make a new topic. If so, let me know and it will be done.

I've now set my multiplier to 45x and my vcore to 1.25, 1.255, 1.26 and 1.265 and got the respective BSODs:

072813-4945-01.dmp 28/7/2013 00:26:10 DRIVER_OVERRAN_STACK_BUFFER 0x000000f7 fffff880`09a91fc0 00002b99`2ddfa232 ffffd466`d2205dcd 00000000`00000000 nvlddmkm.sys nvlddmkm.sys+17a496 x64 ntoskrnl.exe+75c00 C:\Windows\Minidump\072813-4945-01.dmp 8 15 7601 292,584 28/7/2013 00:27:18


072813-5085-01.dmp 28/7/2013 01:05:00 APC_INDEX_MISMATCH 0x00000001 00000000`777d170a 00000000`00000001 00000000`00000000 fffff880`0a1e4b60 ntoskrnl.exe ntoskrnl.exe+75c00 NT Kernel & System Microsoft® Windows® Operating System Microsoft Corporation 6.1.7601.18113 (win7sp1_gdr.130318-1533) x64 ntoskrnl.exe+75c00 C:\Windows\Minidump\072813-5085-01.dmp 8 15 7601 292,776 28/7/2013 01:06:11


072813-4867-01.dmp 28/7/2013 01:16:30 MEMORY_MANAGEMENT 0x0000001a 00000000`00041287 00000000`00000000 00000000`00000000 00000000`00000000 ntoskrnl.exe ntoskrnl.exe+75c00 NT Kernel & System Microsoft® Windows® Operating System Microsoft Corporation 6.1.7601.18113 (win7sp1_gdr.130318-1533) x64 ntoskrnl.exe+75c00 C:\Windows\Minidump\072813-4867-01.dmp 8 15 7601 292,528 28/7/2013 01:17:12

EDIT: Regarding this error, I will run Memtest86 after Prime95 and up the voltage to 1.55v instead of 1.5 if an error is detected.


072813-5569-01.dmp 28/7/2013 01:24:10 BAD_POOL_HEADER 0x00000019 00000000`00000003 fffff800`03660760 fffff800`03660760 fffff800`03660860 hidusb.sys hidusb.sys+4b17 x64 ntoskrnl.exe+75c00 C:\Windows\Minidump\072813-5569-01.dmp 8 15 7601 333,408 28/7/2013 01:27:25


The main issue is that they're all different and most are not part of any BSOD code list for Ivybridges I've found out.

I've decided to put everything on optimized defaults and adjust my memory from 1333 to 1600 mhz, 9-9-9-24, DRAM Command Rate set to 2 and voltage from Auto to 1.5v (while on auto it was already 1.5v) those are the tested values on the Corsair website and currently I'm on my 13th hour of a Prime95 torture test with no errors. I'm doing this as recommended by TotallyDubbed to test for faulty hardware. The max temperatures on Real Temp for this 13 hour torture test at 3.9 are so far 69 72 75 and 69. I have no way to measure the temperature but it's neither hot or cold, no heating or AC on, it's 19º outside so I would say perhaps 22-25º.

Because of the 1st BSOD I got, I ran Malwarebytes and found no issues and also uninstalled and reinstalled my nVidia 320.49 drivers as the code was referring mainly to malware or GFX driver malfunction. The BSODs I got after were all different but it seems to be all related.

Right now, with Prime95 running at 3.9 ghz (stock turbo) I'm at 1.212 - 1.224 - 1.236 voltage (this is on auto as it's default). I'm not sure if by those numbers alone you could say if my chip is below average and this is nothing but a voltage issue or if it could be something else.

I'd like to take screenshots of the UEFI for you to see all my settings when running the OC but I didn't want to stop the prime95. I saw on TotallyDubbed's video that a 24 hour prime95 run would be ideal.

Still, I can give you the things I changed on my Gigabyte Z77 UD5H guiding myself:

PWN Phase Control - eXm Perf (extreme)
CPU Clock Ratio - 35
Intel Turbo Boost Technology - Auto
Turbo Ratio for cores 1 2 3 4 - 45
Turbo Power Limits(Watts) - 300
Core Current Limit (Amps) - 200
C3/C6 State Support - Disabled (the C1E is enabled)
CPU Thermal Monitor - Disabled
CPU Vcore you already know, from 1.250 to 1.265 increasing 0.005 at a time
Tried reducing the CPU PLL to 1.7 but left it at 1.800v afterwards
The first crashed caused me to hang right as I entered Windows so I enabled then CPU PLL Overvoltage (was on Auto)
I noticed now on some videos, people change the Vcore Voltage Response from Auto to Fast, this is the only thing I haven't done.

I'll give you guys any information you wish and please, if I should make a new thread for this please let me know as well.

Thank you for taking your time reading this.

Vcore Loadline Calibration: Turbo (75% impact on the VDROOP) I didn't touch the CPU Vtt Loadline Calibration, not sure if I should have.
Edited by Furialz - 7/28/13 at 6:18am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Ocing 3770k causing the BIOS/UEFI to freeze