Asrock P67 Extreme4 UEFI & Drivers thread [Extreme6 & Pro3 owners welcome]

[burebista]

But using Real Temp 3.67, whenever I launch some stress test, be it Prime95, Linx, OCCT the VID reading goes from ~1.22v at idle to 1.366v at full load!

RealTemp reads a register for that VID, it's not a sensor value.
For (almost) accurate software voltage readings I'd trust CPU-Z/whatever program who read a sensor not a register set by Intel.

[Caparroz]

RealTemp reads a register for that VID, it's not a sensor value.
For (almost) accurate software voltage readings I'd trust CPU-Z/whatever program who read a sensor not a register set by Intel.

Thanks.

I was suspecting that. I was reading about how it reads that register from Sandy Bridge chips and how Real Temp uses it to calculate the CPU power consumption.

Well, that's great. Testing at 45x with 1.235v now for a stable 24/7 OC, so far so good. I think I may have a good clocker in my hands now. Being through a couple of really bad 2500k til now. I'll test max clocks later this week to see if I can get past 5ghz with sane vcores...

[Ket]

Thank you Ket.

C-state Enable?

No C-State. I have all CPU power saving features off as I run BOINC.

Ket I don't belive that you need as much CPU PLL and VTT voltage.
1,709V for PLL
and 1,1V for VTT because you have high clocked RAM would be sufficient
VCCSA should be leaved on Auto (0,925V) as it has no impact on stability.

No piece of hardware is the same. I've used lower voltages and all I get for my troubles is BSODs. People really do need to remember nothing is set in stone and everything is variable based on many.. well variables

Another happy (for now... ) Extreme4 user here.

I'm using the latest 1.70 bios from asrock site. I did some minor adjustments in the UEFI:

2500k @ 45x w/ fixed 1.275v set in bios, PLL ov off, Turbo on, all power savings/C-states on, LLC at lv2, all other volts settings manually set but pretty much default.

Everything looks OK, temps are great but one thing is bugging me. UEFI/CPU-Z/HWmonitor/etc. all read the vcore as expected with given settings at idle and load.

But using Real Temp 3.67, whenever I launch some stress test, be it Prime95, Linx, OCCT the VID reading goes from ~1.22v at idle to 1.366v at full load! No other SW reading does that, they all stay at ~1.27v. I know SW readings are not fully accurate but that Real Temp load VID reading is really bugging me. I don't know for sure if I'm stable at 1.275v or 1.366v...

I read this thread from start to finish, searched the Real Temp thread but found nothing like what I'm getting here.

Any ideas?

I get the exact same thing.

RealTemp reads a register for that VID, it's not a sensor value.
For (almost) accurate software voltage readings I'd trust CPU-Z/whatever program who read a sensor not a register set by Intel.

Not necessarily. I've observed the exact same spikes and at the point of those spikes temps have also spiked and not just in RealTemp, but in Asrocks AXTU software as well, though no other software registers the VID spikes.

[pythonse]

dual..or maybe triple boot this morning...ugh

[Ket]

Clear UEFI and load defaults. Something is deffinately wrong, perhaps too high memory OC or not enough vDIMM?

[[XC] Synthetickiller]

I plan on picking up an Extreme4 to go along w/ an i7 2600k. I'd prefer the 6, but it has more bugs.

I was going to do air cooling, but I have so much WCing gear, I did a 4 hour search and finally found a Kryos HF copper top, so I should be sitting at nearly ambient, lol.

Just warning you guys since I'm an 1155 n00b.

[Ket]

Don't worry, I got your back I offered to bug test the Extreme6 if they can send me a board when I emailed one of my contacts but that contact hasn't replied yet. No skin off my nose eitherway, but it would be nice to give people more options than just the Extreme4.

[EzyRyder]

Don't worry, I got your back I offered to bug test the Extreme6 if they can send me a board when I emailed one of my contacts but that contact hasn't replied yet. No skin off my nose eitherway, but it would be nice to give people more options than just the Extreme4.

Hi Ket,

Are you going to be supporting Z68 Extreme 4 users here or do we need another thread? I am asking because I posted a message a few days ago and I didn't get an answer.

Basically I am having a few problems with eSATA and Waking up from Sleep in my Z68 Extreme 4

By the way I did swap the TIM in the PCH!

Cheers!

[Ket]

I'm sure people will offer what advice they can in this thread. The Z68 is practically identical to the P67 after all. As to your specific issue as I haven't used the Asrock Z68 board I can't really say what your problem could be.

[EzyRyder]

THanks ...

I have read something about not being able to wake back from sleep on P67 boards. Is this something to do with the overclocking? my overclock is only a multiplier change, I haven't actually touched the over-voltage options yet...

and no one had eSATA problems with P67?

[pythonse]

Clear UEFI and load defaults. Something is deffinately wrong, perhaps too high memory OC or not enough vDIMM?

should i reset cmos after bios update? if so when do i do it? when the pc is off and unplugged from the wall?
also my vdimm or ram voltage is 1.6 and its 1.5 stock..should be plenty and 1833 mhz is ok with the ram i have

[Ket]

THanks ...

I have read something about not being able to wake back from sleep on P67 boards. Is this something to do with the overclocking? my overclock is only a multiplier change, I haven't actually touched the over-voltage options yet...

and no one had eSATA problems with P67?

It could be the same problem experienced by P67 owners in that case, AFAIK intel are aware of the problem and working on a fix.

should i reset cmos after bios update? if so when do i do it? when the pc is off and unplugged from the wall?
also my vdimm or ram voltage is 1.6 and its 1.5 stock..should be plenty and 1833 mhz is ok with the ram i have

Assuming you already updated the UEFI;

- Pull power cord, remove CMOS batter, set jumper to "clear"
- leave for a few mins, then power up

Keep everything at stock and if you don't experience any problems, something was wrong with your OC.

[pythonse]

thx ket...jumper has to be set to the right i believe, ill get the manual out
i take it u cant just hit the cmos clear button?

[radier]

Press button for ~15s.

[Ket]

Pressing the Clear CMOS button won't fully clear the UEFI, some settings will still be retained unless the battery and jumper method is used.

[radier]

Please....

[Ket]

Thats not any kind of response, but to clarify; Press the Clear CMOS button then enter the UEFI. Time, date, year, etc will still be as they should be. I never checked to see what other settings were retained, but those were. So again, using just the clear CMOS button will not fully reset the UEFI, its more of a hardware load setup defaults button.

[bat4]

Indeed, the CMOS clear button doesn't clear everything. Removing the battery is preferable for a full CMOS reset.

[[XC] Synthetickiller]

Well, I bit the bullet and ordered one.

Ewiz has them onsale for $150 ($10 promo code) if you're in the US. Heck of a price for an Extreme4.

I'll need to do some reading on voltage settings for this guy before I start asking lazy questions.

[Ket]

Scroll back a page or 2, I upped some screens with base values that'll get anyone going

[BranFB]

Yeah.
If your cpu is good.
any voltages will do it

[Ket]

Been talking with one of my contacts again and I was informed that Asrock sent some devices to Etron 3 weeks or so ago and they have been looking at the problem but as yet don't seem to have a solution. I'll let you know as soon as I know any more info.

[paschov]

Hello everybody! I'm reading this forum for months already, but this is my first post. I have P67 Extreme4 B3 with 2600K chip and G.Skill P67 memory. The day before yesterday I found Ket's settings on page 46, tried them, they worked perfect, I even upped clock to 46,ran Intel Burn Test without any problem. Everything was OK, no double boot. I was happy like an elephant, shut down my rig, went to sleep. Next morning, no cold bug (YESSS!), BUT: blinking prompt in left upper corner of display. And since that time, no matter which settings and BIOSes I try (I tried quite a bunch, believe me!) on any multiplier higher than 43 I get a blinking prompt. This chip was working on old B2 MB on 48 multiplier without a problem. What should I do? RMA the MB? The chip? Both of them? Sell RMA'd parts on fleebay and forget about Sandy Bridge? Any help will be highly appreciated. Thank you.

[Ket]

Hmm... that issue is starting to become a reocurring problem. If many more users report that on B3 boards I think I'll have to ask my contacts about it. For now I would advise the following;

- Load UEFI defaults, save and exit.
- Download UEFI 1.3 (DOS version) and flash to it
- Update UEFI to 1.6A, 1.64A or 1.7
- Shutdown after the update completes successfully
- Remove power cord, CMOS battery, and set the CMOS jumper to "clear"
- Leave for a few mins and press power button 3-4 times
- Boot back up and see what happens.

[pythonse]

ok so do u leave the jumper still in the clear position and reboot? or...do u put it in the clear position for a few minutes and then put it back,plug in the system and reboot?