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

[Martin.v.r]

voltset whit max MHZ ? on cpu try 5.3 ore 5.4
it becomes difficult after 5.1

[mean6]

Hmm my OC is perfectly stable (Linx using AVX instructions on Win7 SP1 runs for hours, no issues with memtest, and gaming for hours never crashes either). I assumed the sleep issue was due to PLL enabled since the board sleeps fine at stock speed, but not when I overclock to 4.9 GHz (only requiring enabling PLL and bumping the CPU voltage in the UEFI). Perhaps a higher multiplier is the cause then.

Just an update. I have been trying to get past the 4.8 oc, but as soon as I go over the multi of 48, no matter how many volts etc, I can get it linx stable, but like the you said, it will only wake to a black screen. I am just assuming it means its not stable. Just from past experiences, this is my guess. It's funny though, I am currently running @4.8 @1.365 v, and can't get passed this mark. Any suggestions? Could it be just a bios bug?

[metalion71]

I have a P67 Extreme6 mean6 but i think the definitive item to overpass 4,8 Ghz(with my 2500k) was to have the internall pll overvolt in enabled and the core current limit in 300.The voltages you are saying are damm sweet and is possible you can raise your cpu to 5 ghz with no problem at all.Beyond it it can be a mistery...but you are almost touching heaven with the pulps of your fingers.
And of course,if Asrock guys read this tread,please,add oore memory dividers.I hate the big step betwen 1888 and 2133 because I have a 2000 kit.Except that,the motherboard is the best bang for the buck i got in ages.Nice and impressive work Asrock...hope to se you again when socket 2011 is launched.

[gurneyhck]

@Clockz, 2600k with HT on results indeed in lower Gflops, so he is fine.

@Mysterfix, "additional turbo voltage" is a setting where your motherboard will give your cpu extra voltage (according to what you set there) when he goes into turbo mode. So this is a very useful setting when you still want to use all powersaving features.
Example: you can have while browsing a cpu speed of 1.6GHz@0.85V and under full load it will increase to 5GHz@1.4V

[Cy4n1d3]

Using Offset Voltage over 0.100 doesn't have any effect for me, anyone else noticing this?

[ClockZ]

I'm playing with the offset voltage features and working on a stable 5ghz clock, I like the fact that it backs down the voltage when not needed. One thing I have noticed is that LinX is very easy to make passes on and shows stable but Prime95 running blend or large FFT im having a hard time passing more than 5 loops before a core drops out??? And it seems to drop a core no matter where im at clock speed and voltage wise, anyone else have this?

[Cy4n1d3]

Are you using LinX with AVX? This is definitely more demanding than Prime.

[ClockZ]

Are you using LinX with AVX? This is definitely more demanding than Prime.

Nope,I'll download SP1 and do that at some point today....not a big fan of all the stress testing more interested in being game stable at my settings, but I will try that out to see how it goes.

[Cy4n1d3]

Just a suggestion. If your system survives LinX with AVX and max mem setting, Prime will not be a problem anymore, believe me

[mean6]

I have a P67 Extreme6 mean6 but i think the definitive item to overpass 4,8 Ghz(with my 2500k) was to have the internall pll overvolt in enabled and the core current limit in 300.The voltages you are saying are damm sweet and is possible you can raise your cpu to 5 ghz with no problem at all.Beyond it it can be a mistery...but you are almost touching heaven with the pulps of your fingers.
And of course,if Asrock guys read this tread,please,add oore memory dividers.I hate the big step betwen 1888 and 2133 because I have a 2000 kit.Except that,the motherboard is the best bang for the buck i got in ages.Nice and impressive work Asrock...hope to se you again when socket 2011 is launched.

I have tried both those settings, and they will run fine through both Linx, and Prime using Windows 7 SP1. The problem is having the black screen after coming out of sleep mode. It seems that any multi after 48, no matter how stable the OC is with stress tests, it just wakes to a black screen. Can anybody else confirm this?

[Cy4n1d3]

I have no problems waking up my pc...

[Mysterfix]

I'm playing with the offset voltage features and working on a stable 5ghz clock, I like the fact that it backs down the voltage when not needed. One thing I have noticed is that LinX is very easy to make passes on and shows stable but Prime95 running blend or large FFT im having a hard time passing more than 5 loops before a core drops out??? And it seems to drop a core no matter where im at clock speed and voltage wise, anyone else have this?

Have you checked your ram for stability? I was having the same problem getting instant fail in blend and large fft running prime. I upped my ram voltage and all seems fine now. I seem to have gotten some bad ram, there rated for 1.35v but I have to use 1.45v for stability at default settings.

[dnottis]

Just a suggestion. If your system survives LinX with AVX and max mem setting, Prime will not be a problem anymore, believe me

I wish I could agree. I passed linx and IBT, prime would reboot me and games would BSOD me. I needed a bump in vcore to be stable in Prime and games over what IBT / AVX and LinX was giving me.

There have been others that found they were crashing in games after passing hours of stress testing. These chips are different than anything we've seen.

[Cy4n1d3]

Well I could pass Prime Custom/Small/Blend + Linx w/o AVX, run Black Ops or World of Warcraft for hours but Linx with AVX was still failing on me, whereas Prime stability most of the time didn't mean game stability for me..

[Martin.v.r]

form 5.2HGZ an an up 2134MHZ on mem ? try 2T

[Ket]

I've had more information passed onto me from Asrock and they have been able to verify half of the issues I've highlighted regarding UEFI 1.5 for the Extreme4. I've forwarded further information to Asrock regarding the issues their not sure on to help them identify and fix the problems. Currently, UEFI 1.42A still remains by far the most problem free for Extreme4 owners.

[swampy]

I've had more information passed onto me from Asrock and they have been able to verify half of the issues I've highlighted regarding UEFI 1.5 for the Extreme4. I've forwarded further information to Asrock regarding the issues their not sure on to help them identify and fix the problems. Currently, UEFI 1.42A still remains by far the most problem free for Extreme4 owners.

What do you think of version 1.51?

[Mysterfix]

What do you think of version 1.51?

What is the proper procedure for flashing the UEFI? I had a hell of a time going back to the 1.42a! I flashed it several times but the 1.5 bugs stayed with me. I finally got it to work after clearing cmos and flashing 4 times back to back. The 1.51 seemed to be pretty good until I tried messing with the timings on my ram.

I must have caused a melt down by doing that because I got a no post, reset and all my settings were gone. After that the bugs started running wild and nothing was working. I couldn't even get it to boot with the ram at stock settings. So I loaded the UEFI default settings, rebooted and started my flashapalooza! It took a while to get rid of the bugs though.

[Ket]

What do you think of version 1.51?

Theres very little data available for 1.51 and I'm unable to find any kind of changelog. From what I know, 1.51 is 1.5 just with the LED options. Based on that, stay with 1.42A.

What is the proper procedure for flashing the UEFI? I had a hell of a time going back to the 1.42a! I flashed it several times but the 1.5 bugs stayed with me. I finally got it to work after clearing cmos and flashing 4 times back to back. The 1.51 seemed to be pretty good until I tried messing with the timings on my ram.

I must have caused a melt down by doing that because I got a no post, reset and all my settings were gone. After that the bugs started running wild and nothing was working. I couldn't even get it to boot with the ram at stock settings. So I loaded the UEFI default settings, rebooted and started my flashapalooza! It took a while to get rid of the bugs though.

Whenever flashing the UEFI It's wise to take the following steps and things should go without a problem;

1. Load UEFI defaults, save and exit
2. Update UEFI either through DOS or built-in flash program (in the cases where a DOS flash isn't possible)
3. Load UEFI defaults after the flash
4. Remove battery, power cord, and set CMOS jumper to "clear". Leave for a minute or so.
5. Power up.

That routine will ensure the "cleanest" flash possible.

[Mysterfix]

I will remember that for next time, thank you.

[burebista]

ASRock P67 Extreme4 UEFI 1.53
Changes.

[radier]

People report that 99,8MHz BCLK issue is still present with this release. I will wait for some official BIOS.

[Cy4n1d3]

They say SS-off bug is not that easy to fix as it is a bug inside Intel's MRC/ME code. Fixing this is going to lower bclk overclockabiliy they told ASrock_User. So don't keep your hopes too high :\

[radier]

I just got an answer from Taiwan.

Since change Spread Spectrum and overclock CPU frequency will relate to clock generator behavior, full reset once is necessary for system stability.
So once you change settings and an error occurs, you would have to reset the BIOS, in order to work properly.

[Bei Fei]

Seems there is a v160 available which fixes the ss issue.