Haswell Overclocking Guides / Results

[profJim]

Re: Haswell Overclocking Guides / Results

That makes sense.

So how do you set adaptive on GB boards? I can only see an option for Offset in the BIOS.

Using GTL? EasyTune? Intel Extreme Tuning Utility (XTU)?

Does pressing CTRL + F1 do anything to unhide some settings?

[stasio]

Re: Haswell Overclocking Guides / Results

Hi TomazzZ.......welcome to forum.

Your GFlops in LinX is very low (should be arround 115-120).Setting is OK (no need 50 runs...20 is also OK),but which version you are running?
Did you run on Win7 SP1?

If you testing stability with latest AIDA64 3.00,be sure to set Preferences/General....Memory block size minimum 512MB.
Even 1 GB is better to full stress your memory together with CPU.

Not good setting I saw in Sin0822 Z87 OC Guide...20 runs finished in less the 5 min......proper 20 LinX runs take 1 hour+

Prime95 27.9 needs some updates for Haswell. Normally I would trust LinX or IBT only since it tell us the GFlops which shows the efficiency and accuracy of our overclocks. Eg.(sample under water):

4.4GHz @ 1.25v - 125 GFlops, 85C
4.5GHz @ 1.3v - 135 GFlops, 90C

The higher the GFlops, the more performance the overclock yield. So, 1.3v is picked as our preferred vCore.

That was on win7 SP1 with AVX instruction AND LinX with AVX libraries build 10.3.7.

On win7 non SP1 OR LinX non AVX, you would get:

4.5GHz @ 1.15v - 60 GFlops, 60C

lower GFlops and reduced performance of AVX supported apps and might sometimes get BSOD on routine works.

Thats why I said the hotter or the more vCore needed, the more stressful the app. Also, the more vdroop (lower load vCore) the more stressful.

Edit:
OCCT 4.4.0 (CPU Linpack) is also OK, if you set per pic bellow:



(untick 85C limit in setting)

[boondocks]

Re: Haswell Overclocking Guides / Results

From the Overclockers link in post #130:
Caveat – big, massive, honking, pay attention to this – caveat! When using adaptive voltage, the top of the curve isn’t necessarily the whole story. ASUS drove this point home when we met and I’m doing so now; we all need to do this as a public service to our users and readers. Even though you set 1.25 V as your maximum voltage, under certain very heavy loading conditions (i.e. stress testing), the voltage can and will exceed the maximum you have set. Let me say that again – if you stress test using adaptive voltage, even with the maximum set to 1.25 V, the CPU will – guaranteed – request and receive more voltage than you have told the motherboard to deliver it. It’s just the way this works.

In the ASUS demonstrations, setting a maximum 1.25 V in this scenario and then running something like AIDA64, Prime95, etc – any stress testing application designed to stress more than normal loads – the CPU requests and gets ~1.36 V. There is nothing you can do about this and it will happen whether you want it to or not. The only way to prevent stress testing programs from pulling extra voltage is to use a manually set voltage, which takes away your CPU’s ability to reduce voltage when idle.

All is not lost though! You probably noticed every time I told you what would happen it went hand in hand with stress testing. That’s because those are the only scenarios that will lead to this behavior. In all other circumstances so far as ASUS can tell, the CPU will cap at the set 1.25 V and never exceed it. Even if you’re Folding@Home it should maintain the 1.25 V cap (though those folding at home are under 100% load all the time and I’d suggest using a manual voltage for those machines just in case). Likewise, video encoding, audio encoding, compression, rendering – any CPU-intensive process, your CPU will maintain the 1.25 V cap.

So, now that you know the very important caveat, set 1.25 V as your maximum voltage with adaptive voltage, enable C1E and EIST (which throttle your multiplier and voltage when not under load) and go on with your merry life knowing your CPU is as relaxed as it can get when idle.

Better to follow Sin's guide than all that ASUS stuff. On the Gigabyte boards Extreme LLC will stop droop but the board is still going to allow some voltage increase with a set vcore.
If you keep all the "power saving" options on Auto the board will downclock and downvolt just fine at idle speeds because with Haswell voltage regulation is on the cpu.

[profJim]

Re: Haswell Overclocking Guides / Results

FWIW, it looks like MSI provides Adaptive voltage settings in some of their motherboards.
From the MSI Z87-GD65, Z87-MPOWER and Z87-MPOWER MAX motherboard manuals:

CPU Core Voltage Mode/ CPU Ring Voltage Mode/ CPU GT Voltage Mode [Auto]
Selects the control modes for these voltages.

• [Auto] This setting will be configured automatically by BIOS.
• [Adaptive Mode] Sets adaptive voltages automatically for optimizing the system performance.
• [Override Mode] Allows you to set these voltages manually.

I wonder if some of the higher end Gigabyte motherboards support adaptive cpu voltage control.
With Gigabyte motherboards, I would definitely follow Sin0822's recommendations.

[Acebmxer]

Re: Haswell Overclocking Guides / Results

From the Overclockers link in post #130:
Caveat – big, massive, honking, pay attention to this – caveat! When using adaptive voltage, the top of the curve isn’t necessarily the whole story. ASUS drove this point home when we met and I’m doing so now; we all need to do this as a public service to our users and readers. Even though you set 1.25 V as your maximum voltage, under certain very heavy loading conditions (i.e. stress testing), the voltage can and will exceed the maximum you have set. Let me say that again – if you stress test using adaptive voltage, even with the maximum set to 1.25 V, the CPU will – guaranteed – request and receive more voltage than you have told the motherboard to deliver it. It’s just the way this works.

In the ASUS demonstrations, setting a maximum 1.25 V in this scenario and then running something like AIDA64, Prime95, etc – any stress testing application designed to stress more than normal loads – the CPU requests and gets ~1.36 V. There is nothing you can do about this and it will happen whether you want it to or not. The only way to prevent stress testing programs from pulling extra voltage is to use a manually set voltage, which takes away your CPU’s ability to reduce voltage when idle.

All is not lost though! You probably noticed every time I told you what would happen it went hand in hand with stress testing. That’s because those are the only scenarios that will lead to this behavior. In all other circumstances so far as ASUS can tell, the CPU will cap at the set 1.25 V and never exceed it. Even if you’re Folding@Home it should maintain the 1.25 V cap (though those folding at home are under 100% load all the time and I’d suggest using a manual voltage for those machines just in case). Likewise, video encoding, audio encoding, compression, rendering – any CPU-intensive process, your CPU will maintain the 1.25 V cap.

So, now that you know the very important caveat, set 1.25 V as your maximum voltage with adaptive voltage, enable C1E and EIST (which throttle your multiplier and voltage when not under load) and go on with your merry life knowing your CPU is as relaxed as it can get when idle.

Excellent info! The only time I have see my gigabyte board do something similar to this is when setting offset vcore. At 4.5ghz in bios with vcore set to normal and offset +0.000 I get 1.111v +075v = 1.198ish in windows at light load. Under full load vcore will jump to 1.325. I have tried dropping llc from extreme to turbo and no difference on full load vcore.

Better to follow Sin's guide than all that ASUS stuff. On the Gigabyte boards Extreme LLC will stop droop but the board is still going to allow some voltage increase with a set vcore.
If you keep all the "power saving" options on Auto the board will downclock and downvolt just fine at idle speeds because with Haswell voltage regulation is on the cpu.

On my z87-OC board I have never been able to get vcore to drop at idle when using fixed vcore. I have left everything on Auto, Ive tried Turbo disabled, c1e enabled, c3 enabled EIST enabled and c6/7 disabled and any other combition I have see floating around on the forms that other say worked for them but has yet to work for me.

FWIW, it looks like MSI provides Adaptive voltage settings in some of their motherboards.
From the MSI Z87-GD65, Z87-MPOWER and Z87-MPOWER MAX motherboard manuals:

CPU Core Voltage Mode/ CPU Ring Voltage Mode/ CPU GT Voltage Mode [Auto]
Selects the control modes for these voltages.

• [Auto] This setting will be configured automatically by BIOS.
• [Adaptive Mode] Sets adaptive voltages automatically for optimizing the system performance.
• [Override Mode] Allows you to set these voltages manually.

I wonder if some of the higher end Gigabyte motherboards support adaptive cpu voltage control.
With Gigabyte motherboards, I would definitely follow Sin0822's recommendations.

z87-OC only has fixed vcore and offset vcore. I have never been able to get voltage to drop at idle with fixed vcore. Offset vcore will overvolt at full load.

[TomazzZ]

Re: Haswell Overclocking Guides / Results

Hi TomazzZ.......welcome to forum.

Your GFlops in LinX is very low (should be arround 115-120).Setting is OK (no need 50 runs...20 is also OK),but which version you are running?
Did you run on Win7 SP1?

If you testing stability with latest AIDA64 3.00,be sure to set Preferences/General....Memory block size minimum 512MB.
Even 1 GB is better to full stress your memory together with CPU.

Not good setting I saw in Sin0822 Z87 OC Guide...20 runs finished in less the 5 min......proper 20 LinX runs take 1 hour+

Prime95 27.9 needs some updates for Haswell. Normally I would trust LinX or IBT only since it tell us the GFlops which shows the efficiency and accuracy of our overclocks. Eg.(sample under water):

4.4GHz @ 1.25v - 125 GFlops, 85C
4.5GHz @ 1.3v - 135 GFlops, 90C

The higher the GFlops, the more performance the overclock yield. So, 1.3v is picked as our preferred vCore.

That was on win7 SP1 with AVX instruction AND LinX with AVX libraries build 10.3.7.

On win7 non SP1 OR LinX non AVX, you would get:

4.5GHz @ 1.15v - 60 GFlops, 60C

lower GFlops and reduced performance of AVX supported apps and might sometimes get BSOD on routine works.

Thats why I said the hotter or the more vCore needed, the more stressful the app. Also, the more vdroop (lower load vCore) the more stressful.

Edit:
OCCT 4.4.0 (CPU Linpack) is also OK, if you set per pic bellow:



(untick 85C limit in setting)

Thanks Stasio ,

I went up to vcore 1.18 at 4.6 , linx AVX (Linpack 10.3.7.012) geting about 120 GFlops
Temp is very high ...
Just wondering is it normal , to have max at 92-93C under full load on air at 1.18 Vcore with average cooler like Cooler Master V6 or even for this cooler it is too high ,
maybe 2 years old thermal paste can be an issue ?

How hight should be the temp on average air with 1.18 4.6 under linx AVX ???


Thanks

[Acebmxer]

Re: Haswell Overclocking Guides / Results

Excellent info! The only time I have see my gigabyte board do something similar to this is when setting offset vcore. At 4.5ghz in bios with vcore set to normal and offset +0.000 I get 1.111v +075v = 1.198ish in windows at light load. Under full load vcore will jump to 1.325. I have tried dropping llc from extreme to turbo and no difference on full load vcore.




On my z87-OC board I have never been able to get vcore to drop at idle when using fixed vcore. I have left everything on Auto, Ive tried Turbo disabled, c1e enabled, c3 enabled EIST enabled and c6/7 disabled and any other combition I have see floating around on the forms that other say worked for them but has yet to work for me.




z87-OC only has fixed vcore and offset vcore. I have never been able to get voltage to drop at idle with fixed vcore. Offset vcore will overvolt at full load.

Ok I stand corrected. vcore is dropping at idle, its cpuz that does not show the drop. I just verified this with a DMM.

[profJim]

Re: Haswell Overclocking Guides / Results

See if the latest beta version of HWINFO64 does a better job of showing accurate voltages.
HWiNFO64 Download

[wazza300]

Re: Haswell Overclocking Guides / Results

@ ace switch to power saver profile in windows,cpu-z will show correct idle clocks then

[Acebmxer]

Re: Haswell Overclocking Guides / Results

I switched to version 1.64.0 with the OC skin and that version will show vcore drop at idle.

[ehume]

Re: Haswell Overclocking Guides / Results

I did a web search for cpu-z 1.64.3 , dl'd it, and everything is OK on idle.

[Acebmxer]

Re: Haswell Overclocking Guides / Results

how low can you go at idle?????

[stasio]

Re: Haswell Overclocking Guides / Results

I did a web search for cpu-z 1.64.3 , dl'd it, and everything is OK on idle.

Web search for CPU-Z 1.64.3?
It's already 3 weeks in my sticky here.

how low can you go at idle?????

I saw very close to 0 (zero),somewhere.

[Acebmxer]

Re: Haswell Overclocking Guides / Results

Yeah for some reason 1.64.3 would not show idle voltage, always fixed vcore. 1.64.0 with OC skin show me idle voltage. Some times you may have to reopen it, max 3 times (so far for me) for it to show idle voltage. Most of the time you need to leave it open for 20 sec or so before it will show idle voltage.

[Acebmxer]

Re: Haswell Overclocking Guides / Results

Video of DMM vs CPU-z voltage. LLC was set to extreme.