Asus P8P67 Hangs after being idle for some time

Mine did it twice yesterday, all with the lastest drivers installed...
 
Yes it's really strange... I hope too to see this update on the Intel's website !
 
No freezes so far since I disabled both C3 and C6 reporting in the BIOS.
--Mark--
 
Now up rock solid for 4 days including 3 nights idling.
With no sleep in Windows but with monitor power saving after 10 mn
C3/C6 disabled

Starting to sound good :cool:

I have spent quite a lot of time on S0-S5 Windows OS states and C0-C6 CPU states. Not easy to find information but I believe I am starting to get an idea about how this works :confused:
 
If Wikipedia is up-to-date, the C3 and C6 report states are used for a "deep sleep" of the CPU and the C1E state for a "cool sleep" (without loosing of the cache and others).

So maybe we could think about an incompatiblity between Windows (or the chipset drivers from Intel) and the reports generated by this new hardware ?

I had a similar hang issue corrected by uninstalling the RST (probably an incompatibilité Samsung F3/RST for the moment). But with yours reports and if I take a look closer to my installation, I have always some softwares in background (e.g. PS3 Media Server), so maybe in my case the cpu never report a C3 or a C6 state to the OS !

So, just in case, it's maybe a good advice to desactivate the C3/C6 report for everyone ?
 
Well actually if understand correctly it is far more complex than this :rolleyes:
Let me try to give a short explanation (may be incorrect on some points) .

On one hand we have the CPU c-states. C0 is full active and C6 is deep deep sleep with wake-up time inversely proportional.
On previous release of Intel ix core the turbo boost mode was fired up when one or more cores where sleeping. So if 3 cores where in C3/C6 states then the forth core would get a bump of 4x. Therefore with this generation of CPU it was goos to report C3/C6 to get benefit of turbo boost. But this also caused trouble on many mobos. Reason beeing that the mobos needs to "collaborate" to enter C3/C6 and this is not always implemented correctly.
Now with Intel ix second generation (Sandy Bridge) the TBoost is in version 2 and if remember correctly from Intel IDE presentation it is more based on power consumption than on core usage. In other word you can have a core "turbo boosted" even if the three others are also working... What that means is that it looks less critical to report C3/C6 to the OS than it used to be on previous generation. So until Asus get it right (?) it is probably a good idea to disable reporting (wont see much difference in perf / power saving).

Now on the other hand we have the power saving state of the OS as defined by ACPI spec. They range from S0 (full active) to S4 (hibernation) and even S5 (turned off). Apparently Windows 7 implement S1, S3 (sleep), S4. S3 and S4 can be activated or not in power management of Windows. Also S1 state is supported by Windows (type powercfg /a to see all states) I do not know if it used and when?

The last piece of the puzzle is to find out how windows uses the CPU C-states. First of all you need to understand that S-States are not directly related to C-States. It is almost impossible to know in details how Windows uses the C-states but we have some hints.
Here are my understandings/speculations on the subject:
Windows is a preemptive multi-tasking OS and therefore it uses a scheduler to dispatch the tasks/threads to be executed to the cores available. So if there is no much things to do you will have several threads assigned to one core and the rest of them sleeping. Even on heavy load, for example if you run super-pi, you can have one core at 100% and the other ones sleeping.

Ok this is a somewhat speculative and simplified explanation but it should help to understand the subject....

By the way as I already mentioned previously when you disable some of the c-states reporting to the OS that means that OS will not use these states (which is useful if they are not correctly implemented in the mobo). The C3/C6 reporting has been turned on by default only in the latest BIOS version. This indicates that Asus believe they support these states correctly, but my tests seems to show that turning them off greatly improve stability especially on the "hanging problem". Plus you still getting TB V2 even if they are disabled....

Hope this makes sense :)
 
Thanks DrCoolZic,

That does make sense.

C3 & C6 reporting are staying off for me!

This CPU idles cool enough that my system is "green" enough for me without individual cores getting turned off.
 
Thanks a lot for your really greats explanations!

So the only doubt is about the activation of the TurboBoost v2 on used cores without the C3/C6 states reported by others cores? Unfortunatly, The TurboBoost v2 seems to only add this functionnality :

"In the Sandy Bridge microarchitecture this technology (i.e. TBoost) was improved in order to allow the CPU to exceed its TDP (Thermal Design Power) for up to 25 seconds – i.e., to dissipate more heat than officially allowed."

http://www.hardwaresecrets.com/article/Inside-the-Intel-Sandy-Bridge-Microarchitecture/1161/5

This information about the TurboBoost Technology may be also valid for the TurboBoost v2 Technology, so it seems than it will not be used with the C3 state report disabled :

"When the operating system instructed one of the active cores to enter C3 sleep state using the Advanced Configuration and Power Interface (ACPI), the other active core(s) dynamically accelerated to a higher frequency."

http://en.wikipedia.org/wiki/Intel_Turbo_Boost

But maybe the activation of the TurboBoost v2 with the C3/C6 report disabled can be tested with the Intel Turbo Boost Technology Monitor software?

http://www.intel.com/support/processors/sb/CS-031038.htm

Thanks again for sharing your researchs!

Edit: add a link to the Intel Turbo Boost Technology Monitor software.
 
Last edited:
I am also running stable with uptime since feb. 4th last week with the last changes at C3/C6 disabling.

Still want to undo all the other changes I made so that I can be 100% sure it is that & only that.
 
spliffrider said:
I am also running stable with uptime since feb. 4th last week with the last changes at C3/C6 disabling.

Still want to undo all the other changes I made so that I can be 100% sure it is that & only that.
Click to expand...
For your information I have reinstalled from scratch a genuine Windows 7 X64 US on a separate HDD from which I can boot. This partition is kept with bare minimum programs.
I have installed the latest of all the drivers we have talked about here.
Installed AI suite from CD and updated to 1.01.14.

The only interesting settings changed from default BIOS are
- AI overclock set to XMP so I use my memory at spec 1600
- PLL overvoltage disabled
- C3/C6 report disabled

In windows I have only turned off sleep/hibernate as well as disk & set turn off display: 10mn. I did not changed any other advance power saving options (USB, LAN, ...).
So this is an easy setting no mysterious changes to be performed :)

And with these settings everything works great :cool:

So now I am very satisfied with the current status of my P8P67.
We will see what the new BIOS 1035 and the new driver from Intel will bring ...
 
I'm up to two weeks now completely stable with C3 and C6 off, no bad effects that I've noticed, this box is on at least 15 hours a day in that time frame. I've left it up overnight trans-coding movies a couple times as well, no more issues. :) Max temps 70c at max 1.4volts, typically 33c, it is certainly scaling the CPU like it should.
 
ibnarabi said:
I'm up to two weeks now completely stable with C3 and C6 off, no bad effects that I've noticed, this box is on at least 15 hours a day in that time frame. I've left it up overnight trans-coding movies a couple times as well, no more issues. :) Max temps 70c at max 1.4volts, typically 33c, it is certainly scaling the CPU like it should.
Click to expand...
You must be overclocking. With CPU fully loaded (prime95) max temp is about 55C @3.8GHz.
This morning the system was at about 22C (idle for night)
 
Hi everyone,

About my precedent post in this thread, I'm sharing with you the results of my tests to try to understand :

How the C3 (and C6?) Report is interacting with the TBoost v2?

My configuration

- i5 2500K @ Stock speed : 3,3Ghz - Box Turbo Boost annonced : 3,7Ghz
- Asus P8P67 Pro Motherboard with Bios 1204 (1253 did no difference here)

The softwares used in my test were

- IntelBurnTest v2.50 with 1, 2, 3 and 4 threads.
- Intel Turbo Boost Technology Monitor software v2.0.

Just a brief reminder of the TBoost v2 with 1/2/3/4 core(s) used



Here are the results

With the CPU C3/C6 Report activated in bios with 1/2/3/4 core(s) used:






With the CPU C3/C6 Report desactivated in bios with 1/2/3/4 core(s) used:






Sadly it seems my fear explained in my precedent post is real, without the C3 Report enabled, the TBoost v2 doesn't activate the others 100Mhz steps of TBoost accessible with less than 4 cores.

To resume, the steps after the initial TBoost of 100Mhz with 4 cores won't be activated with only 3/2/1 core(s) and the C3 Report disabled. And the maximum TBoost reachable will be +100Mhz in this case and not +100Mhz for each core in C3 State. It's only a +100Mhz increase and not the normal +400Mhz recheable of the TBoost with one Thread.

Of course, it's preferable to avoid an hang issue than having a TBoosted CPU with some hangs. But the tests are showing than these bios tweaks aren't without consequences in term of performances in some circumstances. I don't know if others brands of Motherboard have the same problem, but if it's the case, it's probably a Win7/Intel drivers problem which I hope will be fixed soon.

Note : I have only done my tests with the C3 and C6 Report desactivated because I didn't experiment the hang issue and I wanted to place me in the same condition than yours. So finally, I have a question for people who solve the hang issue with the C3/C6 Reports desactivated : Did you make a try with only the C6 Report desactivated? Maybe it's only this parameter which is faulty? If the hang would disapear too, it will be a good compromise to avoid the hang issue and to keep the TBoost fully functional.

Hoping to have done, even a small, step forward in understanding this problem and his consequences. Have a nice day!


Edit: Add a link to my precedent post.
 
Last edited:
Thanks very much for the information this quite interesting.
I was under a wrong impression on improvement of TB V2 compared to TB V1!

What I will do, as you suggested, is to try to re-enable C3 and see if system does not freeze.

In fact I think that the most important factor in solving the issue was to disable the PLL overvoltage. I believe I have disabled the C3/C6 at the same time and it may not be the causing the problem after all?

So many thing has changed since day one (specially better drivers and BIOS) that disabling the C3/C6 is may be not needed.

PS: I also made some test with TB monitor and was disappointed it does not show TB on each core :(

do you have a pointer to the Intel Burst test (or I just google :) )
 
Hi DrCoolZic!

For the IntelBurnTest I used, here is the official link : IntelBurnTest v2.50

It seems that many people in this thread have resolved their hang issue with only the C3/C6 Report desactivated.

So maybe in your case (and to be sure), you could leave the PLL overvoltage on AUTO (if I remember, the settings are auto/enabled/disabled?) and keep the C3/C6 disabled. And after that, and if all is fine, you could re-enable the C3 Report ?

Oh and yes, TB monitor is really ... basic!

May the good settings be with you! :p

Edit: I have made some tests with only the C3 Report enabled and the C6 Report disabled, the results are exactly the same as the C3 and C6 enabled. It was just to confirm that only the C3 Report is used by the TBoost functionality!
 
Last edited:
Sundae said:
Hi DrCoolZic!
So maybe in your case (and to be sure), you could leave the PLL overvoltage on AUTO (if I remember, the settings are auto/enabled/disabled?) and keep the C3/C6 disabled. And after that, and if all is fine, you could re-enable the C3 Report ?
Click to expand...
:confused: Are you running with PLL set to Auto :confused:

This is recognized as a buggy feature by Asus representative and causing all sort of problems
This is the first thing they recommend to disable ???

And thanks for the pointer
 
Wow! Yes I'm on AUTO setting!!! Could you give me a link about this concern? Maybe it's because I didn't give a try to overclock my CPU for the moment that I didn't have any problem with this setting ? :eek:

Hey, it's my turn, thanks for the advice!
 
Sundae said:
Wow! Yes I'm on AUTO setting!!! Could you give me a link about this concern? Maybe it's because I didn't give a try to overclock my CPU for the moment that I didn't have any problem with this setting ? :eek:

Hey, it's my turn, thanks for the advice!
Click to expand...

If you are overclocking with Internal PLL Voltage on auto, it will normally enable this feature once you are past a 44x multiplier or have set Bclk past 103. Once this is enabled, most sleep/hibernate routines will not work properly, this is internal to the processor although Intel is working on a possible solution.

As for the C-States, depending on the system configuration (OC, PM, etc), disabling C3/C6 will generally allow for improved sleep resume activities. In the 1302 UEFI build arriving in the next few days (hopefully sooner) we will set those to disable once you manually overclock the system or if you run an auto overclock and it exceeds the rules we have designed. This should assist those users who are overclocking and have resume problems. Also, you might see a 1% bump in 3DMarks and HD benches. ;)
 
Whew! Thank you Gary for your clarification! So leaving it on AUTO without an overclock keep it disabled. But I will right now desactivate it by security.

For the C-States, you will disable the C3/C6 by default only on manual O/C or exceeded auto-O/C and not by default at Stock Speed? Am I right? (Otherwise it will desactivate the TurboBoost by default!)

Thanks again for your support Gary!
 
I've had this happen several times also. I run Media Center for my TV and it lasts all night no problem (leave TV on). I shut it down before I leave for work, then I come home and find there's no power to my keyboard and mouse (none of the lights are on), and no signal to the monitors. It's like it went into a sleep mode and is unable to wake (even though it's setup not to sleep or hibernate in windows).

This is with no OC, and my DDR3-1600 (ADATA 2x4 GB) running at 1333
 
Last edited:
Just a suggestion, I had a similar problem on an older build. I notice you are running raid.

I had a similar problem on a P5Q motherboard, I traced it down to the Power Management of Windows 7.

Windows 7 defaults to spin down drives after 10 minutes. So I would encounter strange lockups and long pauses when using the machine after idle. Once I changed the power scheme to never spin down the hard drives the problem went away.
 
compbl said:
Windows 7 defaults to spin down drives after 10 minutes. So I would encounter strange lockups and long pauses when using the machine after idle. Once I changed the power scheme to never spin down the hard drives the problem went away.
Click to expand...
Thanks for your advice, but it's already done : http://hardforum.com/showpost.php?p=1036810446&postcount=31

We are investigating around C3 / C6 / Pll Overvoltage settings to find which of these parameters is the faulty one ! DrCoolZic will do some test around it (I assume when he will have the time).

Like I have shown in one of my precedent post, the C3 state is used for the TBoost technology, so we hope it will not be the faulty one.

PLL Overvoltage must be disabled for security purpose (look at the Gary and DrCoolZic precedent posts in this thread).

The C6 State (deep-deep-sleep) is for me the real guilty, but, we must wait some tests from peoples who have experienced the problem and who have corrected it by disabling C3/C6 settings.
 
Just for info I discovered that my C6 was enabled and C3 disabled !!!
Dont know why, but I have tweaked the BIOS so often it must be a left over.
However I do not know for how long I had this setting but I did not experience any hang !

I now have C3 enabled and C6 disabled. I will see if system is stable.

As far as the HD spin down: on one of my windows build (I have 4 bootable disks with different experiments going on) I forgot to turn it off and apart from waiting for the disk to spin up I did not experience any hang.

I also found a nice page on FAQ for the Asus board http://support.asus.com/faq/faq.aspx?SLanguage=en-us&model=P8P67 DELUXE&product=6&os=30 that also talks about the S3 problem. You may want to have a look

Really not related but if you are ready to experiment with EFI GPT boot disk I wrote something http://hardforum.com/showthread.php?t=1584795
 
Last edited:
For people who have the C3/C6 stability issue, did you make a try with the new bios and the C3/C6 enabled (Auto=Disabled, so you must put it to enabled manually...) ? ^_^

@DrCoolZic: Thanks for your report, so it was maybe the C3 Report which was the faulty one ?

My first test of the new bios is really really strange ! The TBoost seems to work completely differently with the 1302 bios! Even with the C3/C6 disabled and with 1/2/3/4 Thread(s), the TBoost always increase all the 4 cores to the max (3700Mhz in my case) ! I have confirmed my test with TMonitor v1031 (which could show the speed of each core). Wow !?
 
Sundae said:
For people who have the C3/C6 stability issue, did you make a try with the new bios and the C3/C6 enabled (Auto=Disabled, so you must put it to enabled manually...) ? ^_^

@DrCoolZic: Thanks for your report, so it was maybe the C3 Report which was the faulty one ?

My first test of the new bios is really really strange ! The TBoost seems to work completely differently with the 1302 bios! Even with the C3/C6 disabled and with 1/2/3/4 Thread(s), the TBoost always increase all the 4 cores to the max (3700Mhz in my case) ! I have confirmed my test with TMonitor v1031 (which could show the speed of each core). Wow !?
Click to expand...
I now have C3 enabled and it is working fine.
I am a little bit reluctant to update to BIOS 1302 Beta.

I now have a system working perfectly with the following driver & utilities
Note that I am using the latest MEI (1197) from Intel. From what I have heard a new RST is coming soon.
****************************************
* Asus P8P67 Deluxe Divers & Utilities *
****************************************

Drivers
-------
From Designation Version Date File Name

Intel Serie 6 Chipset Device V9.2.0.1021 PV 01/04/2011 infinst_autol.exe
Intel Rapid Storage Technology V10.1.0.1008 12/15/2011 iata_cd.exe
Intel Management Engine Driver V7.0.4.1197 02/09/2011 MEI_allOS_7.0.4.1197_PV.exe
Intel Network Adapter Driver V16.0 01/27/2011 ProWin32.exe/ProWin64.exe
ASUS Audio Realtek V5.10.0.6235-V6.0.1.6235 12/29/2010 Realtek_Audio_V5106235_WinXp_V6016235_VistaWin7
ASUS JMicron JMB36X V1.17.58.2 12/30/2010 JMB36X_V117582_XpVistaWin7
ASUS Marvell 9128 & AHCI Drivers V1.0.0.1051 12/30/2010 Marvell_9128_V1001051_XpVistaWin7
ASUS LAN Realtek 8811E V5.770.909-V7.31.1025 12/29/2010 Realtek8111E_LAN_V5770_6243_731_XPVistaWin7
ASUS USB NEC USB 3.0 V2.0.4.0 12/29/2010 NEC_USB_3_V2040_WindowsXP_Vista_7
ASUS Bluetooth Atheros Drivers V7.2.0.40 12/30/2010 AW_NB051_V64519_XP_V72039_Vista_V72040_Win7
ASUS Bluetooth Atheros WLAN V9.0 12/30/2010 AW_NB037_V90_XpVistaWin7

Not Instlled:
Browser Configuration Utility 1.0.10.0

Asus Utilities -- AI Suite II V1.0.14
-------------------------------------
Asus Mobilink 1.00.37
Asus Update 1.01.02
BT GO! 1.00.33
DIGI+ VRM 1.00.17
EPU 1.02.08

Not Intalled
AI Charger AI Charger V1.00.09
AI Marvell MSU Utility V4.1.0.1909


Intel Drivers Download Links
http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=19596&ProdId=3301&lang=eng
http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=19607&keyword=RST&lang=eng
http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=19691&lang=eng
http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=18713&lang=eng
Click to expand...

I would appreciate if people can write about their experience with the new BIOS.

I do not understand when Gary says
Gary said:
I am releasing the beta 1302 for Pro, Deluxe, EVO and Standard boards. There will not be any official support for these releases except by ASUS Technical Marketing members in this forum (Gary, JJ, Mason), so do not call Technical or Customer support, please PM one of us. In addition, this is a beta with new Intel microcode so there will not be a possibility of rolling back
Click to expand...
New Intel microcode? Is this in the BIOS? or new drivers?
Not possible to rollback? The BIOS ? the drivers?

This sounds a little bit scary :(
 
DrCoolZic said:
I now have C3 enabled and it is working fine.
I am a little bit reluctant to update to BIOS 1302 Beta.

I now have a system working perfectly with the following driver & utilities
Note that I am using the latest MEI (1197) from Intel. From what I have heard a new RST is coming soon.

I would appreciate if people can write about their experience with the new BIOS.

I do not understand when Gary says

New Intel microcode? Is this in the BIOS? or new drivers?
Not possible to rollback? The BIOS ? the drivers?

This sounds a little bit scary :(
Click to expand...

The new Intel microcode is in the Bios, (the firmwares update for Marvell and JMicron also). The 1302 Beta Bios prevents to downgrading to a previous Bios version.

I have updated to 1302 two days ago, the USB performances are better (more a feeling than a real bench, just my observation of the speed in a Win7 copy window). And I confirmed the TBoost v2 which is working completely different : TBoost max on all Cores and not anymore 100Mhz per core used, and it doesn't care about the C3 state. At the moment, I didn't see any functionnalities degraded in my case.
 
DrCoolZic said:
I now have C3 enabled and it is working fine.
:(
Click to expand...


Thanks for the update and hard work, this motherboard is on my watch list. Have you had any issues with the ethernet? For me the only reason to spend the extra would be for the intel ethernet chip vs the one used on the standard P8P67
 
I seem to be having a similar issue with my new build, as well. I just got this system up and running last weekend. The BIOS was been flashed to 1204 and, aside from disabling a couple of onboard devices, I'm using all default values in the BIOS. Since I just built the system, my priority is on stability so I'm not overclocking yet. I've been load testing to ensure that my temps are good and stable. I also ran memtest for ~ 36 hours with 0 errors. Under load, everything seems fine. Unfortunately, the system has locked up twice while just sitting idle. It was locked hard and the only thing I could do power cycle the system. Event viewer shows nothing other than the fact that the system didn't shut down cleanly. :(

System specs are:

Asus P8P67 Deluxe
Intel i7 2600K w/Corsair H50
Corsair Vengeance 8GB DDR3 1600
2x EVGA GeForce GTX 580 SC
Crucial RealSSD 300 (SATA 6G)
Western Digital Caviar Black 640GB 7200 (SATA 6G)
LITE-ON iHAS524
Antec HCP-1200
Corsair Obsidian Series 800D
Windows 7 x64
 
glen said:
Thanks for the update and hard work, this motherboard is on my watch list. Have you had any issues with the ethernet? For me the only reason to spend the extra would be for the intel ethernet chip vs the one used on the standard P8P67
Click to expand...
The Intel Ethernet has been working fine for me at 1 GB :)
Only problem was originally connection was detected at 100 MB and I had to specify manually 1GB but now everything works great.
 
bmt22033 said:
I seem to be having a similar issue with my new build, as well. I just got this system up and running last weekend. The BIOS was been flashed to 1204 and, aside from disabling a couple of onboard devices, I'm using all default values in the BIOS. Since I just built the system, my priority is on stability so I'm not overclocking yet. I've been load testing to ensure that my temps are good and stable. I also ran memtest for ~ 36 hours with 0 errors. Under load, everything seems fine. Unfortunately, the system has locked up twice while just sitting idle. It was locked hard and the only thing I could do power cycle the system. Event viewer shows nothing other than the fact that the system didn't shut down cleanly. :(

System specs are:

Asus P8P67 Deluxe
Intel i7 2600K w/Corsair H50
Corsair Vengeance 8GB DDR3 1600
2x EVGA GeForce GTX 580 SC
Crucial RealSSD 300 (SATA 6G)
Western Digital Caviar Black 640GB 7200 (SATA 6G)
LITE-ON iHAS524
Antec HCP-1200
Corsair Obsidian Series 800D
Windows 7 x64
Click to expand...
This was the starting point of this thread!
I thought originally that C3/C6 and PLL overvoltage were responsible for system hanging but this does not seems to be the case as now it works with C3/C6 enabled.

The symptom was exactly what you describe working fine when system actively working but would hang if left idle.

I cant tell for sure what fixed the problem as this does not happen very often.
So my advice is the following:
  • Disable PLL overvoltage (unless doing overclocking)
  • Make sure you have the latest drivers as shown in my previous post. I believe this is the most important point specially the latest Intel MEI and chipset. This really helped. Do not take the Intel drivers from Asus they are lagging by one or two versions.
  • I have also disabled in Windows the sleep and hibernate mode, but I turn off the monitors after 10mn (originally also suspected this as a problem but works fine).
  • Stick to 1204 till system stable.

There is also another subject related to hanging that is not often mentioned: the Windows 7 activators :)
Of course nobody use them ;) but I have seen a post from Gary (Asus marketing) saying that this could be a problem. So I have tried several of them and indeed some of them breaks the communication between the OS and the HW. It is a little bit complex to explain but usually they install SLIC table in memory so the OS believe it is in the BIOS. The problem is that UEFI is a somewhat different beast than the traditional BIOS and the ACPI communication channel can get confused if tables are not placed properly... My explanation is probably not accurate but it gives an idea of the problem.
One easy way to find out if the OS-HW communication is not working properly is by starting the Fan Xpert in AI Suite. Normally the fan should be in standard mode (or what ever you have set in the BIOS). If the communication is broken the fan mode would be in disabled mode and switching between CPU and chassis reset all your fan settings... This has been reported as a bug in AI suite but in fact it is not a problem with AI suite :D
If you are in this case:eek: you have a very high probability of system hanging as communication between OS and HW is not handled properly.
 
So basically (some) people are using pirated software and then coming on forums for help.

LAME.

Either BUY windows like the rest of us, get a copy from a friend who has MSDN, or friggin flash an SLIC 2.1 enabled BIOS Which should not be hard since Asus is a legal OEM. But then again who knows...unlike traditional bios, there aren't any bios tools for modders with UEFI....

If no SLIC 2.1 bios exists, well too bad. People should have been grateful they existed for previous boards. Never had to bother with a loader (I actually used and flashed a SLIC 2.1 modded bios on a p5w simply because I was bored, even though I have a normal W7 retail key).
 
Hi All

P8P67- M pro bios 601

Was having idle lockups virtually every session everything in bios was at stock.

My memory is rated 1600 XMP and was set at the default 1333, so I set it to the XMP profile at 1600 and I've had no lockups since, been perfectly stable for 3 days.

Still have sleep and hibernation off in windows

Hope it helps
 
Hi DrCoolZic. Thanks for the suggestions! You've given me some things to check tonight when I get home.

DrCoolZic said:
  • Disable PLL overvoltage (unless doing overclocking)
  • Make sure you have the latest drivers as shown in my previous post. I believe this is the most important point specially the latest Intel MEI and chipset. This really helped. Do not take the Intel drivers from Asus they are lagging by one or two versions.
  • I have also disabled in Windows the sleep and hibernate mode, but I turn off the monitors after 10mn (originally also suspected this as a problem but works fine).
  • Stick to 1204 till system stable.
Click to expand...

I have PLL overvolt disabled and I'm using the same drivers as you listed in a previous post. I did NOT yet disable sleep/hibernate in Windows so I'll try that. And I'll definitely stick with 1204 for a bit.

DrCoolZic said:
There is also another subject related to hanging that is not often mentioned: the Windows 7 activators
Of course nobody use them but I have seen a post from Gary (Asus marketing) saying that this could be a problem. So I have tried several of them and indeed some of them breaks the communication between the OS and the HW. It is a little bit complex to explain but usually they install SLIC table in memory so the OS believe it is in the BIOS. The problem is that UEFI is a somewhat different beast than the traditional BIOS and the ACPI communication channel can get confused if tables are not placed properly... My explanation is probably not accurate but it gives an idea of the problem.
One easy way to find out if the OS-HW communication is not working properly is by starting the Fan Xpert in AI Suite. Normally the fan should be in standard mode (or what ever you have set in the BIOS). If the communication is broken the fan mode would be in disabled mode and switching between CPU and chassis reset all your fan settings... This has been reported as a bug in AI suite but in fact it is not a problem with AI suite
If you are in this case you have a very high probability of system hanging as communication between OS and HW is not handled properly.
Click to expand...

My OS load is Windows 7 x64 RTM from MSDN. It has not, however, been activated yet because I was waiting on the release of SP1 to MSDN yesterday. I was using RTM so I could get started with stability testing but didn't want to waste one of my MSDN keys for something that I knew I'd be replacing in a week or so. Now that I have the new SP1 slipstreamed ISO downloaded, I'll be wiping the system and reloading very soon. I'm guessing that the "activators" you're referring to are the patches intended to circumvent Windows product activation? If so, that's definitely not the issue for me (but thanks for the info nonetheless). If I'm misinterpreting your statement, though, and you think this issue could be related to the OS having not being activated yet then please let me know. Either way, I'll be reloading and activating Win7 SP1 in the next day or two and hopefully that, combined with some of your other suggestions will resolve this for me. Thanks again for your help!
 
bmt22033 said:
If I'm misinterpreting your statement, though, and you think this issue could be related to the OS having not being activated yet then please let me know. Either way, I'll be reloading and activating Win7 SP1 in the next day or two and hopefully that, combined with some of your other suggestions will resolve this for me. Thanks again for your help!
Click to expand...
No you have interpreted correctly. Hang problem is not related to activation.
One last thing to look at is memory. I never had any problem running my memory at 1600 but looks like problems have been reported in this area.
ONe thing that seems to help a lot is to perform a clean CMOS reset.
Procedure is turn off power supply wait until capacitors are discharged (all led on the board turned off) and press the clean cmos button for 15/20 sec. Restart system and set parameters. According to Gary this solve 95% (!) of problems related to memory?

Side note:
I was using RTM so I could get started with stability testing but didn't want to waste one of my MSDN keys for something that I knew I'd be replacing in a week or so.
Click to expand...
Are you sure you need to use another MSDN key when applying SP1? Am about to apply SP1 so I like to now.
 
Last edited:
You must log in or register to reply here.
Back
Top