Intel Haswell Overclocking Clubhouse.

[erixx]

i found a litle sweet spot at 4600, with idle low volts, etc, and pass all tests and games...
I also could run benches at 48x and fixed 1,32 v, but I want it to clock and volt down when idle.
I am still green with this processor and all the VDIO and vAIO will have to mature...


making some screenshots for you of a solid 46x OC:

 

[TheHunter]

Try this,




EDIT: ah I thought it was you fullinfusion posting these bios screens , but still you erixx can tweak like that too

Is it ok *safe* without degration to run the CPU Boot strap this high?

None, cpu is designed to run this way ..Mine doesnt want to run 125 or 169mhz for some reason, or i set it wrong.. Always fails by ram or gpu check.. <_<

Well are you sure you dont have those ohm settings, they only appear if you select main source clock, kinda like you did now when you selected 169mhz strap

ok, well. Setting both analogue I/O voltage and Digital I/O voltage to 1,25v did not work with the video benchmark..

Sidenote: if I set Adaptive voltages vor Core and Alternative voltages, I have no option to set the above values. I have to go to full manual mode. Will try to leave some on auto, but set vDIO, AIO to 1,25v....

Nah you can keep vIAO alone, just vDIO but still vDIO is not the cause imo, VCCSA would be sooner or cpu current % or even Cache multi if OC'ing.

With Asus M7G Uefi bios 1104 I am able to run a lot of things with 4790K@4800 Mhz (voltages tested at auto, 1,25, 1,29 and 1,30), but not to finish many tests. h-264 encode bench version 5.0.1 resets my pc just before finishing.
I have been playing around with adaptive voltage, but no change so far.

Apart, what do you guys think of the build in "Gamers Profile", that is, different speeds per core: 47, 46, 45, 44 Mhz.... ? Personally I am looking for a 24/7 solid overclock, that runs stable, cool enough, and makes sense in terms of real world performance.

If it resets, did you try higher cpu current %? Its at auto by default in DIGI+, set to at least 120 - 130% and try again.

Got a blue screen for the first time in like 4+ weeks today. Haven't changed anything. Haven't done much to investigate it yet, but thought id post it here.

0x00000019 (0x0000000000000003, 0xfffff901406c4010, 0xfffff903406c4010, 0xfffff901406c4010) Bad_Pool_Header

http://aumha.org/a/stop.php#0x19

something with routers, lan drivers, even ram OC issue..

STOP 0x00000019: BAD_POOL_HEADER (go to top of page)
Usual causes: Device driver

MSDN Listing (Win2K ResKit): http://msdn.microsoft.com/en-us/library/ms818860.aspx
MSDN Listing (WDK): http://msdn.microsoft.com/en-us/library/ff557389(v=VS.85).aspx
AUMHA Listing: http://aumha.org/a/stop.php#0x19

Problem w/ESET and SandBoxie (thanks to RedScare!): http://windows7forums.com/blue-screen-death-bsod/84555-bsods-shortly-after-startup.html#post272422

Knowledge Base Articles:
KB 156731 Windows NT Err. Msg: Stop 0x00000019 in NTFS
KB 179129 STOP 0x0000000A or 0x00000019 Due to Modified Teardrop Attack (NT)
KB 295624 "Stop 0x00000050" or "Stop 0x00000019" While Printing Extended Characters to a PostScript Printer (NT)
KB 883516 You receive Stop error code 0x00000019 or Stop error code 0x00000050 in the Tcpip.sys file on a Windows 2000 Server-based computer
KB 884585 When you back up to EMC Clarion storage devices in a SAN environment, your Windows Server 2003-based computer may stop responding after you restart
KB 892260 You may receive a "STOP: 0x00000019" error message on a Windows Server 2003-based computer
KB 905795 When you try to control a Systems Management Server 2003 client from a remote location, you experience a Stop error on the Systems Management Server 2003 client
KB 925259 Error message when a Delayed Write Failure event is reported in Windows Server 2003: "Stop 0x00000019 - BAD_POOL_HEADER" or "Stop 0xCD PAGE_FAULT_BEYOND_END_OF_ALLOCATION"
KB 931479 The print server experiences a Stop error when you use the Point and Print feature to print to a shared printer in Windows Vista (XP, Server 2003)
KB 934177 FIX: Error message when you connect a scanner or a digital camera to a computer that is running Windows Vista or Windows XP Service Pack 2: "BAD_POOL_HEADER STOP"
KB 934326 FIX: Stop error message when you run the Client for NFS service in Microsoft Windows Services for UNIX 3.5: "Stop 0x00000019"
KB 935920 When the Emulex Elxsli2.sys driver is installed, you may receive a Stop error message after you upgrade your computer to Windows Server 2003 Service Pack 1
KB 946226 FIX: You receive a "Stop 0x00000019" error message or a "Stop 0x000000c4" error message when you access NFS resources by using User Name Mapping (MS Svcs for Unix)
KB 947979 FIX: Stop error message on a computer that has the Gateway for NFS component of Windows Services for UNIX 3.5 installed: "Stop 0x00000019"
KB 948289 Error message on a Windows Server 2003-based computer: "Stop error code 0x00000019"
KB 949531 Stop error message when you try to install Windows Server 2008 on a computer that is connected to an iSCSI target LUN by using an iSCSI bootable network adapter: "STOP 0x00000019"
KB 955015 Stop error message on a Windows Server 2008-based computer when the NFS feature is enabled: "Stop: 0x00000019"
KB 970693 Stop error message in Windows Server 2008 or Windows Vista after you uninstall or reinstall a wireless network adapter driver: "0x00000019 BAD_POOL_HEADER"
KB 971280 Random Stop errors on a Windows Server 2003-based computer that performs input/output to file shares hosted on another computer or to a network-attached storage device
KB 973026 Error message on a computer that is running Windows Vista or on Windows Server 2008 that has the IPMI driver installed: "Stop 0x00000019"
KB 973097 Stop error message when the Smb.sys driver is unloaded on a computer that is running Windows Server 2003: "Stop 0x00000019"
KB 974201 Stop error message when you retrieve WMI connection statistics for iSCSI after you change the iSCSI configurations on a computer that is running Windows Server 2008 or Windows Vista: "0x00000019 BAD_POOL_HEADER"
KB 976443 Stop error message when you retrieve WMI connection statistics for iSCSI after you change the iSCSI configurations on a computer that is running Windows Server 2008 R2 or Windows 7: "0x00000019 BAD_POOL_HEADER"
KB 2280732 You receive Stop error message 0x000000CD or 0x00000019 on a computer that is running Windows Server 2003 when corrupted registry hive files are loaded
KB 2669443 "0x00000019" Stop error when you use Internet Explorer 8 to browse webpages in Windows Vista or in Windows Server 2008
KB 2712746 "0x00000019" Stop error when you run an SNMP application on a computer that is running Windows Vista or Windows Server 2008
KB 2832829 "0x00000019" Stop error when you try to shut down a Windows 7 or Windows Server 2008 R2-based computer

WinDbg Help File Entry:
The BAD_POOL_HEADER bug check has a value of 0x00000019. This indicates that a pool header is corrupt.

Parameters
The following parameters are displayed on the blue screen. Parameter 1 indicates the type of violation. The meaning of the other parameters depends on the value of Parameter 1.

Parameter 1 Parameter 2 Parameter 3 Parameter 4 Cause of Error
0x2 The pool entry being checked The size of the pool block 0 The special pool pattern check failed.

(The owner has likely corrupted the pool block.)
0x3 The pool entry being checked The read-back flink freelist value The read-back blink freelist value The pool freelist is corrupt.

(In a healthy list, the values of Parameters 2, 3, and 4 should be identical.)
0x5 One of the pool entries Reserved The other pool entry A pair of adjacent pool entries have headers that contradict each other. At least one of them is corrupt.
0x6 One incorrectly-calculated entry Reserved The bad entry that caused the miscalculation The pool block header's previous size is too large.
0x7 0 Reserved The bad pool entry The pool block header size is corrupt.
0x8 0 Reserved The bad pool entry The pool block header size is zero.
0x9 One incorrectly-calculated entry Reserved The bad entry that caused the miscalculation The pool block header size is corrupted (it is too large).
0xA The pool entry that should have been found Reserved The virtual address of the page that should have contained the pool entry The pool block header size is corrupt.
0x20 The pool entry that should have been found The next pool entry Reserved The pool block header size is corrupt.

Cause
The pool is already corrupted at the time of the current request.

This may or may not be due to the caller.

Resolving the Problem
The internal pool links must be walked to figure out a possible cause of the problem.

Then you can use special pool for the suspect pool tags, or use Driver Verifier on the suspect driver. The !analyze extension may be of help in pinpointing the suspect driver, but this is frequently not the case with pool corrupters.

WinDbg Output Example:
BAD_POOL_HEADER (19)
The pool is already corrupt at the time of the current request.
This may or may not be due to the caller.
The internal pool links must be walked to figure out a possible cause of
the problem, and then special pool applied to the suspect tags or the driver
verifier to a suspect driver.
Arguments:
Arg1: 00000020, a pool block header size is corrupt.
Arg2: 89d147d8, The pool entry we were looking for within the page.
Arg3: 89d14808, The next pool entry.
Arg4: 08060007, (reserved)

^
from http://www.carrona.org/bsodindx.html

 

[erixx]

HAPPY Been playing and testing hours, the whole evening, at 4800 Mhz, only remains another run of that movie bench of yours.
Aida stress fine, Grid 2 and Insurgency gaming fine (prone to crash with bad OC's). Intel burn, fine.
Temps up to 82ºC while stressing.
1,30 V "OC Voltage" + 0,05 Offset (Adaptive). I left on "auto", Mr. Asus increases it to 1,4 V which is fine for pure results, but a bit hot... lol
CPU Cache voltage, adaptive, but with 0 offset.
All others auto, apart from power 130% (thanks mate! ) and and Phase control on "optimized" (on Auto or Extreme, idling on low votlage does not work)

RAM, I have set it to 1T (stock is 2T) and increased it to 1.66 V. Results in Aida thread.

 

[fullinfusion]

EDIT: ah I thought it was you fullinfusion posting these bios screens , but still you erixx can tweak like that too



None, cpu is designed to run this way ..Mine doesnt want to run 125 or 169mhz for some reason, or i set it wrong.. Always fails by ram or gpu check.. <_<

Well are you sure you dont have those ohm settings, they only appear if you select main source clock, kinda like you did now when you selected 169mhz strap

Nope there no ohm setting below the strap.
If I change the strap from 100, 167, 250 the bus changes from 100 to 99.8 but alternates with what setting I chose. I can run any strap except the 250 one lol... The memory starts at 3000and goes over 5000+ so I didn't even bother. The 167 strap was easy. I found you need to lower the memory or set it as close to its rated speed and loosen the timings a tad. Also the cpu cache multi needs to be dropped to because its just like the cpu multi. It almost caught me off guard but Dave said to watch for it because it will change and will need to be lowered to the 4ghz mark. I sure find the system snappy as all hell running the bus at 167 vs the 100mhz setting. The bios on restart did its thing when I clicked save and boot.. It took 3 power up and down modes I guess setting the frequency's then boot up was painless.

 

[erixx]

HAPPY Been playing and testing hours, the whole evening, at 4800 Mhz, only remains another run of that movie bench of yours.
Aida stress fine, Grid 2 and Insurgency gaming fine (prone to crash with bad OC's). Intel burn, fine.
Temps up to 82ºC while stressing.
1,30 V "OC Voltage" + 0,05 Offset (Adaptive). I left on "auto", Mr. Asus increases it to 1,4 V which is fine for pure results, but a bit hot... lol
CPU Cache voltage, adaptive, but with 0 offset.
All others auto, apart from power 130% (thanks mate! ) and and Phase control on "optimized" (on Auto or Extreme, idling on low votlage does not work)

RAM, I have set it to 1T (stock is 2T) and increased it to 1.66 V. Results in Aida thread.

I quote myself to avoid confussion. Above settings are fine, BUT i did NOT PASS the x264_Benchmark_HD_v5.0.1. Not sure if I should just leave it this way, untill I find a more annoying problem...

 

[fullinfusion]

@erixx ditch that x-264 encoder. I seen ppl complaining that the latest version is buggy as all hell.

How about XTU stress or AIDA64 stress can it pass an hour or more?

 

[TheHunter]

Nope there no ohm setting below the strap.
If I change the strap from 100, 167, 250 the bus changes from 100 to 99.8 but alternates with what setting I chose. I can run any strap except the 250 one lol... The memory starts at 3000and goes over 5000+ so I didn't even bother. The 167 strap was easy. I found you need to lower the memory or set it as close to its rated speed and loosen the timings a tad. Also the cpu cache multi needs to be dropped to because its just like the cpu multi. It almost caught me off guard but Dave said to watch for it because it will change and will need to be lowered to the 4ghz mark. I sure find the system snappy as all hell running the bus at 167 vs the 100mhz setting. The bios on restart did its thing when I clicked save and boot.. It took 3 power up and down modes I guess setting the frequency's then boot up was painless.

Well you need to first set 100mhz in it then it will show up.. At least that's what it does here..

Cache it depends, mine starts to get wonky at 4.3ghz +, had at 4.4ghz but it was kinda all over the place.. Got lower benchmark scores instead of higher, guess i didnt feed enough volts and or it just doesnt want to go over 4.2ghz @ 4.7ghz+ cpu multi.


Btw i replied your mail

 

[TheHunter]

HAPPY Been playing and testing hours, the whole evening, at 4800 Mhz, only remains another run of that movie bench of yours.
Aida stress fine, Grid 2 and Insurgency gaming fine (prone to crash with bad OC's). Intel burn, fine.
Temps up to 82ºC while stressing.
1,30 V "OC Voltage" + 0,05 Offset (Adaptive). I left on "auto", Mr. Asus increases it to 1,4 V which is fine for pure results, but a bit hot... lol
CPU Cache voltage, adaptive, but with 0 offset.
All others auto, apart from power 130% (thanks mate! ) and and Phase control on "optimized" (on Auto or Extreme, idling on low votlage does not work)

RAM, I have set it to 1T (stock is 2T) and increased it to 1.66 V. Results in Aida thread.

Dont keep voltage @ auto with adaptive, its the same as if you used default auto. By total turbo voltage enter your stable fixed voltage, now it will boost to that max.


Same by cache, but then again if you used fixed cache it won't matter much, it will run at same voltage all the time.. If you set minimum cache multi to 35 and max ie 40x it will now lower accordingly.


Idk if you found your fixed cpu voltage @ x264 ie 1.30v use this voltage, but dont stress test anymore with it or it WILL overvolt for 0.09v, 1.30 will become 1.39v. Same by Aida64 stress test even at just ram testing - no avx involved.

x264fhd is ok for stress testing, dunno about x264 5.05? benchmark.
http://downloads.guru3d.com/x264-FHD-Benchmark-v1.0.1-64bit-download-2825.html


as for 1T you can keep ram voltage at same 1.65v.

 

[fullinfusion]

Why adaptive, why not offset?

 

[TheHunter]

Because its better and no need to fiddle until you find the right offset voltage. Also offset will raise its default base multi voltage and you dont really want that.. Unless you really need to but you can always use a combo of both.

For example

cpu offset 0.010v
additional turbo voltage 1.25v

total adaptive turbo voltage 1.260v (0.010 + 1.250v)

 

[fullinfusion]

im just using offset at 0.005
and under full demand the vcore hits the same voltage I had keyed in when in manual mode.

Edit.....

 

[erixx]

just to update you, thank you a lot! This is not over, because, without apparent reason, this evening nothing works! Not 48x, not 47x... Just gone back to auto-ALL to enter here... Maybe it is the heatwave we have here in the Iberian Peninsula, constantly high temps, and AirCo demanding also un upgrade to 64bit! LOL
We will keep this going...

 

[fullinfusion]

just to update you, thank you a lot! This is not over, because, without apparent reason, this evening nothing works! Not 48x, not 47x... Just gone back to auto-ALL to enter here... Maybe it is the heatwave we have here in the Iberian Peninsula, constantly high temps, and AirCo demanding also un upgrade to 64bit! LOL
We will keep this going...

Reflash the bios! Asus bioses flake out generally after 8-10 bsod's

If you have saved a few overclock profiles save them to a jump drive then load optimized settings, enter to restart and go back into the bios and flash the bios chip.

Be just like new again

 

[Skywalker12345]

Hello everyone its been a while since i have been able to get back into OCing! Finally got around to ordering the following:

ASUS Z97I-PLUS LGA 1150 Mini ITX & i5-4460

Please point me in the right direction for a newb Haswell OCer, ive had experience with AMD in the past but no Intel CPUs.

Also i have a stock CPU cooler ATM (i know this is not sufficient i am planning on upgrading ASAP)

Thanks!

 

[fullinfusion]

Well got my low volt c-states working and stable as can be

@TheHunter thanks for all the added settings I need to change. You were right, some settings were just set to high for no reason

 

[Kursah]

@lucasweir

Well, you might be able to BCLK that chip, but only chips with the K at the end of their model are allowed to be overclocked with unlocked multipliers. And if you do go and buy a K-series chip I also recommend spending the extra $25-ish on the Intel Performance Tuning plan, because your warranty is void as soon as you OC beyond turbo speeds without it.

 

[fullinfusion]

This cpu strap at least gives me some freedom like my ol AMD rigs for bus speed enhancements..

This is nice! I've been playing with this setting all night and not a single crash.

Ill post the bios screen shots for others if you guys want to give it a go.

One note I must add is it sure beats the 100mhz bus intel has on these chips..

Snappy as a mofo to say the least

 

[TheHunter]

Reflash the bios! Asus bioses flake out generally after 8-10 bsod's

If you have saved a few overclock profiles save them to a jump drive then load optimized settings, enter to restart and go back into the bios and flash the bios chip.

Be just like new again

This, but usually its enough to reset bios to optimized defaults, apply (reboot) and re-apply your OC.

 

[fullinfusion]

I did it, I FUCKING DID IT!!!!

I benched Cinibench @ 5GHz!!! and it passed!

It may not be a big deal to all, but to me, what an accomplishment!
Want bios shot's let me know

 

[fullinfusion]

Kinda funny, I reset the bios and decided to run stock with low power enabled.

Look at the voltage vs the core clocks.

I don't even need that voltage @ 4.7ghz and this is at stock!!!

@asus #asus wtf programmers?

 

[fullinfusion]

Asus Maximus Formula VI owners, a new bios is out.

I hope the multi is fixed this time

 

[fullinfusion]

Ok a bit of an update on this bios.

From what I've seen so far there's lots of added features in the DIGI+ Power control section.
Theres a new memory timing control called Latency compensator
In CPU power management (advanced) New CFG Lock??? whatever that is??
And biggest thing I see is core voltage in windows is a lot tighter and not bouncing around like it used to.

Will add more when I find more new options

Edit look how tight the cpu core is now! LLC is at 6

Just at Idle

Under load

Sorry here's the cpu-z on this bios

 

[fullinfusion]

Have to share another DP

Low Power states work despite not being set in the control panel.

I have mine set to MAX Performance and all I must say is NICE!!

Idle is 4.0GHz and turbos up to 4.4ghz. Low volt is 0.320 @ 4.2ghz

This is stock setting, Load Optimized settings, nothing else changed..

Kudo's to you @asus

 

[cadaveca]

Nice to see they've finally getting a handle on it

 

[fullinfusion]

Nice to see they've finally getting a handle on it

Yes but my 4.8Ghz setting is gone with the wind over the 1505 bios, I think I have it figured out but until morning I cant be sure.

But then again I'm using the 166.6 cpu strap lol and had to lower the memory to 2332MHz and use your 2400MHz timings... 11-13-13-35-2T ext ext.........

So far so good, I never had all that much luck using the 125 strap but if you send me those Avexir 3000+MHz sticks I can try the 250 cpu strap... If I remember the lowest memory setting on it is 3000MHz and goes over 6k lol