Jonny, MonsterDK... I use Windows 7 RC 64bit and have had zero issues with ET6. When you double-click it wait 10 seconds, then check your systray as that's where it will be hiding. :)
Printable View
Jonny, MonsterDK... I use Windows 7 RC 64bit and have had zero issues with ET6. When you double-click it wait 10 seconds, then check your systray as that's where it will be hiding. :)
Nice one mate.....found it (feels a bit like a plank :rofl:)Quote:
Originally Posted by Kougar
Hi,Quote:
Originally Posted by Jonny Manc
So so,
Just trying to improve on a 1.35 VCore tried 1.325 and I got a reset. :(
Not sure what the UNCORE does but I've put that back to 3200 and I'm trying that with a 1.325 VCore to see what happens.
I was using a 3400 UNCORE with a 1.35 VCore with no problems priming after 2 hours and small 10k FFTs.
Hi
Is this memory Ok to run with the extreme?
OCZ Blade Series 6GB (3x2GB) DDR3 PC3-16000 Triple Channel (OCZ3B2000C9LV6GK)
As far as I know, UNCORE should be set to twice the speed of your RAM.....dont know what it actually does though :shrug:Quote:
Originally Posted by MonsterDK
Hi,Quote:
Originally Posted by Jonny Manc
Well priming with UNCORE 3200 and VCORE 1.325 seems stable up to now.
So for my system to reset at UNCORE 3400 with VCORE 1.325, whats this telling me? The more UNCORE the more VCORE needed?
Anyone know????
Yes I found the more UNCORE needed more vCore, just use 16x uncore if your memory is set to 8xQuote:
Originally Posted by MonsterDK
Also you can try QPI/VTT 1.355v
Use LinX to test stability, it will fail fast if your overclock is not stable, 25 passes with max mem available better than waiting hours and hours priming
Quote:
Originally Posted by Jonny Manc
Quote:
Originally Posted by phsinc1
I'm using QPI/VTT 1.335 at the moment is lower not better with this???
my corsair XMP set the QPI/VTT to 1.355v, it is safe to useQuote:
Originally Posted by MonsterDK
ok,, but what is better here... lower voltage here?Quote:
Originally Posted by phsinc1
Just wondered if there is a reason I should change the QPI higher if it seems lower?
sure lower is better, I'm just giving a tip if you still have instability problemsQuote:
Originally Posted by MonsterDK
I'm using this same ram on my UD5 and seems to working fine.Quote:
Originally Posted by Ikari1
So before I go reading all 100+ pages, is there a "sticky" as to how to OC this thing =D, yes I'm another newbie to i7 and x58
Can someone help me on my OC im using bios F7c and i got it up to 3800 i have a C0 920, i would appreciate some help thank you.
Quote:
Originally Posted by yang88she
Try this site: http://forums.tweaktown.com/f69/ga-e...locking-31636/Quote:
Originally Posted by Godmyster
Lsdmeasap does a good job explaining what to disable/enable and set when overclocking.
Still, ET6 Resets Turbo Multiplier. Makes this Program totally useless Im afraid. To bad, couse the program got it all that you need.
Use uncleweb turbo program to increase the multi back.:) But it still annoying to tick and untick the box everytime changing setting using ET6.:mad:Quote:
Originally Posted by Infa
Well I hope they working on it because the last time ET6 being updated is on April.:down:
When using Linx to test the system for stability.
What configuration within Linx will be best to ensure its stability?
At the moment I've selected Problem size: 23118 and Memory size: 4096 MiB
EDIT:
I get LinX 64 .exe stopped working... no system reboot though.
Will linX cause the system to fall over fully if the OC is unstable?
I usually try to use the max memory available, more problem size means more stress, try to disable all the autorun programs to free as much memory you can.Quote:
Originally Posted by MonsterDK
Any error is a sign of instability such as windows errors, BSOD or system reset.
Thank You!Quote:
Originally Posted by SimpleTECH
ok so LinX gives a windows error in Windows 7 after about 2 runs.Quote:
Originally Posted by phsinc1
Any suggestions please on what I should try ?
a Tip: Why you don't try 191x20 [with Turbo enabled] so you get 21x multi [191x21= 4GHz], I'm using this settings [191x21] and it is much easy to get it stable than [200x20]
I built the MultiBoost program into the latest version of RealTemp. Just add this to your RealTemp.INI file to keep your multiplier at its max:Quote:
Originally Posted by Sparda
MultiBoost=1
http://www.fileden.com/files/2008/3/...mpBeta_330.zip
It works good for the DFI boards with 12X multi issues.
Thanks for the tips.:up: Will try them.Quote:
Originally Posted by unclewebb
Oh..I miss that. Thanks.Quote:
Originally Posted by stasio
Quote:
Originally Posted by Godmyster
Quote:
Originally Posted by MonsterDK
Use this thing so people have some idea of your setting and rig.Quote:
Originally Posted by yang88she
PHP Code:Advanced CPU Features:
CPU Clock Ratio ................................ [ x]
Intel(R) Turbo Boost Tech ...................... [Enabled]
CPU Cores Enabled .............................. [All]
CPU Multi Threading .............................[Enabled]
CPU Enhanced Halt (C1E) ........................ [DISABLED]
C3/C6/C7 State Support ......................... [DISABLED]
CPU Thermal Monitor ............................ [DISABLED]
CPU EIST Function .............................. [Enabled]
Virtualization Technology ...................... [DISABLED]
Bi-Directional PROCHOT ......................... [DISABLED]
Uncore & QPI Features:
QPI Link Speed .............................. AUTO
Uncore Frequency ............................( x)
Isonchronous Frequency ......................[Enabled]
Standard Clock Control:
Base Clock (BCLK) Control .......................... [Enabled]
BCLK Frequency (MHz) ................................[ ]
PCI Express Frequency (MHz) ........................ [Auto]
C.I.A.2........................................... .. [Disabled]
Advanced Clock Control:
CPU Clock Drive ..............................[800mV]
PCI Express Clock Drive ................... [900mV]
CPU Clock Skew ............................. [0ps]
IOH Clock Skew ............................. [0ps]
Advanced DRAM Features:
Performance Enhance ...................... [TURBO]
Extreme Memory Profile (X.M.P) ........... [Disabled]
System Memory Multiplier (SPD) ........... [AUTO]
DRAM Timing Selectable (SPD) ............. [QUICK]
Channel A + B + C
Channel A Timing Settings:
##Channel A Standard Timing Control##
CAS Latency Time ......................
tRCD ..................................
tRP ...................................
tRAS ..................................
CR.....................................
Advanced Voltage Control:
CPU
Load Line Calibration ................. [ENABLED]
CPU Vcore ..............................[ ]
QPI/VTT Voltage 1.150v .................[ ]
CPU PLL 1.800v .........................[Auto]
MCH/ICH
PCIE 1.500v ...........................AUTO
QPI PLL 1.100v ........................AUTO
IOH Core 1.100v ....................... 1.1v
ICH I/O 1.500v ........................AUTO
ICH Core 1.1v .........................AUTO
DRAM
DRAM Voltage 1.500v ..............[ ]
DRAM Termination 0.750v.......... [AUTO]
Ch-A Data VRef. 0.750v ...........[AUTO]
Ch-B Data VRef. 0.750v ...........[AUTO]
Ch-C Data VRef. 0.750v ...........[AUTO]
Ch-A Address VRef. 0.750v ........[AUTO]
Ch-B Address VRef. 0.750v ........[AUTO]
Ch-C Address VRef. 0.750v ........[AUTO]
But i need help on a UD5 not on a UD4P pleaseQuote:
Originally Posted by SimpleTECH
Here are settings.
Please give advice on any changes. Thanks!
Advanced CPU Features:
CPU Clock Ratio ................................ [ 20x]
Intel(R) Turbo Boost Tech ...................... [Enabled]
CPU Cores Enabled .............................. [All]
CPU Multi Threading .............................[Enabled]
CPU Enhanced Halt (C1E) ........................ [DISABLED]
C3/C6/C7 State Support ......................... [DISABLED]
CPU Thermal Monitor ............................ [DISABLED]
CPU EIST Function .............................. [Enabled]
Virtualization Technology ...................... [DISABLED]
Bi-Directional PROCHOT ......................... [DISABLED]
Uncore & QPI Features:
QPI Link Speed .............................. AUTO
Uncore Frequency ............................( x16)
Isonchronous Frequency ......................[Enabled]
Standard Clock Control:
Base Clock (BCLK) Control .......................... [Enabled]
BCLK Frequency (MHz) ................................[200 ]
PCI Express Frequency (MHz) ........................ [101]
C.I.A.2........................................... .. [Disabled]
Advanced Clock Control:
CPU Clock Drive ..............................[900mV]
PCI Express Clock Drive ................... [900mV]
CPU Clock Skew ............................. [0ps]
IOH Clock Skew ............................. [0ps]
Advanced DRAM Features:
Performance Enhance ...................... [STANDARD]
Extreme Memory Profile (X.M.P) ........... [Disabled] - DO NOT SEE THIS
System Memory Multiplier (SPD) ........... [AUTO]
DRAM Timing Selectable (SPD) ............. [MANUAL]
Channel A + B + C
Channel A Timing Settings:
##Channel A Standard Timing Control##
CAS Latency Time ......................8
tRCD ..................................8
tRP ...................................8
tRAS ..................................24
CR.....................................1
Advanced Voltage Control:
CPU Intel Core i7 920 D0 Stepping
Load Line Calibration ................. [ENABLED]
CPU Vcore ..............................[ 1.3250 ]
QPI/VTT Voltage 1.150v .................[ 1.335]
CPU PLL 1.800v .........................[Auto]
MCH/ICH
PCIE 1.500v ...........................AUTO
QPI PLL 1.100v ........................AUTO
IOH Core 1.100v ....................... AUTO
ICH I/O 1.500v ........................AUTO
ICH Core 1.1v .........................AUTO
DRAM
DRAM Voltage 1.500v ..............[ 1.640 ]
DRAM Termination 0.750v.......... [AUTO]
Ch-A Data VRef. 0.750v ...........[AUTO]
Ch-B Data VRef. 0.750v ...........[AUTO]
Ch-C Data VRef. 0.750v ...........[AUTO]
Ch-A Address VRef. 0.750v ........[AUTO]
Ch-B Address VRef. 0.750v ........[AUTO]
Ch-C Address VRef. 0.750v ........[AUTO]
Anyone have a system identical to mine?
Quote:
Originally Posted by MonsterDK
Those settings look fine for 200x20 except the vcore a little bit high for D0. Try playing with the vcore for the linx and up the vdimm a notch or two..:shrug: I use prime blend to find my vtt/qpi. Usually if not enough vtt/qpi or vdimm it will fail in large ftt in prime blend. Most of the time it depend on the uncore speed. If it still fail try play with the IOH Core or QPI PLL or CPU PLL if there any effect.:shrug:Quote:
Originally Posted by MonsterDK
edit: No wonder. I missed the x21 multi. Well hopefully the temperature isnt an issue.;)
try QPI/VTT Voltage 1.150v .................[ 1.355]Quote:
Originally Posted by MonsterDK
Also in the sub timmings,`make sure tRFC is set to "88"
http://www.imgx.org/pthumbs/small/17290/pifast.jpg
http://www.imgx.org/pthumbs/small/17291/sp1m.jpg
920 D0 3851A227, UD5 F8B, crucial ddr3 1333, true + 130 cfm fans
whats the bclk wall for this board on air?
How do you guys generate that below? Is that within EasyTune or someother program.
Would make it much easier for me to document my settings :)
Quote:
Originally Posted by MonsterDK
This is about as far as I've overclocked my Xeon. Trying to get stable over 200 BCLK is a PITA.
http://img41.imageshack.us/img41/5756/41371200v.jpg
I've looked on google, but can't seem to locate a download of Prime Blend. Is it known as something other or a feature of another program?Quote:
Originally Posted by Sparda
Heh, the program is Prime95 and Blend is one of the tests you can run on it :up:Quote:
Originally Posted by MonsterDK
I'm not sure if it would help everyone but since getting my 950 i've noticed that over a certain bclk (205~) i needed to increase my NB volts to maintain stability, i'm not sure if it's because of my video cards.Quote:
Originally Posted by SimpleTECH
Any info on how to get the F8x BIOS to install?
I am using 8b modded with new ahci rom from noangel's post in stasio's tweaktown thread which works great including S3 works perfectly. Max bclk is 222 with normal volts/settings and 229.x with pcie at 103 (max on my mobo without freezing).
I tried F8x modded one in link below, no change in OC, no change in max bclk still 222 normal and 229.x given same settings, only difference is S3 is completely broken on F8x, and my computer kept shutting down randomly, so went back to F8b.
But here is post with F8x
http://www.xtremesystems.org/forums/...postcount=2835
Ok, I've been testing AHCI vs IDE mode, and i cant find any reason to change? everest disk said the same numbers more or less identical, I've read around a little, and the benefit are mostly Hotswap function using eSATA...
been checking my bootup time, and While in AHCI it said 25.6s while in IDE 18.2s .. what's the benefit of the new AHCI updated room? i dont get the version number really, but i think it says 2006? have anyone modded the UD5 bios to new AHCI?
AHCI increases read speeds vs ide, usually highest percentage gain is small read speeds that OS typically uses, (large percentage gains .5 to 8Kb in atto). Write speeds can also be affected depending on particular drive/firmware, pic below is mine booted up in ide vs ahci, and why I use ahci mode.
Boot up time is same if using new ahci rom, which does not pause for ~7-8 seconds while searching for drives. If not using a modded bios with new ahci rom, then boot up time is annoyingly 7-8 seconds longer.
ahci vs ide on mine both run today.
http://img26.imageshack.us/img26/3104/ahcivsidej.jpg
now thats a diffrence, but for me it was identical.. and im not using ssd either... so perhaps for an old raptor 16mb 36gb it doesnt do jack shijjt :)
i dont care about the other disk, they dont have to be fast.. but i will use this and see.. noone modded ud5 with the new?
i'm still running the F8x on my UD5, with no S3 or shutdown problems
hoping to do the pci-e resistor mod sometime, maybe next week
I have not seen UD5 bios modded, but I think the next new beta bioses for UD5 and Ext will have updated ahci rom, hopefully coming soon...at least all new betas on other mobo coming out have new ahci rom.
Bit of a noob on the IDE / AHCI thing but the results look interesting.
Can I just change to AHCI in BIOS?? No OS re-install or anything required??
Usually you download the latest driver, ie ICH10.
Then at start of OS installation, ie before OS installed, you get the sceen to load any drivers first, then you load the appropriate driver.
Then you install the OS with Ahci enabled in bios.
If you just try to boot up with ahci enabled in bios without installing driver, you will get bsod.
You can install ahci drivers after the fact with windows vista 64bit, assume same with windows 7, but I have not done either...can search google and find out how...though I have never tried. But much easier to do beforehand.
On other hand when you install under ahci with driver, you can then boot up in either mode just by changing setting in bios.
if you don't feel like crossflashing with the extreme BIOS, unstrain modded the F8b BIOS for the UD5 with the latest RAID and AHCI roms here:Quote:
Originally Posted by Infa
http://forums.tweaktown.com/300786-post1025.html
to enable ahci or raid without reinstalling, there's a guide here:
http://forums.hexus.net/hexus-hardwa...ch9-ich9r.html
Here are settings.
Please give advice on any changes. Thanks!
Advanced CPU Features:
CPU Clock Ratio ................................ [ 20x]
Intel(R) Turbo Boost Tech ...................... [Enabled]
CPU Cores Enabled .............................. [All]
CPU Multi Threading .............................[Enabled]
CPU Enhanced Halt (C1E) ........................ [DISABLED]
C3/C6/C7 State Support ......................... [DISABLED]
CPU Thermal Monitor ............................ [DISABLED]
CPU EIST Function .............................. [Enabled]
Virtualization Technology ...................... [DISABLED]
Bi-Directional PROCHOT ......................... [DISABLED]
Uncore & QPI Features:
QPI Link Speed .............................. AUTO
Uncore Frequency ............................( x16 @ 3200)
Isonchronous Frequency ......................[Enabled]
Standard Clock Control:
Base Clock (BCLK) Control .......................... [Enabled]
BCLK Frequency (MHz) ................................[200 ]
PCI Express Frequency (MHz) ........................ [101]
C.I.A.2........................................... .. [Disabled]
Advanced Clock Control:
CPU Clock Drive ..............................[800mV]
PCI Express Clock Drive ................... [900mV]
CPU Clock Skew ............................. [600ps]
IOH Clock Skew ............................. [0ps]
Advanced DRAM Features:
Performance Enhance ...................... [STANDARD]
Extreme Memory Profile (X.M.P) ........... [Disabled] - DO NOT SEE THIS
System Memory Multiplier (SPD) ........... [AUTO]
DRAM Timing Selectable (SPD) ............. [MANUAL]
Channel A + B + C
Channel A Timing Settings:
##Channel A Standard Timing Control##
CAS Latency Time ......................8
tRCD ..................................8
tRP ...................................8
tRAS ..................................24
CR.....................................1
Advanced Voltage Control:
CPU Intel Core i7 920 D0 Stepping
Load Line Calibration ................. [ENABLED]
CPU Vcore ..............................[ 1.3500 ] (trying to get stable)
QPI/VTT Voltage 1.150v .................[ 1.355]
CPU PLL 1.800v .........................[Auto]
MCH/ICH
PCIE 1.500v ...........................AUTO
QPI PLL 1.100v ........................AUTO
IOH Core 1.100v ....................... 1.220V
ICH I/O 1.500v ........................AUTO
ICH Core 1.1v .........................AUTO
DRAM
DRAM Voltage 1.500v ..............[ 1.640 ]
DRAM Termination 0.750v.......... [AUTO]
Ch-A Data VRef. 0.750v ...........[AUTO]
Ch-B Data VRef. 0.750v ...........[AUTO]
Ch-C Data VRef. 0.750v ...........[AUTO]
Ch-A Address VRef. 0.750v ........[AUTO]
Ch-B Address VRef. 0.750v ........[AUTO]
Ch-C Address VRef. 0.750v ........[AUTO]
In CPU-Z is shows 21 x 200 @ 4200 but Turbo is off under advanced Clock Control , its set to Standard.
Despite some recommendations what to try. I can't get it stable :(
I've tried various things and as soon as the rig deals with 10k FFT's it bumps out with a BSOD or the rig resets.
Temps are good... On water with Quad rad... temps are around 64c max.
There are 3 lotteries you play when you buy a chip.Quote:
Originally Posted by MonsterDK
vcore lottery, qpi/vtt lottery, and TDP (watts/temps) lottery.
My i940 was above avg for C0 in vcore lottery, was excellent in VTT lottery for C0 (prime 210 bclk with 1.3qpi/vtt), and OK in TDP lottery.
My i950 will do 4.4 23x192 with 1.34 vcore, 1.34 qpi/vtt.
But to prime 4.3ghz 21x205, I need 1.46 qpi/vtt. Others can do 21x205 with 1.2-1.3 qpi/vtt. In other words my i950 did well in the vcore lottery and poor in the qpi/vtt lottery.
My guess is your chip did not do so well in qpi/vtt lottery either (or vcore). And playing with the other volts may save a notch or 2, but I would fix the 90% problem first. If you need 1.42 or 1.44 or 1.46 qpi/vtt, you are not going to obviate that need playing with other things...you might make .01 difference, but worry about backing down a notch or two, after you fix the 90% problem. Above 195 bclk, my chip starts eating qpi/vtt to be prime stable.
But unless you have a ram problem, which I doubt, you likely are deficient in either vcore or qpi/vtt. You may or may not want to run your chip that high 24/7 (everyone has own view on this), but rather than wasting your time, I would set it there, prime it, and if that solves your problem, then back down as much as you can. Then play with other volts and see if you can take it down another 1 notch.
So with Vcore 1.35. I should try going higher on qpi/vtt?Quote:
Originally Posted by rge
If so how much further should I go MAX... where would I not go near as far as qpi/vtt voltage?
@MonsterDK
"Intel(R) Turbo Boost Tech ...................... [Enabled]" set to [Disabled]
this will give you 20x200
Try vcore 1.37, and qpi/vtt 1.42. If that does not work, your on your own:p:Quote:
Originally Posted by MonsterDK
If that works then walk down vcore, then qpi/vtt until you fail and find lowest setting for each...much faster with linx MAX memory testing with 5-10 runs than prime, but either works, you can run single prime once you decide what to run. Then, once you know what you require, YOU need to decide what max for vcore and qpi/vtt your willing to run at. But a few prime/linx runs at those volts are not going to hurt anything...providing you can control temps.
You might find sweetspot of your cpu is 21x196 or 4.1 instead of 4.2.
I am at 4.4 23x192 with 1.34vcore, 1.34 qpi/vtt. To get 4.5 to prime 1 hour I need 1.41 vcore and 1.44 qpi/vtt, in other words scaling goes to crap on mine after 4.4, and I am not going to use that much higher volts for 100mhz for 24/7.
And if you need high qpi/vtt for bclk 200 (like mine) then you may want to run at 21x196, no point in cranking either vcore or qpi/vtt (other than to figure it out) for 100mhz.
Anyone have a system identical to mine? need some help one getting passed 3800mhz
I don't have a Vcore 1.37 :( so will the nearest of 1.36875V be ok?Quote:
Originally Posted by rge
its pretty close ain't it? Other wise its 1.375V
For QPI/VTT I don't have a 1.42 :( I have a 1.415V and a 1.435V
But I don't want to go and blow up this :banana::banana::banana::banana::banana:... lol.... so I wanted to be sure what to go with here so I can start the testing.
Thanks
You are fine because there a chip using odd number like 1.315v and others using even number 1.32v. Nothing special.Quote:
Originally Posted by MonsterDK
Have you tried test your qpi/vtt using prime95(blend) like I said before. >1.4v is too much for 3200 uncore.
I assume your vdimm & vcore is enough and RAM timing is correct ; lower the qpi/vtt down until it will fail in the first bunch 1024k of the prime95(blend) and increase it step by step until it will pass the first bunch 1024k of the prime95(blend).
Then increase it a notch or two for the final qpi/vtt.:) After that you can start meddle with other settings like vcore , etc.
qpi/vtt is not just for uncore. Higher bclk independent of vcore will scale with qpi/vtt on some. I agree >1.4 for bclk of 200 with 16x uncore is not winning the cpu lottery...but it happens, just like needing 1.4 vcore for 4 mhz happens...although they both suck. Although I do agree with your method for finding qpi/vtt, but since he has been trying for few days and getting frustrated, i was just trying to see if possible to get stable then work down.Quote:
Originally Posted by Sparda
I can pop in my i940 and do 21x202 with 18x uncore (3636 uncore) and run prime 15+hrs with 1.34 qpi/vtt.
My i950 is prime stable 12+hrs, 4.4ghz, 23x192, 17x (3264 uncore) with 1.34 vcore and 1.34 qpi/vtt, 1.62 dram. But running 21x202 or 4.24 with 1.4qpi/vtt and near same uncore 16x (3232 uncore) fails in 2 minutes prime blend every time. Raising qpi/vtt to 1.44 and it will run prime for hours, no other voltage has any effect. My i950 did well in vcore lottery, but sucks in qpi/vtt one. I know you have had a few cpus...just wait til you get one with crappy mem controller then you will be a believer:D
see pic failing prime with 1.4qpi/vtt in 1-2mins. Raise qpi/vtt to 1.44-1.46, runs all day. For bclk 196 need 1.4-1.42 (for prime). For bclk 192 need 1.34 qpi, 190 need 1.32. Above 194 bclk mine eats qpi/vtt, just like my i940 ate vcore above 4.7 to 4.9 ghz.
http://img34.imageshack.us/img34/158...primecrash.jpg
I just assume he got a normal D0 that why I said it too much.:shrug: But I did outline how to find his qpi/vtt for him so that he can start meddle with other settings:) I did got one before that arent good with IMC but it C0. It need 1.38v qpi/vtt for 3600 uncore and cant even touch 4000 uncore stable.
On both of my D0 qpi/vtt voltages increase exponentially with the uncore ; 1.315v for 3600 uncore and 1.475v for 4000 uncore.
Yep, not disagreeing with uncore scaling with qpi/vtt, my i950 does as well at some point, just adding that bclk can also scale with qpi/vtt independent of uncore. My i950 will prime uncore of 3610 with 1.34 qpi/vtt if done by uncore 19x190 bclk, but needs 1.44 qpi/vtt for uncore only 3232 if done by uncore 16x202. Even tried slow mode to rule out qpi...but it is just bclk that is sucking up qpi/vtt on mine. And actually I like your way of finding qpi/vtt, it is more efficient that way for most.
Update on my D0
With the below I'm prime stable for 12hrs, I've only ran 10 passes of LinX but so far better than before and 3-4 degs less
Advanced CPU Features:
CPU Clock Ratio ................................ [ 20x]
Intel(R) Turbo Boost Tech ...................... [Enabled]
CPU Cores Enabled .............................. [All]
CPU Multi Threading .............................[Enabled]
CPU Enhanced Halt (C1E) ........................ [Enabled]
C3/C6/C7 State Support ......................... [Enabled]
CPU Thermal Monitor ............................ [Enabled]
CPU EIST Function .............................. [Enabled]
Virtualization Technology ...................... [Enabled]
Bi-Directional PROCHOT ......................... [Enabled]
Uncore & QPI Features:
QPI Link Speed .............................. ( x36) [6876MHz]
Uncore Frequency ............................( x16) [3056MHz]
Isonchronous Frequency ......................[Enabled]
Standard Clock Control:
Base Clock (BCLK) Control .......................... [Enabled]
BCLK Frequency (MHz) ................................[191]
PCI Express Frequency (MHz) ........................ [100]
C.I.A.2........................................... .. [Disabled]
Advanced Clock Control:
CPU Clock Drive ..............................[800mV]
PCI Express Clock Drive ................... [900mV]
CPU Clock Skew ............................. [0ps]
IOH Clock Skew ............................. [0ps]
Advanced DRAM Features:
Performance Enhance ...................... [TURBO]
Extreme Memory Profile (X.M.P) ........... [Profile1]
System Memory Multiplier (SPD) ........... [AUTO]
DRAM Timing Selectable (SPD) ............. [MANUAL]
Channel A + B + C
Channel A Timing Settings:
##Channel A Standard Timing Control##
CAS Latency Time ......................8
tRCD ..................................8
tRP ...................................8
tRAS ..................................24
CR.....................................1
Advanced Voltage Control:
CPU Intel Core i7 920 D0 Stepping
Load Line Calibration ................. [DISABLED]
CPU Vcore ..............................[1.4] (1.312v under load)
QPI/VTT Voltage 1.150v .................[1.335]
CPU PLL 1.800v .........................[1.3]
MCH/ICH
PCIE 1.500v ........................... [1.5]
QPI PLL 1.100v ........................ [1.3]
IOH Core 1.100v ....................... [1.16]
ICH I/O 1.500v ........................ [1.5]
ICH Core 1.1v ......................... [1.1]
DRAM
DRAM Voltage 1.500v ..............[ 1.66]
DRAM Termination 0.750v.......... [AUTO]
Ch-A Data VRef. 0.750v ...........[AUTO]
Ch-B Data VRef. 0.750v ...........[AUTO]
Ch-C Data VRef. 0.750v ...........[AUTO]
Ch-A Address VRef. 0.750v ........[AUTO]
Ch-B Address VRef. 0.750v ........[AUTO]
Ch-C Address VRef. 0.750v ........[AUTO]
General Questions about Voltages and more.
Q1: When giving voltage to a chip. Is there a stage where you can give the chip just enough in one area, be it VCORE or vtt or other, where the chip will work JUST within its limits also won't fail but also will not give the best results?
What I am asking is, if the voltage means there are no reboots under intense load through out the chip will the voltage be perfect as in its just within the scales or can you still hold the chip back by it performing on limited power?
Is it a perfect working chip situation or a problematic one at some stage? no in between?
Hope that comes across how I intended it.
Q2: Prime95 Testing: If all 1024K threads pass at different times (not all within 5 secs of each other) does this mean the chip is suffering on some cores? Or is that normal?
Thanks...
Seems a lot of vcore!? How are the temps??Quote:
Originally Posted by sunn
My Theory:Quote:
Originally Posted by MonsterDK
I believe the bigger differences are caused by open programs that require processing in one form or another thus slowing down cores ability to pass tests at the same speed as others.
Can anyone verify / shoot me down on this??
Yep. Varying background tasks running on cores occupy cpu cycles temporarily so slows prime to different degrees on cores. I found old post of mine that had quote from prime site regarding priority and sharing of cpu cycles of prime and other tasks.Quote:
Originally Posted by Jonny Manc
http://www.xtremesystems.org/Forums/...8&postcount=25
Is this a new thing for D0? On my 940 the only thing that would scale my uclk up was vdimm.Quote:
Originally Posted by rge
I look again at this matter and the ioh core just simply scale up with qpi/vtt at certain point if I put the ioh core on AUTO. :shrug:Quote:
Originally Posted by phsinc1
http://img199.imageshack.us/img199/7615/iohqpivtta.jpg
http://img194.imageshack.us/img194/8650/iohqpivttb.jpg
http://img3.imageshack.us/img3/1847/iohqpivttc.jpg
Advanced CPU Features:
CPU Clock Ratio ................................ [20x]
Intel(R) Turbo Boost Tech ...................... [Disable]
CPU Cores Enabled .............................. [All]
CPU Multi Threading .............................[Enabled]
CPU Enhanced Halt (C1E) ........................ [Disable]
C3/C6/C7 State Support ......................... [Disable]
CPU Thermal Monitor ............................ [Disable]
CPU EIST Function .............................. [Disable]
Virtualization Technology ...................... [Disable]
Bi-Directional PROCHOT ......................... [Disable]
Uncore & QPI Features:
QPI Link Speed .............................. [AUTO]
Uncore Frequency ............................(x19)
Isonchronous Frequency ......................[Enabled]
Standard Clock Control:
Base Clock (BCLK) Control .......................... [Enabled]
BCLK Frequency (MHz) ................................[200]
PCI Express Frequency (MHz) ........................ [100]
C.I.A.2........................................... .. [Disabled]
Advanced Clock Control:
CPU Clock Drive ..............................[800mV]
PCI Express Clock Drive ................... [900mV]
CPU Clock Skew ............................. [0ps]
IOH Clock Skew ............................. [0ps]
Advanced DRAM Features:
Performance Enhance ...................... [TURBO]
Extreme Memory Profile (X.M.P) ........... [Disable]
System Memory Multiplier (SPD) ........... [AUTO]
DRAM Timing Selectable (SPD) ............. [QUICK]
Channel A + B + C
Channel A Timing Settings:
##Channel A Standard Timing Control##
CAS Latency Time ......................7
tRCD ..................................7
tRP ...................................7
tRAS ..................................20
CR.....................................1
Advanced Voltage Control:
CPU Intel Core i7 920 D0 Stepping
Load Line Calibration ................. [ENABLE]
CPU Vcore ..............................[1.175]
QPI/VTT Voltage 1.2v .................>>>>>>>>test 1.3v / 1.34v / 1.38v
CPU PLL 1.800v .........................[AUTO]
MCH/ICH
PCIE 1.500v ........................... [AUTO]
QPI PLL 1.100v ........................ [AUTO]
IOH Core 1.100v .......................>>>>>>>>>test AUTO
ICH I/O 1.500v ........................ [AUTO]
ICH Core 1.1v ......................... [AUTO]
DRAM
DRAM Voltage 1.500v ..............[ 1.66]
DRAM Termination 0.750v.......... [AUTO]
Ch-A Data VRef. 0.750v ...........[AUTO]
Ch-B Data VRef. 0.750v ...........[AUTO]
Ch-C Data VRef. 0.750v ...........[AUTO]
Ch-A Address VRef. 0.750v ........[AUTO]
Ch-B Address VRef. 0.750v ........[AUTO]
Ch-C Address VRef. 0.750v ........[AUTO]
My i940 which was not very qpi/vtt sensitive, scaled uncore primarily with dram v, though to some degree with qpi/vtt.Quote:
Originally Posted by Xello
but my i950, holding bclk constant at 190, boots up fine at 19x uncore (3600 uncore) with 1.34 qpi/vtt. But change to 21x uncore (near 4000 uncore) will not boot. increase dram to 1.8, will not boot. Increase qpi/vtt to 1.44 and return dram to 1.64, will boot and prime for short time.
Probably just typical chip to chip variability, not sure if its chips in general that need more qpi/vtt that scale uncore more with qpi/vtt or not...need test many to make any generalizations.
How about with LinX? Sometimes I get a run that is 10% slower than the rest and other times it runs slower from the start at the same frequency.Quote:
Originally Posted by rge
If you run linx and prime at same time, linx will run nearly 2x slower, more apps running, less cpu time for linx.Quote:
Originally Posted by JKDC
I mean running LinX by itself I will get variation. I assume it must be background programs running though as I test it on a fully loaded OS. I tried to stabilize it for several hours the other night, but it won't happen with any setttings. I guess I have to live with it. lol
MonsterDK what batch of D0 do you have? I got 3851A that needs 1.4 set in bios to be stable :s
Hi,
First time poster on here, and I am wondering whether anyone can give me some help with the problems I am experiencing with my new i7 build using the UD5 mobo...
I put the system together about 3 weeks ago, and everything went perfect with the build, I then installed Vista x64 and then started testing for stability using Prime95, IntelBurnTest, OCCT, and 3d Mark Vantage...
But everytime I would run any of these apps, at stock CPU frequency and my RAM running @ 1600MHz, the system would crash after a short while, either a BSOD, completely lock up, or reset itself...
At first I had suspected that it was my old PSU, so I went and purchased a Be Quiet! Dark Power Pro 850w PSU, and also a new set of 3GB Corsair XMS3 1333Mhz RAM for testing purposes..
But the problems were still present after installing the new PSU and the new RAM, so I then shipped the CPU off on RMA for testing, and the company found no faults with the processor..
So I got it back and began more testing again using various different hardware configurations with all the stuff I have around the house, and each configuration produced the exact same results...
These problems seem to be extremely intermittent though, because one day I could run Prime95 Blend test for 14 hours without any issues, then the next day with the EXACT same BIOS settings and hardware config, it would crash after 10 minutes!!!
Also, at the start of last week just before I took my build apart so that I could RMA the motherboard, the first DIMM channel appeared to die on me, as the system would only pickup 2 out of the 3 RAM modules that I had installed, at first I thought one of my modules had died, so I tried out my 3GB set and it would only pickup 2GB of RAM.. so I tried just running the system with 1 stick in the first white slot, and another in the 2nd white slot, and it would only see 1 stick, so I shutdown again, and moved the stick from the 1st white slot into the 3rd white slot and started the PC up and this time it picked up both sticks of RAM...
I then tried booting the system with just one module installed in the first white slot but it wouldn't even POST...
I then tried running my 6GB set with 1 module in the 2nd blue slot, 1 in the 2nd white slot, and the other 1 in the 3rd white slot, and booted up, this time it recognized all 6GB but it was only running in dual channel obviously...
So I figured the first DIMM channel had died, so I got an RMA number and shipped the board off.. but when they tested out the mobo to see if the DIMM channel was bricked, they said that it picked up all of their 6GB of RAM without any issues!!!
Is it possible for a DIMM channel to stop working and then startup again all by itself???
They then tried to tell me that there is nothing wrong with the mobo because they managed to get it to run Prime blend for 1 hour!!
I spent 3 weeks with this system, and I couldn't get it stable @ stock speeds, and as I said, I tried loads of different hardware configs, I tested 3 different GFX cards, 3 different HDD's, 2 sets of RAM, 2 PSU's, built outside the case, inside the case, and I also sent the CPU off for testing...
Everything that I done here at home seems to point at the motherboard as the culprit for all these issues I have experiencing with this rig, but the people that I purchase it from are trying to say that it isn't the mobo because they spent 1 hour testing it, and it didn't crash!!!
How can they test for any flaws in just one hour!?!?
I really don't know what else I can do with this system, I have spent over £1800 in total on this rig, and it is pretty much useless, it wont even run stable at stock frequency, and the main reason I bought it in the first place so was that I could have loads of fun overclocking it, but instead I have had nothing but problems!!!
Could a RAM channel die intermittently like this??
Why would it stop working, then start up again all on its own?!?!
I hope someone here can help push me in the right direction as to what I can do next, as I am at a loss here, I really don't know what I can do regarding this matter...
Thanks in advance people...
Have you tried flashing the BIOS? It's possible it could be corrupted. If that doesn't work I would buy a replacement motherboard from somewhere that you could return the old one if it turns out that it is the problem. If the manufacturers won't help I don't see any other way. You'll have to pay restock fees and shipping to find the cause. You can't be stuck with what you have that's for sure.
Use a bootable usb key/floppy and spiflash.exeQuote:
Originally Posted by JKDC
This is not looking good at all if it on stock.:shakes:. Probably the board.Quote:
Originally Posted by BLunTSmoKeR
Since you have done lots of troubleshooting I assume you already flash to latest f8b bios , set the ram voltage and timing properly and test the sticks with memtest86+ right ?
View Single PostQuote:
Originally Posted by BLunTSmoKeR
http://img218.imageshack.us/img218/6057/ud5rama.th.jpgQuote:
Originally Posted by BLunTSmoKeR
Hello everyone, i`m very new at this, i have been reading and tried some templates but no lucky at all something must be missing here >( i would like to know if anyone could help me out i just want to get my memories at their speed because atm they are at 1066.
GIGABYTE EX58-EXTREME INTEL X58 SK1366
INTEL CORE I7 920 2.66GHZ 8MB Stepping D0
OCZ REAPER DDR3 3X2GB 2000MHZ TRIPLE CHANNEL CL 9-9-9-30 (CAS-TRCD-TRP-TRAS)
thanks in advance.
Hi guys, and thanks so much for the replies..
I have already flashed the BIOS, the mobo came with F4 BIOS installed, so at first I flashed to F6, and now I am on F7...
I also tried F8b, but it caused my system to BSOD as soon as Prime95 started working, so I reverted back to F7 again...
Also, I checked the CPU socket for bent pins, I have done this several times, and there are none whatsoever... As for voltages, I have messed around with the main voltages in the BIOS, increasing them slightly higher than the stock settings to see if it is due to there not being enough voltage being supplied, but I still experienced the exact same problems...
I also set all RAM timings manually, acording to my RAM's spec, and I have tried loosening timings up to no avail... nothing I have tried seems to work at all for me, and it is really starting to drive me up the wall...
The mobo is with the people I bought it from as we speak, but their testing methods are poor, so the instabilities I have been experiencing wont show up with them, and also the board is now recognizing all memory installed once again, which also puzzles me, because it definitely stopped working for me, that was the reason that I decided to return the board!!!
They ran Prime95 for 1 hour, and because it never crashed for them, they are telling me the mobo is fine, but it can't be fine if it isn't stable at stock frequency!!!!
I had to ask them to tell their technicians to try running Prime Blend overnight to see if it does the same thing that it does for me, so now I am just waiting the results of this overnight test to find out whether the board is the cause...
What else could it be if it isn't my mobo, that is what I want to know... I am starting to regret building this system tbh... this is my first Intel system I have build myself, and the experience has been a real nasty one to say the least!!!
It is really frustrating when you read posts about other people with very similar systems to mine running them @ 4+GHz and theirs being completely stable... while I am sitting here struggling like crazy to have a stable system @ stock frequency!!!!
:edit:
@jar600, you need to raise your memory multiplier in your BIOS to increase your RAM frequency, and also change those timings that you posted to their correct values to...
@ Sparda, why have you posted me a page out of the mobo manual about RAM configuration??
I have already seen that page, and I tried installing one stick of RAM in DDR3 1 for single channel, and it refused to POST completely, with error "b9" on the LED debug thing on the board... this is why I figured the first DIMM channel was dead...
And I only tried installing all 3 modules in the order mentioned above so that I could see if the board would pickup all three modules, which it did, but it was only running in dual channel...
When I tried running 3 modules in the correct order (all sticks installed in the white slots!) the mobo only seen 2 out of 3 sticks... this was the same with both sets of RAM...
Which slot do you class as DDR3 1??
I class the first white slot as DDR3 1, or at least thats what the mobo manual says...
::edit::
Also, yes all modules have been tested with memtest+86 boot disc, 1 set at a time, and also tested each stick indivdually and they are all fine...
hey BLunTSmoKeR . i just did that and finally i can start working with it, i add 9 9 9 30 1 and changed from 1.50 to 1.66 v cause this was the main problem because i couldnt pass 133 on cpu frequency, thanks again, its a start :).
@ BLunTSmoKeR, I would ask them to take the mobo back and pay a restocking fee if necessary, and send you a new one. That is why I will only buy some things like motherboads from places like newegg. If it does not work, you rma it, none of this in house "testing". Only one I trust testing would be manufacturer.
I would have 1st tried different RAM and slots (which you did). Second tried different cpu, though that would be very low yield and only since easy and if I had one and then I would have rma'ed the mobo, you have to try another mobo, that is next logical troubleshooting option. Anything else in inefficient. I would just not take no for an answer.
One other thing, I assume you have tried using load optimal defaults, and then ran that using prime WITHOUT changing your memory to 1600, ie leave it at what boots up. Changing RAM means OCing uncore and OCing ram (not according to your ram specs but Ocing according to cpu/mobo) because if works at true stock, ie optimal defaults then might be a setting. Though the passing sometimes failing others...again I would try another mobo.
I just post it since you seems struggle putting the ram in the slot. Nothing special.Quote:
Originally Posted by BLunTSmoKeR
Like I said maybe the board is no good because you got no post when you put single stick in the white slot near the cpu socket.
When I first got my board I only use single stick for some time at the same 1st white slot and it works fine. So it probably the board since you already know you got a working cpu and ram.
Ask them to put/test only one stick on the first white ram slot like you did. If they said it post ; something might be fishy or wrong since you already tried it with 2 different set of RAM.:shrug:
Hi again Sparda, I didn't mean to sound nasty in my last post, but I thought that by you posting the mem config page from the manual that you thought that I hadn't bothered to read it myself, or that I was installing the mem in the incorrect slots!!!
My mistake!
I really dont know what to do anymore with regards to this system, I still haven't heard anything back from the people that are testing it neither, so I am going to take it that they are still currently testing the mobo for me...
I really dont know what else it could be if they test the mobo over the course of a few days and it doesn't crash with them..
About flashing the BIOS, can you use any USB flash drive, or do they have to be a special type of flash drive?
I used the @BIOS software to flash via Vista, but I have just read on here that using this software is a bad idea... I dont have any floppy discs, or a floppy drive for that matter, and I no longer have a USB flash drive, but I can easily go buy one...
I flashed mine using a normal USB drive (formatted to FAT32) and am happy running on f8b and experienced no issues through the flashing process :PQuote:
Originally Posted by BLunTSmoKeR
Talking about flashing...anyone know when F9 bios will be available for UD5?
This is the exact same thing that happened to me!!!Quote:
Originally Posted by andyr(uk)
So I RMA'd the mobo for testing, and when they fired it up with 6GB it picked up all 3 modules!!!
Same here & I'm using spiflash because earlier bios has some Q-Flash issue.Quote:
Originally Posted by Jonny Manc
Hello!
Working with a user-summer-friendly OC @3,36GHz (21*160) for now what with all the warm weather and temps in Greece.
With xmp enabled i get 1.65v (1,64 actual) and 8-8-8-24 timings just fine with no manual tuning.
Any suggestions, specially in volts would be appreciated, I think that maybe some need a lower value than auto.
(Auto on vcore gave me 1,36v! Hit 95C on the first 20 seconds in prime :D )
A suggestion on hyper-threading would be appreciated too.
Since i m not running any benchmarks and the pc is for gaming mostly should I turn it off? (as i understand more volts are needed when its on... )
F7 bios, changes made:
Advanced CPU Features:
CPU Clock Ratio ................................ [20x]
Intel(R) Turbo Boost Tech .................. [Enabled]
CPU Cores Enabled ............................ [All]
CPU Multi Threading .......................... [Enabled]
CPU Enhanced Halt (C1E) ................... [Enabled]
C3/C6/C7 State Support .................... [Enabled]
CPU Thermal Monitor ......................... [Enabled]
CPU EIST Function ............................ [Enabled]
Virtualization Technology ................... [Disabled]
Bi-Directional PROCHOT ..................... [Disabled]
Uncore & QPI Features:
QPI Link Speed .............................. x36
Uncore Frequency .......................... 21
Isonchronous Frequency ..................[Enabled]
Standard Clock Control:
Base Clock (BCLK) Control ................ [Enabled]
BCLK Frequency (MHz) .....................160
PCI Express Frequency (MHz) ........... [auto]
C.I.A.2 [Disabled]
Advanced Clock Control:
CPU Clock Drive ..............................[800mV]
PCI Express Clock Drive ................... [900mV]
CPU Clock Skew ............................. [0ps]
IOH Clock Skew ............................. [0ps]
Advanced DRAM Features:
Performance Enhance ...................... [turbo]
Extreme Memory Profile (X.M.P) ......... [Enabled]
System Memory Multiplier (SPD) ........ [10] (auto gave me 12 with ram@1920)
DRAM Timing Selectable (SPD) .......... [auto]
Advanced Voltage Control:
CPU
Load Line Calibration ................. [Εnabled]
CPU Vcore ...............................1.125
Everything else missing, on default or auto.
Results in the attached image with prime running in the back.
http://img4.imageshack.us/img4/3089/ocsetts.png
Thanx in advance mates :)
Hi
What is the newest/best chipset drivers to use with the extreme? Can you just use the Vista ones?
I rma'd the board as i did find some bent pins on the zif socket , i have problems with my ram , it wont boot with two of the sticks i have so i am down to one stick in slot 1 , waiting on corsair to get there rma into gear.Quote:
Originally Posted by BLunTSmoKeR
I might get some patriot to tie me over, very good price for 6 gig atm
I have also RMA'd my mobo, but when they set the mobo up and installed 6GB of RAM the board picked up all 3 modules...
I also checked the CPU socket for bent pins, and they were all fine.. I checked the socket every single time I had to remove the CPU for any reason and there is definitely no bent pins in there...
If I were to RMA the mobo directly with Gigabyte UK, and they found the board to be faulty, would they actually replace the mobo for me, or just repair the current board??
I sent the board back to the e-tailer for testing, but so far their methods of testing have been extremely poor.. to begin with they told me that the board wasn't faulty because they installed 6GB of RAM and it all showed up, they never installed Windows and tried any stress testing, they just viewed the POST screen...
Then I called up and explained how it used to see all 6GB of RAM for me when I first set the board up, and how it was due to the board being really unstable that I shipped it back to them, so they ran Prime95 Blend for 1 hour and tell me yet again that there is nothing wrong with the board...
I then had to plead with them to leave the system setup overnight running Prime Blend for the problems to showup, so that is what I am awaiting the results of as we speak... I am praying to god that the board acts up for them this time, because I am 99.9% positive that it is a faulty mobo...
With the 3 weeks I spent with this mobo and all my testing, everything points towards the mobo as being the problem... so if they turn around and tell me that it isn't faulty, I have no idea what steps I can take next??
I am also annoyed at the fact I had to pay £21 to ship the mobo off, and now I am also expected to pay them another £23 to get it sent back if it isn't faulty... that is a joke IMO... When I RMA'd my CPU and it was found not faulty, I wasn't charged by that company to get my CPU back, they covered those costs themselves...
The only thing I can think of doing next if they say it isn't faulty is to test it out again at home, and if the problems are still present with me, then I should ship it off to Gigabyte to see what they say...
But these problems have been with the mobo since the first day I set it up, it initially had F4 BIOS installed, and I have a D0 chip, but the board still booted up with F4, even though Gigabyte say you need to have F5 minimum for proper D0 Stepping support, but I did try some stress testing with F4 BIOS, and it was extremely unstable with that BIOS, so I went onto F6, again the problems were still present, but there were times when the system would Prime away overnight without any problems, but then the next day with the exact same settings the system would crash within 10 minutes of Prime...
Then when F7 BIOS came out, I installed that, and it didn't really change anything, the system was still unstable.. then someone made me aware of beta BIOS'es from tweaktown I think it was, so I tried F8b and as soon as I started Prime it would BSOD on me every single time... and all this is at stock frequency remember, I haven't really done any overclocking on this thing so far... because of my inability to get the system stable @ stock!!! :mad:
Does anyone here have any idea as to what else could be causing my problems??
Like I said, I have tried as many different hardware configurations as possible with all the hardware I have here at home, and every single one gave me the exact same results... also with the 1st DIMM channel dying on me and then miraculously working again when the retailers set it up is a mystery to me...
I hope that someone here could guide me in the right direction with where I can go from here with this rig, because I am at a complete loss as to what I can do now...
The only hope I have left is that the tech who is testing the mobo actually gives the board a good thorough testing and that he also experiences the same issues as I was having... then I will get a replacement mobo sent out and everything will be fine, but if they only test if over the course of 1 day and it behaves itself like it has done for me a couple times in the past, I dont know what else I could do...
Thanks again for your replies so far people, it is much appreciated...
P.S. here is a rundown on all the hardware I have used for testing in total:
GFX cards: 8500GT 512MB, 8800GTX 768MB, GTX280 1GB (All cards have been tested in another machine and each one works perfectly)
RAM: 3 x 1GB Corsair XMS3 1333MHz, 3 x 2GB Corsair Dominator 1600MHz (tested with stock timings and much looser timings on both sets, each module has been tested indivdually and also as a set running memtest bootdisc and they are all fine)
PSU: Hiper Type-M 630w, Be Quiet! Dark Power Pro 850w. (lots of people had told me the issues I was experiencing was probably caused by my old Hiper, so I went and bought the Be Quiet! only to find the problems were still present!!!)
HDD: I have tested the system using 3 different SATAII drives, to see whether it was a faulty drive that was causing the problems, but again the issues were still present..
I tested the rig using about 5 different SATA cables as well to see if it that made any difference to, but nothing I tried made a blind bit of difference...
The rig was built up outside the case with the absolute bare minimum connected to it, also tried using just one RAM module (this was before the RAM channel died on me) installed in the first white slot, and I thought that everything appeared to be OK with just one stick installed, so I decided to leave it Priming overnight, but when I returned to the PC about 7 hours later, it has BSOD'd on me again...
As you can see I have tried lots of things with my rig, and nothing I have done seemed to make any difference at all..
Did you buy your kit from the same place?? If so, send it all back to them for some proper testing!!
Hey guys... I'm interested in this motherboard with an Intel W3520, anyone running the two of these ? Any BIOS issues with the Xeon versus the 920 D0?
Thanks
http://www.hwupgrade.it/forum/showpo...postcount=7085
:yepp:
ANYONE TRY?
Flashed and are working fine, AHCI is 1.20E and no delay in HDD detection...
BUT! this bios have som changed default values, so check settings, and worst of all? the uncore multiple are somehow limited to 16x then it goes to AUTO, and in this bios auto means the highest against BCLK, so 200 bckl gives auto 3800mhz. and either that or 16x or lower, why remove such feature?
Quote:
Originally Posted by Jonny Manc
I got the mobo, Be Quiet! PSU, and both sets of RAM from one place, and I got my PSU, GFX, etc etc from another online shop...
They did ask me to send my RAM back also, but they were ordered at different times, so I have been told if I send it back and the RAM is found not faulty, I will then have to pay shipping and testing fees for the RAM also, which I am not willing to do as I know that there isn't anything wrong with my RAM..
I am 99.9% sure that it is a flaky mobo that is the cause..
I regret dealing with these people now, because of the hassle I am having with this system... Since buying this setup, I have seen an excellent bundle deal on the website where I got my CPU from, which consists of a D0 i920, the same mobo as me, 6GB of Patriot 1600MHz RAM, and a Titan Fenrir, all O/Ced to 4GHz for around £539, this made me feel sick when I first saw it, as I paid around about the same for all my gear, and it is unable to run stable @ stock, nevermind 4GHz!!!!
At least if I bought that bundle deal it would have been thoroughly tested first before being shipped out to me, so at least then I would know that the hardware I bought was working properly first before I built it up...
Can anyone tell me whether Gigabyte would replace the whole mobo if I returned it to them for testing, or do they just repair the damaged board???
Always a bummer when you try to get best deals that end up causing more hassle. I assume that its OCUK that you are dealing with!?Quote:
Originally Posted by BLunTSmoKeR
Never had an issue personally, but there is always a first :D
If they asked you to send the RAM back, then I would not accept any charges in relation to testing as THEY are requesting it.
Personally, I would speak to GigaByte and see what they suggest, they may 'lean' on the shop to test it fully to try an ascertain if there is a problem.
Nah man, OcUK are top notch, it's Scan that I am having probs with...Quote:
Originally Posted by Jonny Manc
It is OcUK that has got the 4GHz overclocked i920 D0 bundle with the same mobo that I have, though when I got my mobo and RAM from Scan OcUK didn't have the 4GHz bundle available, and Scan had the UD5 and my Corsair Dominator RAM slightly cheaper than OcUK...
Now I wish that I just paid a little more and went with Overclockers instead, as it was them that I bought my CPU from, and I RMA'd that to begin with as I thought the chip was the cause of my problems, but they tested it overnight and it was fine... so when I got it back from them I installed it again and started running more tests but continued getting the problems that I mentioned, and then the 1st DIMM channel stopped working, which is why I decided to send the mobo back....
God knows what to do now... I am just hoping that the board fails so that I can get a replacement sent out...
the online shop i got my mobo from wasnt ocuk but it didnt respond to my request about the problem with my board so i emailed taiwan and they gave me a link to Gigabyte uk , they dont normally deal direct with customers but if you explain that you arent having any joy from your etailer then they will agree, you have to enclose a cheque for £10 for the return postage and if they cant fix it they will send you a new one , mine came from Holland , took 2 weeks but good service really, where do you live in the uk m8
OCUK are good, no arguments there, but I agree that Scan can be a pain in the arse. Where abouts in the UK are ya....cos you can always go in and kick off with em, I would, but I am only about 1 1/2hrs away from em :p:Quote:
Originally Posted by BLunTSmoKeR
andyr(uk) - sounds like a great idea from GByte, Asus do something similar but outsource the 'support' to a 3rd party in the UK (yet operate under Asus UK Support !?) and they suck and are really unhelpful unless you show em the cash!!
Hey guys, I am from North East of Scotland, Elgin to be precise...
I talked to Gigabyte when I first setup this system about their RMA service, and they were quite happy to test the mobo for me, but I didn't ask them whether they would send me a new mobo if this one was faulty or if they just repaired the current mobo... They also said I had to enclose a cheque for £10 as well...
I just wish now that I had sent it to them in the first place instead of messing around with these idiots at Scan... though I figured that the RAM channel was dead, so as soon as they seen that, they would just send me a new mobo, but the darn thing started working again!!!
I sent 3 e-mails since early yesterday morning asking them if there is any word on my mobo yet, and no-one even bothered to reply to me, that is really poor IMO...
I'm not caring if they haven't had a chance to thoroughly test it yet, but they should still reply to my e-mails to tell me what is going on with MY mobo when I ask them!!!!
This rig has been a nightmare from the very beginning... I built this system so that I could have loads of fun overclocking it to 4GHz, and playing all my games, but instead I have ended up with a piece of crap!!!!!
Oh bit far away to come round and test your bits really ,mine looked to me to be a bad memory channel ,but since i got this new board it looks like i have two dead sticks of ram , one stick works fine in slot A1,the second wont get into windows and the 3rd wont boot at all and corrupts my OS.
I wish that I knew of someone else in my area with an i7 rig so that I could test out my hardware on another system, but I don't know of anyone at all...
Everybody I know with PC's are all running crappy old P4's, or Dual Core Athlons...
I am just hoping that the techs at Scan are thoroughly testing my mobo over the weekend, and that it starts acting up for them... if it doesn't play up, then god knows what else I can do, as I have tried absolutely everything I can think of...
I really dont think it is my RAM that is at fault, because each stick tested individually running memtest boot disc will run flawlessly overnight without a single error, and they also pass with flying colours when running all 3 modules together as well...
All the hard drives that I tested with work perfectly in my Fiancee's rig, as do all 3 graphics cards, and 2 PSU's, I also tested with a bunch of different SATA cables to make sure I never had faulty leads, but they are all fine...
As I have already said, I returned the CPU back to OcUK for testing, and it ran perfectly for them... although they only ran Small FFT's in Prime95, and 32bit Linpack in OCCT for 3 hours each, which isn't really thorough in my book...
I have tried pretty much everything... and I am not getting anywhere at all... Sometimes I wish that I had just stuck with AMD and built a Phenom II rig instead!!!!
Its not at the benchmark it should be
it should be next to the 975, like the fpu bench.
Im using a EX-58 extreme with f7 original.
http://img41.imageshack.us/img41/8940/alunotright.jpg
Elgin eh.....not too far from my mum (Portsoy) but a bit too far from me (Cumbria) lol.Quote:
Originally Posted by BLunTSmoKeR
If the muppets (Scan) find no error with your board, I would trial and error all your components in your system. ie start with the least to get the system powered, then add more drives, RAM etc and see if that brings up an error. After that, if your still having problems, I would take up Giga's offer and send the board to them for proper testing.....
Hi Johnny,
Are you originally from up here as well, or did your mum move to go and stay in Portsoy??
My Dad lives even closer to her, he lives in Buckie...
As for the Trial and Error thing, I have already tried this, by building it up outside the case with the bare minimum connected to it, but the problems were still present...
There really isn't anything else that I can think of trying other than sending it to Gigabyte if the clowns at Scan tell me it isn't faulty...
All the components that I have used for testing (except the CPU and RAM) have been tested out on my Fiancee's system to see whether it is those that are faulty, but they are all perfectly fine... and as I said in a previous post the CPU has been tested by OcUK and no faults were found...
Though OcUK's testing was pretty poor IMO, 3 hours of small FFT's and 3 hours of 32bit Linpack isn't what I would call thorough, especially when I told them before they tested the CPU that it passes Small FFT's and 32bit Linpack without any issues, and that it is Large In-Place FFT's and Prime Blend that causes the problems.. but when they recieve the CPU they someone only run the two tests that I told them were OK!!!!
And when I called them up to complain about this, they told me to just go ahead and return the CPU again!!!
I might just look into purchasing a new mobo to see if that makes a difference... although TBH I really dont want to spend another penny on this rig, because it has already cost me over £1800 all in!!!!
hi
is anybody using the 8b x bios with the revised rom and is it worth it
if you need to test your hardware i live Yorshire if its any help
cheers
mark
My mum and dad moved up there a few years ago, around the same time I moved to Cumbria.....not had the pleasure of a visit yet, but I have seen loadsa pics (including Elgin) and cant wait to get up there ;)Quote:
Originally Posted by BLunTSmoKeR
I would defo return the board to GB and mention how crap and unhelpful Scan were in trying to solve the problem.