Printable View
I read something about it somewhere, the same instability was being experienced with a quad, and setting CPU margin enhancement to 'performance' rather than 'optimized' solved the problem apparently. I would have thought it would make it worse, but maybe it's worth a try. If it has :banana::banana::banana::banana: ed your cpu, it can't hurt it much more to give it a go, either way they owe you a cpu if it has damaged it.
Hi there!!
Here i am again with another doubt. :confused:
I´m using Prime95 and MemTest to test the stability of the system. All tests run smoothly, until it reaches to the blend test, where i get an error in the core#2 after a few minutes.
What can it be? vNB? vFSB? vRAM? RAM timings?
Rig specs:
Q6600 (cooled by Tuniq)
P5Q Deluxe
GSKILL PC2-8500 PK F2-8500CL5D-4GBPK
Powercolor HD4870
2x Seagate 250GB@RAID0
Corsair HX620
BIOS settings:
AI Overclock tuner: Manual
CPU Ratio Setting: 8X
FSB Frequency: 425
FSB Strap to North Bridge: 266
PCI-E Frequency: 100
DRAM Frequency: 1066
DRAM CLK Skew on Channel A1: AUTO
DRAM CLK Skew on Channel A2: AUTO
DRAM CLK Skew on Channel B1: AUTO
DRAM CLK Skew on Channel B2: AUTO
DRAM Timing Control: Manual
CAS# Latency: 5
DRAM RAS# to CAS# Delay: 5
DRAM RAS# Precharge: 5
DRAM RAS# Activate to Precharge: 15
RAS# to RAS# Delay : AUTO
Row Refresh Cycle Time: AUTO (defaults to 52)
Write Recovery Time: AUTO
Read to Precharge Time: AUTO
DRAM Static Read Control: Disabled
DRAM Read Training: AUTO
MEM. OC Charger: AUTO
AI Clock Twister: Moderate
AI Transaction Booster: 8
PULL-INS: All Disabled
CPU Voltage: 1.29
CPU GTL Voltage Reference (0/2): 0.630
CPU GTL Voltage Reference (1/3): 0.670
CPU PLL Voltage: 1.54
FSB Termination Voltage: 1.36
DRAM Voltage: 2.00
NB Voltage: 1.26
NB GTL Reference: 0.630
SBridge Voltage: 1.10
PCIE SATA Voltage: 1.50
Load Line Calibration: Enabled
CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
CPU Clock Skew: AUTO
NB Clock Skew: AUTO
CPU Margin: Optimized
Advance CPU Settings
CPU Ratio Setting: 8
C1E Support: Disabled
Max CPUID Value Limit: Disabled
Intel Virtualization Tech: Disabled
CPU TM Function: Disabled
Execute Disable Bit: Disabled
Blasar, have you tried upping the CPU GTL Voltage Reference (0/2): 0.630 to 0.067. A lot of times raising this setting will help when it is always the same core failing.
True, I don't think it has messed his up though. I'm thinking this realy does just fall onto the 9450s and alike. I even tried getting Vista to boot by changing those settings and it did me no good but, you are right I remember the manager saying that he hit on that as well some where about switching it and the issues stopped for some people.Quote:
Originally Posted by CryptiK
Well it is sold so me no care about it .. :)Quote:
Originally Posted by CryptiK
If it dies on friend well he can send it in to intel - 3 year warranty as its only about 2 months old, but 3 weeks ran as i had to wait for mobo to arrive.
I ran Optimized before always like my Signature Template is my settings for the 3.6ghz and now it not do linpack anymore .. raising PPL or NB or em did nothing.. like before i had to raise PLL and NB for 1.4 to pass and those are the settings i used always which is my template settings.
Now can do 3.4ghz and maybe 3.5-3.540 if i am lucky, but 3.560 BSOD, just like 3.6 ghz does now.
If Asus owes me a CPU for it, np's as i won't bother as i getting E8600 which will smoke that Q6600 in gaming and many other things.
If the E8600 gets :banana::banana::banana::banana:ed by all of this then i will demanding a new CPU for sure.
Now until it arrives and i test it out to see if it is a good batch number, but most do 4.0ghz easy and if i can't pass linpack 1.4 . (well settings have to be good of course) and if they are good and can't pass ... then i would state/claim its a mobo issue.. if i tried all i can do and what others suggest and still no go on e8600 .. then i will scream at Asus for a new mobo.
I am only guessing its degration and if so then so be it as it would have done so over longer period if never linpacked.. if it is Asus CPU margin enhancement issue then I STATE FOR ALL PPL TO BE CAREFUL as over time if not linpacking for as much as i did it for.... maybe 6 months your CPU maybe come instable from degration.... if linpacked then you could end up like me 3 weeks a total of maybe 6 days of linpack testing here and there like i noted . = degration.
Or maybe Linpack hurt my MOBO... i will know better when i get the E8600 in to see whats what .. but even if new cpu clocks high and linpacks stable using MAX memory available.. it still may degrade sooner then most may think.
I personally think my Q6600 max was 3.6ghz and it ran great at my settings and you can see it was NOT hardcore settings .. and after linpacking/folding and Orthos testing may have broken the CPU in = BURNED in and it no longer can run 3.6ghz linpack stable (again i never tried Orthos since) as i am not wasting 12 hours or 3 or 6 to find out it fails or not... if it linpacked before.. it still should do it now... plain and simple and if it can't ... then CPU is degraded is it Linpack .. is it BURN IN .. is it Asus CPU Margin Enhancement.. i can't say 100% yet.
I did have a Opteron 146 that its max was 3.2ghz, but it was not stable over 2.950 and after about 2 months running at 2.950 at good volts and all .. one day it BSOD on me and i said WTF? and thought maybe a driver issue or what not .... well sure enough it failed prime 95 back then i like prime 95... and i had to back it all the way down to 2.8ghz ... as anything above it... FAILED or BSOD.
Now was it MOBO, was it CPU degration or was it something else... well for instance i took it out and put into another good Ocing mobo and again ... nothing above 2.8ghz would run stable . heck i could not even boot to 3.0 ghz, so forget about the orginal MAX 3.2ghz.
I just hope my E8600 is a good one and i can do 4.0-4.4ghz at stock volts to 1.35-1.39 ... as that would be awesome, but again if degration is an issue on newer chips ... i will find its max and back it down 200-300 mhz.. or let it ride out until it degrades.
Just discovered a weird bug.
Did a fast test to see how far i could go with 4x1gb ram, all went fine and maxed out at 500FSB PL8 and 5:6 divider = 600MHz 5-5-5-15 at 2.2volt.
Now to the weird part, took out two of the sticks, booted up again -same setting, but now it maxes out at 490FSB, all timings in Memset are the same.
Shouldn't it be the other way around :confused::p:
Setup
p5q-deluxe bios 1306
e8600
2x1gb crucial pc5300 + 2x1gb cellshock pc6400
vram = 2.20v
vnb = 1.26
vfsb = 1.26
And a screen
Not really, as you prob needed the 2.20v ram volts with 4 sticks, whereas with 2 you are giving too much voltage is my guess. I'd try less ram volts and see if you get it back up there.Quote:
Originally Posted by noxon
Larry
I now have a really weird stability issue - I was linpack/orthos stable at 4GHz/DDR1186/PL8, now I only pass 1 out of 5 tests in linpack at the same settings as before. It's not my ram, tested it on a mates board and it's fine, tested cpu on mates board too and it's also fine. I have tried going back to bios 0803 and still the same. Then also I can boot up fine, run linpack etc, then when I reboot it fails to even post. I can try numerous times and all I get is a black screen. I have to clear cmos to get it to boot. I want an RMA, this coupled with the inability to run DDR1200 despite everything being tried since I got the board has confirmed in my mind the motherboard must have a fault.
Thing is I cant find Asus Australia's contact details, they have an email sevice but I have to fill out about 50 fields before I can send the report to them and I would much prefer to just talk to someone in their tech department to arrange the RMA. I have tried everything I know, and I have simply had enough of this boards problems.
Now does that not sound a bit familar to my issue .. so is it board or is it CPU in my case .. well i think mine is CPU, but again it can be the memory controller on mobo.Quote:
Originally Posted by CryptiK
Remember the AMD CPU's if you push to hard the on board CPU memory controller will flake out and then POOF error out or BSOD.
CPU margin enhancement to 'performance' rather than 'optimized' ??
Suggestions for setting this for a QX6850?
Are u serious?Quote:
Originally Posted by Big Lar
Well i tried PL9 and then i could to up to 520FSB, 10FSB higher than 4x1gb and PL9, tRD_WR from 7 to 8 gained another 8 fsb ending up at 528fsb - 1266MHz DDR, other settings same as i posted before.
Another thing, i cant boot at 460-533fsb 1:1 if i set PL manual in bios, at 5:6 there is no problem.
So there seems to be ALOT bugs still!
With my cpu in sig I have it set to Optimized which is the default, and so far am good to go.Quote:
Originally Posted by xgman
Larry
More VCore my man, that's what you need. You're too low unless you have a golden Q6600. Try 1.325-1.350.Quote:
Originally Posted by Blasar
Same thing I was thinking.Quote:
Originally Posted by WildstarOCT
Well my Q6600 runs 1.35v (in Bios) ---- 3.4ghz NP's and stable.
What volts are you running? My board will reboot and stall if i set fsb or nb volts too high, over 1.32 for either and it will do it to me if my pc crashesQuote:
Originally Posted by CryptiK
I have found an issue with my board that my be of some help to people.
Yesterday, i posted a picture of orthos running stable for over an hour. Today I thought i'd test it again after booting up and i got bsod after 7 seconds lol! Anyway i rebooted and re-run orthos, running stable again for half an hour so i stopped it. Ran it again, boom, bsod 2 seconds in.
Tried a few different voltages to try and resolve this, no joy. All combinations resulted in same synmptoms as above, stable pc, re-run orthos and then bsod after a few seconds.
Turns out the problem was Memory pull ins. They were all set to enabled. Disabling these has brought me a fully stable system again. Can stop and start orthos as many times as i want, no bsod. Disabled pull ins has had no effect whatsoever on my Everest memory bandwidth scores.
So if anyone is having issues with thier pc's rebooting or bsod and you have got pull ins enabled, try disabling them :up:
EDIT: with pull ins enabled the system was memtest stable too, but like i say it would cause blend to bsod
Well if anyone cares to read my template ..... I always run Pull ins Disabled.... so that ain't my Q6600 issue.
Thanks :D I have 1306 but I've seen tigs say 1603 a lot in various forums, guess he made a typo :pQuote:
Originally Posted by CryptiK
ED - Are people still finding that the 400MHz FSB strap isn't working properly (namely, when you try changing memory dividers) and the PCI-E frequency option still KOs the video signal when set beyond 101MHz?
I get a no post on my P5Q Dlx from 108Mhz on... with the 8800GTX, GTX 260 and 280... seems slower cards have less issues as I've seen a 7600GT do 120mhz...
good to have ya here Ket aka mr Bios :)
I'm back to 803 here, 1306 was too flakey for me. I have my 8800gtx running full OC with no probs here on 1306 or 803 bios. I'm at 8 x 485fsb at the moment as well if that matters, only tested Linpack with 10 runs so far tho. CPU volts 1.30625v with LLC Enabled now.. only way it will run.Quote:
Originally Posted by Leeghoofd
Larry
What is your PCI-E freq at mate ? as soon as I go over 108 even with stock clocked cards I get a no post...
Running regular old 100 for pci-e. For volts=1.50v. Tried Ket's bios, got a winner there :)Quote:
Originally Posted by Leeghoofd
Larry
BSODs days after verifying stability is usually NB voltage, or ease up on your mem setting.
long hours of harsh stress testing doesnt mean much to the P45, it could still crash a simple game in minutes.
I can't go over 103 because the raid controller will flake out and drop my raid. I am going to get the velociraptor and that should enable me to push it as far as I can.