Printable View
I wouldnt think the cpu is limiting me here since I have tried changing the multi down to 6.0 and it didnt help any. I know some of the GB boards are benchable at around the 550 mark, so im wondering if just over 500 is the normal max for the P5Q with a quad.
To get higher you need to get off the 1:1, also better to have 1Gb dimms (max of 2GB) up the FSB term will make you be able to hopefully get higher... I had issues to get the quad stable at 4.5 (9 x 500) with 500FSB and 1000Mhz on the rams ,with 1200mhz ram speed it was easier to do... The GB boards have got the A3 rev of the NB and the board pushes way more volts... for 24/7 I prefer the Asus board as I got the same clockspeed with way less NB and FSB term voltage... for benching the Gb might be better... but does that NB get scorching hot...
My P5Q-Deluxe had a hard time going 500FSB for 24/7 on my quad. I made it stable too, but it seens the board was really at its limit.
Now with Maximus II Formula, things were easier...
Asus Maximus II Formula is doing 500FSB on my 65nm B3 quad for 24/7 now, with DDR2-1200 5-5-5-18 PL8 STRONGER. 100% stable with very low volts on everything. Bios 1901 and 100% aircooled rig / closed case. NB max temp is about 42C during stress test.
That is my 24/7 settings now. I am trying to make it stable @ 508-515FSB now, and I can boot @ 530FSB too.
I am using 2x1GB Corsair Dominator PC10000C5DF.
1.42v CPU
1.35v FSB
1.59v PLL
2.09v DIMM
1.44v NB
1.19v SB1.1
1.55v SB1.5
http://img154.imageshack.us/img154/307/32603469ow2.jpg
http://img296.imageshack.us/img296/2472/mf2qqn8.jpg
http://img388.imageshack.us/img388/6536/full3mj8.jpg
http://img388.imageshack.us/img388/987/full4is8.jpg
[]'s
Simps
nice clocks but you need more fans in your rig
Hehehe! They just don't fit anymore! I only use undervolted <28dB fans, so despite the number of fans, the noise level is really low. Also I keep it with closed door case, and inside of a cabinet I have here... You can't almost hear the thing.Quote:
Originally Posted by Gen
The temps are "water-like". Check out my NB / SB and PWM load temps, during 2h of prime95 @ 3.2GHz.
This is on a Maximus II Formula. Rev02 chipset.
http://img504.imageshack.us/img504/2604/temps1gi4.jpg
[]'s
Simps
55 ~ 67C ?! I think that's way too much for a 45nm... If you want to keep that OC you should definitely go for watercooling. I'm not even comfortable with those kind of temperatures with a 65nm, let alone a 45nm.Quote:
Originally Posted by Rise
Nice OC tho' :) Nice vcore :clap:
Hoss331: every CPu has an fsb limit, which you cant pass, the quads mostly around 460-480( depends on if it is a Kentsfield or York ) , this means lowering the multi( = clock ) dose not help much. 500+ with a quad cpu is a gift,a bonus, what the avarage 4 core cpu-s cannot reach.
Simps: Can this machine lift up in the air at least? Just dont forget to fasten your seatbelt before the take off:hrhr:
Somtihng else...
After suffering with experimentig with BIOS settings to stabilize my DIMMs on 1200, and not getting a step forward, I decieded to leave almost every setting in BIOS on auto, I set vDIMM to 2,1V, fsb strap to 333, CPU multi to 8, vCore 1,23125V, Mem OC charger enabed, Dram static control, read training, CPU / PCIE Spread spectrum to disabled, and tried so, and I surprised, that the system was more stable, but not 100%. So I tweak some on the settings, I set PL from the default 8 to 9, didnt had an efdect on stability, but after that I enabled all the pull in-s ( I thought why not? I never did that before ), and the system was completly stable. After this I just wanted to clear up, what the boadr set on auto for vNB, vFSB, VPLL, cause dont wanna that it some of these get overvolted, and I am curious, what I set before wrong. So first I tried different levels of vPLL to fing out what the board has set, I tried from 1,50V, 1,52V, 1,54V ...and so on, but the system wasnt stable again, I wasted with the test and the system freezes an hour or more, until I set pll to 1,7V, and that was the correct value. I tried it on 1,72V and so it is unstable again. A thought man, whatta heck?:shrug: I dont even thought that the PLL anyhow would influence dram OC, and if it would, then the problem could be just that it dont gets enough voltage, dosent thought that the voltage could be to much. I dunno what the values of the other voltages( vFSB and vNB ) could be, but I dont wanna spend more time probing that. Someone could explain me what these pull- in phases are, and how they effect dram stability, and how pll is related to dram overclock. And of course I 'd fain know somehow without experimenting with BIOS settings what voltage is set on auto for NB and VTT, is there some software what can read this from BIOS? :confused:
Yeah, my chip has always run hot. At stock, it runs 50-55C under full load. When I first got it, it was running as high as 75C with the side of the case on... let's call it a burn test :p: TJmax was still pretty far off, so I'm OK with it as long as its under 65C, on air.Quote:
Originally Posted by Caped Crusader
Right now it's actually not running any higher than 60C (after 12 hours) after opening the window :rofl: But, you're right with the watercooling and I just finished leak testing the setup I'm going to put in there and should be installing by the end of the week sometime. I am going to try and raise the FSB to 466 with the same vcore and see what happens once I have the WC loop setup. I may even put up a work log! :up:
Sorry for the Autoquote.Quote:
Originally Posted by omiez
Leeghoofd what do you think about these runs ?
I am trusting a lot in IBT cause it a Very High Stressing Software to test Stability and i reccomend it to all who want to test stability.
I also tried today to set 430x10 without problem and i also activatet C1E to see if the system crashes in Games but it was 100% running.
4300Mhz with 1.28V on CPU.
But for daily use i think i will use 420x10 or 450x9 (if i can eneble C1E for the last one).
I will let you know further tests.
Yeah, I guess some batches are hotter than others. My Q6700 runs pretty hot too, although I suspect the base of cooler is loosing some contact due to the loosening of one or two push-pins...Quote:
Originally Posted by Rise
Anyway, good luck with your new setup! :)
Need some help getting my P5Q-E stable at 450FSB, can run the CPU test in OCCT but memory and mix fails within 5 minutes and I know it's not the memory as it's rated and tested at higher speeds.
8.5*450fps
Current settings which are failing, memory is only at 900.
Strap: 333
CPU Voltage: 1.375
CPU GTL Voltage Reference (0/2): Auto
CPU GTL Voltage Reference (1/3): Auto
CPU PLL Voltage: 1.56
FSB Termination Voltage: 1.26
DRAM Voltage: 2.10
NB Voltage: 1.30
NB GTL Reference: AUTO
SBridge Voltage: Auto
PCIE SATA Voltage: Auto
System..
P5Q-E
Q9550
4GB G-Skill F2 8500
GTX280
Corsair 750XT
X-Fi XtremeMusic
did you disable static readcontrol ? TRFC 55 ? which PL are you at ? Ram timings ? I would lower NB voltage to 1.26, Dimm voltage to 2.0 and try again... for me Bios 1406 was a requirement to get the E board flying.
Looking good Omiez but don't overdo it with the IBT, it stresses really too hard... I like it for a quick test ( 15-20 runs) but nothing more... for the rams and Nb I use 32Mb Hyperpi ( with an instance for each core )
Yes, don't know what TRFC is, PL? 1.56? 5.5.5.15, memory is tried and tested at 540 on a P35 so changing DRAM votlage seems a bit pointless, theres nothing wrong with the memory.Quote:
Originally Posted by Leeghoofd
Don't see how changing from 1402 to 1406 will make any difference but I guess I could try that.
The Asus board are known to over volt some which is why Leeg suggested you try 2.0v for the ram. 1406 has been the fastest most stable Bios for quite a few of us.Quote:
Originally Posted by Shocky
Larry
Thats not the problem, if it was id have the same issue at 3.6GHz (425*8.5) with the memory at 510(1020) but I don't.Quote:
Originally Posted by Big Lar
Anyway, tried 1406, makes no difference, slightly annoyed as my old P35 did the same settings, this motherboard is no better for overclocking at all so far.
Try these mate :Quote:
Originally Posted by Shocky
And coming from a solid performing P35 platform P45 is not a real upgrade unless you are running crossfire or a dual core for high FSB action...PHP Code:JumperFree Configuration Settings
AI Overclock tuner: MANUAL
CPU Ratio Setting: 8.5
FSB Strap to North Bridge: AUTO
FSB Frequency: 450
PCI-E Frequency: 101
DRAM Frequency: 900
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
1st Information :
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: 55
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: AUTO
AI Transaction Booster: AUTO
CPU Voltage: 1.35
CPU GTL Voltage Reference (0/2): AUTO
CPU GTL Voltage Reference (1/3): AUTO
CPU PLL Voltage: 1.56
FSB Termination Voltage: 1.26
DRAM Voltage: 2.0
NB Voltage: 1.26
NB GTL Reference: AUTO
SBridge Voltage: AUTO
PCIE SATA Voltage: AUTO
Load Line Calibration: ENABLED
CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
CPU Clock Skew : AUTO
NB Clock Skew : AUTO
CME : Compatible or Performance
Advance CPU Settings
CPU Ratio Setting: Manual
CPU VID: Default
C1E Suppport: Disabled
Max CPUID Value Limit: Disabled
Intel® Virtualization Tech: Disabled
Vanderpool Technology: Disabled
CPU TM Function: Disabled
Execute Disable Bit: Disabled
Nope, OCCT crashed then system blue screened.Quote:
Originally Posted by Leeghoofd
did you run Prime small FFT's ? Did you check each ram individually with HCI memtest ? Are you really on 1406 ? as running a previous version gives erratic behaviour with LLC enabled... Tried more Vcore eg 1.42 and LLC disabled ?
First of all yes, ran that for 3 hours with no issues at 3.8, it's mix and ram test that fail instantly.Quote:
Originally Posted by Leeghoofd
Again, it's not the memory for reason posted here , yes i'm really on 1406.
LLC?
Ok, i use to test IBT with maximum 15 runs to see if it is Rock Solid.Quote:
Originally Posted by Leeghoofd
Can you give me Link to Hyperpi Homepage and a howto use ?
TNX a lot.
There is a new Intel Matrix Driver out for those like myself having problems with the latest version...
http://downloadcenter.intel.com/filt...s=45&submit=Go!
Load Line Calibration = LLC, 3 hour small FFT is NOT sufficient to be absolutely sure the CPU is stableQuote:
Originally Posted by Shocky
There must be some ram setting or NB setting not 100%. Can you try one dimm at a time ( so remove one plz )
Here you : HyperpiQuote:
Originally Posted by omiez
Just select the 32Mb test and 2 cores... if the ram survives that you are wel on ya way
Quote:
Originally Posted by Leeghoofd
ahh yes, that is enabled, I know that but it's enough to know that crashing within one minute every time I run the ram and mix test is not the CPU.
Tried with a single stick, same result.
Have you tried another testprogram then OCCT ? Have you tried creating a memtest Cd and test everything outside windows, stress on test 7 and later on 5... dropped CPU multi to 8x ? That the app crashes okay but that it BSOD's is totally out of bounce...
I agree, i'm not sure what to make of it, the memory does pass memtest with no problems at all, it's the exact same limitaton as I had with the P35 P5K, 425FSB is fine, but 450FSB boots but unstable and in both cases the memory will seperately do 530+ without any issues. :brick:Quote:
Originally Posted by Leeghoofd
I guess I could just settle for 3.6GHz, really don't want to though, 425FSB seems prety poor for the P5Q. :(