So excuse me cus' I messed an ep : is the 1804 available for duals or what ?
Any answers for tha ram frequency Cryptic (all my modules are right there \o/)
http://soj.mesdiscussions.net/soj/Me...jet_2355_1.htm
Printable View
So excuse me cus' I messed an ep : is the 1804 available for duals or what ?
Any answers for tha ram frequency Cryptic (all my modules are right there \o/)
http://soj.mesdiscussions.net/soj/Me...jet_2355_1.htm
What is your vcore and why it is not in cpu-z ;> ?Quote:
Originally Posted by ViViD
AI Overclock tuner: ManualQuote:
Originally Posted by Kayso
CPU Ratio Setting: 8.0
FSB Frequency: 480
PCI-E Frequency: Auto
FSB Strap to North Bridge: Auto
DRAM Frequency: DDR2-971 MHz
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: 5-5-5-15
CPU Voltage: 1.56
CPU PLL Voltage: 1.52
FSB Termination Voltage: 1.4
NB Voltage: 1.5
DRAM Voltage: Auto
SBridge Voltage: Auto
All Other EveryThing: Auto
:confused: I don't Know Maybe It's Bug On CPUZ:confused:Quote:
Originally Posted by seban
Gonna To 1.625 CPU Voltage Tommorow And Over 500FSB:shocked:
This is hardcore :D. You dont like your CPU :DQuote:
Originally Posted by ViViD
hey seban have you got your board back from rma yet?
No I Like It But I Like OC More Than :up::DQuote:
Originally Posted by seban
Hi m8,Quote:
Originally Posted by radaja
Sorry to hear about your new/old mobo.
Probably I was not clear enough about my situation. I bought a mobo and because of problems with sensor during first 7 days I just informed shop that I am returning it (during first 7 days when buying from e-tailer I can just send back item without giving any reason and get my cash back - but probably you have same rules in US?).
But shop I was buying from has this additional rule that after they accept my request to return goods I have 28 days for item to be in their mailbox. So altogether I had it for about 30days to do some testing with it and make sure that I can not help anything about damaged sensor.
Despite the fact I had problems with MB sensor which yet alone was more then enough to send it back - the mobo was used before. All plastics were crushed and so on.
Any way to make sure I get my money back and wont be left with 2 MF2 in the hand I packed the old Mobo which came with used parts with brand new parts and plastics from new one (I am currently using) from other etailer... so as you can see it is quit long story (I am waiting for my money at this stage).
So answering your question I did not RMA it to ASUS so I want be able to share experience.
In the end I must say I like the mobo very much but I am still working on it. It's stability during OC is much better then DFI board I had before (which was NF4 chipset :D). Alto craftsmanship for this kind of price tag is just unbelievable joke.
It is hard to believe that Asus can consistantly produce this level of quality,Quote:
Originally Posted by seban
Someone needs to be fired there not doing there job, QA is a very vital part and someone has dropped the ball or simply doesnt give a damn
heres some pictures of the board they sent.clearly this is a used board.http://i41.tinypic.com/xkvonn.jpghttp://i39.tinypic.com/280mkj5.jpghttp://i40.tinypic.com/24177ut.jpghttp://i41.tinypic.com/sw8sn4.jpghttp://i41.tinypic.com/2upwbvd.jpg
That is indeed badly warped MF2... are you planing to send it back again?Quote:
Originally Posted by radaja
------------------------------------------------
About my problem with choppines when moving windows in Vista :) - It looks like the problem is to tight memory timings, but it is not fully confirmed yet (thanks andressergio and Grnfinger).
I agree wit you Grnfinger : wtf with this QA the board lost 100€ on it or what ?? Angry ?? the word would be too small !!
One more time asus lost a costumer : I can't trust anymore when I see what's happenin' on these threads mate :/
Sorry but could someone reply to what I asked please ? Does this bios increase memory BW with C2D or not ?
Hello guys,
I have a question about GTL reference voltage on this board. What does the auto setting do? Is it +0mV? If not, what is it? How can I find out what voltage auto is setting? And if it's not, how can I set +0mV?
Thank you very much in advance
Musho
the auto setting does just that, it sets it to where it needs to be depending on where your overclock is. theres no way to tell except to experiment with it, as far as 1802 increasing bandwith with c2d, I doubt it it doesnt change bandwidth with c2Q so it prolly wont with duals. all it did for me was make boot a bit quicker, and raid works now.
There is no bandwidth increase for bios 1802, tref value is the same as 1307 so bandwidth is unchanged.Quote:
Originally Posted by Eeky NoX
Never had an issue with Raid0 on MIIF with any bios, clean installed OS 5 or 6 times now. What problem where you having with raid?Quote:
Originally Posted by philbrown23
Does any body know what is it and how to turn it off? I cant find any option for it - but maybe I am blind (wait time and esc key)
http://images39.fotosik.pl/41/4cff9799fa4a14f6.jpg
Under the Advanced tab scroll down to LCD Poster and LCD control
Enable the ROG Logo
Not sure if that will work , its a shot in the dark.
Try enabling 'QUICK BOOT' under the BIOS 'BOOT' Tab, it's a 'POST' message...
Hey guys,
Have any of you experienced any positive Vdroop with the 1802 bios? I haven't tried earlier bioses, since I recently bought this mobo and flashed it to 1802 right away. What I mean is, that when my CPU gets loaded with prime95, the Vcore increases according to Everest and CPU-Z. This could give me a hard time trying to get my CPU stable in idle too, since it will drop the Vcore down :(
This is with loadline calibration enable by the way. Without it enabled, getting my quad stable when fully loaded will be impossible except with crazy high idle voltages. I hope this is a bios issue and will be resolved, rather than an issue with my board :(
Thanks for the help guys, my quad's almost stable ;)
Musho
Got another MFII for a build here and it is behaving completely different to the first .....The CPU GTL's required to even boot the same Q9650 at 4.0Ghz are completely different:
Board 1: +10/-40/+10/-40, with NB GTL at +20
Board 2: +40/-10/+40/-10 .......NB GTL at Auto
It also requires slightly more VNB and VFSB.
Very odd.....both are A2's and running 1307 bios.
I have no problems with my m2f, except when I play certain games, it'll lock-up and sometimes the sound will stutter or loop. Troubleshooting this problem, I've backed-off my overclock from 4.5GHz to 3.6GHz, I've swapped my XtremGamer Soundcard for an ASUS Xonar 'Card, & discontinued using my UPS (overload alarm triggered). 'GRID' which locked-up before, now plays ok, but 'GTA San Andreas' locks-up constantly. All that seems left is to start bumping-up some voltages. I bumped the Vcore from 1.3v to 1.38v, and the NB from 1.2v to 1.25v, but GTA still locks-up. Manufacturers suggested voltage for this RAM is 2.1v--using BIOS 1307. Any suggestions or recommendations will be appreciated
Have you tested the ram with memtest?Quote:
Originally Posted by al-
If not test the dimms at stock speed/volts using memtest
I ran Memtest86+ for 2 passes without errors, and Orthos for an hour without any errors, but I read just because a pc is prime or orthos stable doesn't mean it is game stable....Quote:
Originally Posted by Grnfinger
I have been trying repeatedly but I cannot get my M2F to do 542 FSB / 650 MHz SPI32M stable. It locks up or errors out before it's even completed the initial value. My P5Q-Deluxe has no problem at all doing this. The P5Q-D also needs no ram channel delay to stay stable at for example 534 FSB / 640 MHz, whereas the M2F needs 150ps delay on the channels or it will hard lock with over 2.2 vdimm. What I find a bit odd is the fact the M2F is stable at 534 FSB / 650 MHz at reasonable voltages, then is completely unstable only a little higher regardless of the voltages I give it.
I don't know if it's bios differences that are responsible for the different behaviour between the 2 boards as hardware wise they are very similar, they both run the same vNB/vdimm/timings/PL etc up to 534 FSB / 640 MHz (except the M2F needing the ram delay) but after that the M2F stops and the P5Q-D keeps going. The M2F is however slightly faster than the P5Q-D clock for clock at any given setting.
al- i think if you are using the hd48xx card in your sig that you might be having problems with it. drivers looping like blink- black screen and sound continues to play. vpu revovery error that thousands of hd users are experiencing.. i have that problem with mine. i can pass every test for days and can't play games..
Hey guys yesterday i started testings this amazing mems on the MF2
Just booted them at 1,9v at 1200 PL10, they booted on less but i decided to give a try on this that seem so low volts, and put a little NB = 1,3, later i will be testing on lower PL that it'll need more NB and perhaps more vdimm
http://img210.imageshack.us/img210/4...6311yb3zz6.jpg
http://img248.imageshack.us/img248/3...i32mzp3at4.jpg
Quote:
Originally Posted by Grnfinger
Hi guys unfortunately both tips did not help with my strange boot issue. Logo option suggested by Grnfinger does light up the Logo on motherboard (we do not want that ;) ) and quick boot does not affect this strange thing with "esc" key.Quote:
Originally Posted by al-
I will try 1602 from other bios - but I remember I was able to turn it off somehow in 1602. Maybe I got bed flash of 1802.
That is very nice RAM you got there. Sergio can you post everest test for your settings.Quote:
Originally Posted by andressergio
Thanks
andressergio
Yeah, some everest please
No point in everest just yet till he gets the PL down to 8 and tightens up the chipset settings. He'll be pulling low to mid 9K with those settings at the moment.
Hey waterflex I'm getting my way to get PL8, I need more NB and maybe some more vdimm but impressed they passed that 4x SPI 32M on such low volts :DQuote:
Originally Posted by WaterFlex
Will post soon i'm enjoying them by now :D
Cheers
Sergio
Yes mate at PL10 it's no big deal, also a x9 multi for 4050 it has less bandwith to push...but very good mems as i told y ou they booted perfect on less than 1,9 wich impressed me, just a little bit of NB and i get them stable fast.Quote:
Originally Posted by CryptiK
Cheers
Sergio :up:
Kinda stuck here. so any help would be greatly appreciated. This is what I've got so far.
Processor: E8500 E0
Asus Maximus II Formula
2x (8gb) OCZ TECH OCZ Gold XTC 4096MB DDR2 PC6400 KIT (OCZ2G8004GK)
Velociraptor 300g + 2x Samsung SpinPoint 750gb raid0
Antec Sonata III 500w psu door off
XIGMATEK HDT-S1283
Roomtemp 23C
Ai Overclock Tuner: Manual
OC From CPU Level Up: Auto
Ratio CMOS Setting: 9.5x
FSB Frequency: 430 (bluescreens on 440)
CPU Clock Skew: Auto
NB Clock Skew: Auto
FSB Strap to North Bridge: 333
DRAM Frequency: DDR2-861MHz
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 6-6-6-18
DRAM Static Read Control: Disabled
Dram Read Training : Disabled
MEM OC Charger : Enabled
Ai Clock Twister : Auto
Transaction Booster : Manual
Common Performance Level [11] (Tried 10 but I got DET DRAM all the time if I went from 420 to 425fsb)
Pull-in of CHA PH1: Disabled
Pull-in of CHA PH2: Disabled
Pull-in of CHA PH3: Disabled
Pull-in of CHB PH1: Disabled
Pull-in of CHB PH2: Disabled
Pull-in of CHB PH3: Disabled
PCIE Frequency: 100
CPU Voltage: 1.35 (unstable at 1.32)
CPU PLL Voltage: 1.54
FSB Termination Voltage: 1.245
DRAM Voltage: 2.0
North Bridge Voltage: 1.35
South Bridge 1.5 Voltage: 1.54
South Bridge 1.1 Voltage: 1.113
CPU GTL Reference (0): Auto
CPU GTL Reference (1): -30mv (probably wrong, just copied something)
CPU GTL Reference (2): Auto
CPU GTL Reference (3): +30mv
NB GTL Reference: Auto
DDR2 ChA Reference Voltage: Auto
DDR2 ChB Reference Voltage: Auto
North Bridge DDR Reference: Auto
CPU Configuration:
Ratio CMOS Setting: 9.5
C1E Support: Disabled
Max CPUID Value Limit: Disabled
Intel Virtualization Tech: Disabled
CPU TM Function: Disabled
Execute Disable Bit: Disabled
Load-Line Calibration: Enabled
CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled
Increase the vNB ;) mwarf I guess you're so far from my settings wit this shubby 1,35v :D
Hi All Friends
http://valid.canardpc.com/cache/screenshot/465508.png
Valid Link
With 1.65 Volt On CPU
Quote:
Originally Posted by loften
Thanks, I'll take a look at that--
BTW is there a way to make Prime stop all cores when any of cores fails?
loften, I looked (googled) at numerous posts and although the 'vpu recovery error' has definite similarities to my video issue(s), it does not seem to quite fit, and it looks like the 4870 card has 'crash' problems, with different sets of symptoms, that are resolved in different ways. And also, just because one persons' 4870 has the same symptoms as anothers' 4870, doesn't mean that it can be 'fixed' the same way. I think, after trying many things, that I may have it resolved...I deleted everything ATI/CCC, booted into 'Safe-Mode', used 'Driver Sweeper', and re-installed CCC Suite v8.12 from the desktop (along with bumping-up my NB and Vcore). I noticed that prior to using 'Driver Sweeper', my default clock/memory and gpu clock/memory (GPU-Z) was 500/900:750/900, but now its 750/900:750/900, and GTA runs smooth as silk, not choppy like it was before...maybe ATI will pick-up the ball and straighten-out all the problems users are having with HD48XX's-I seriously thought this was just an issue for the maximus2formula bios (voltages), but it looks like it was 'drivers' too (keeping my fingers crossed)...Quote:
Originally Posted by loften
BTW maybe somebody in the future will find it useful - my problems with random choppiness when moving windows around the desktop was resolved by adjusting NBGTL in my case to +60.
Also the same option gave me stability on my Dominators at 1130 5-5-5-15.
And just upping vNB did not helped even to vhigh values.
Nice work, whats the max FSB you can hit on that chip?Quote:
Originally Posted by ViViD
Yes I used to experience the same thing back in the day on my P965 system if i didnt have enough vNB. Weird that your board like positive GTL Ref on the NB, my board hates anything other than auto (0.64x), if I'm pushing the board a little, say 550FSB, setting -5mv or more gives me no boot, and +10mv or more gives me no boot.Quote:
Originally Posted by seban
Good to hear you sorted it out.
I think every board is different, as evidenced by my earlier post here :
http://www.xtremesystems.org/forums/...postcount=2421
The same exact components were used across the two MFII's for testing.
hmm whats the point of doing so high mhz on the ram when ure taking a hit on the performance levels?? willing to bet 1:1 with lower tRD will preform better with lower latenciesQuote:
Originally Posted by andressergio
500fsb 1200mhz CL5 - tRD 8 for 4x1gb
400fsb 1200mhz CL5 - tRD 5 for 4x1gb
Yes my mobo really like high volts. Actually I have a problem with this, because I am ATM stabilizing it at 471 fsb and it needs more than unfamous 1.4v value which I do not want to cross for 24/7.Quote:
Originally Posted by CryptiK
Are there any other ways to lower hunger for Vtt than tuning CPU GTLref? Does increasing vcore might help in lowering vtt?
I need more vtt power but cant go higher with volts on it :D. Does actually anybody run here >1.4v Vtt for 24/7?
Cheers
I'm willing to bet otherwise :) Aside from nVidia chipsets I've never seen where 1:1 did better than a divider with Core 2...Quote:
Originally Posted by cstkl1
Once I got home I will post screenies where 942 1:1 on loose timings does marginal better than 1130 on 5:4.Quote:
Originally Posted by Shiranui Gen-An
It might be coincidence tho.
mate i put this ram 2 days ago...I just started testing them, i posted that as pointless...I know what you meant of course i will find a better bandwith for 24/7...Quote:
Originally Posted by cstkl1
Cheers
Sergio
what do you think this?i was packing the board up to send it and just noticed this under the light.i cant believe they can send this stuff out.http://i42.tinypic.com/24omwe1.jpghttp://i40.tinypic.com/2l896qg.jpg
Hard to tell from teh pics. looks like some sort of insulating spray they use for preventing moisture when cooling with dice but not 100% sure.Quote:
Originally Posted by radaja
How are you making out, did the supervisor call you? I will know today what what mobo I'm getting as a replacement. I tried to squeeze a P6T Deluxe out of them but I was denied:D
i know its hard to tell from the picture but it definitely looks like maybe some spilled (maybe coolant from a WC setup).
Good luck getting a "upgrade" guys.... They beleive in giving you what you paid for, nothing more, EVER......:shakes:
all i want is a new unwarped,no spilled stuff, fully functioning maximusII formula. so i can enjoy it as much as every one else does.
Well then you'd be wrong, got my email and I am offerd.....Quote:
Originally Posted by screwtech02
1.Maximus II Formula
2.Striker II Formula
3.Rampage Formula
4.P5Q Premium
Oh and I'm getting a token Asus T-Shirt as a sorry for the hassles.
So as you can clearly see (altho it took 4ever) the squeaky wheel gets the greese.
thats great news!so which one are you going to pick?if it was me i would go #3
:up:Quote:
Originally Posted by radaja
I wish I had of made that choice. Don't get me wrong, this board has a lot to offer if Asus would only provide us with a Bios worthy of the board's price tag.
I have seen that stuff on a couple of brand new boards. I'm not sure what is it but I don' think it's insulating spray or coolant residue, I think it's something during the manufacturing process. I could be wrong though of course.Quote:
Originally Posted by radaja
That's good to hear, I hope whatever you choose is a solid board. You have certainly been stuffed around.Quote:
Originally Posted by Grnfinger
cryptik, its definitely something spilled,after closer inspection of the top side you can see this stuff leaked down through the gaps of the memory slots.
Wow, i am simply amazed that they are doing that for you.... I tried to get a 975X board after RMA'ing my P5N32-SLI 7 times, and was told "we dont offer upgrades"... Good for you man, what ya gonna get?? :cool:Quote:
Originally Posted by Grnfinger
That's really bad. I've seen light residue before, but if it's leaked down through the ram slots your lucky the board didn't fry you gear.Quote:
Originally Posted by radaja
How could they send you something like that? I would have though they would not even accept something in that condition back let alone send it out.
hi guys this is my result in testing max fsb on my mobo
http://valid.canardpc.com/show_oc.php?id=466061
setup:
cpu@phase ss 1.2hp
nb@liquid cooling ybris k5
here some photo of my rig
http://i527.photobucket.com/albums/c...h_IMG_2149.jpghttp://i527.photobucket.com/albums/c...h_IMG_2152.jpg
http://i527.photobucket.com/albums/c...h_IMG_2152.jpghttp://i527.photobucket.com/albums/c...h_IMG_2152.jpg
http://i527.photobucket.com/albums/c...h_IMG_2159.jpghttp://i527.photobucket.com/albums/c...h_IMG_2168.jpg
http://i527.photobucket.com/albums/c...h_IMG_2169.jpghttp://i527.photobucket.com/albums/c...h_IMG_2171.jpg
here more photo of my setup
I did not use SuperPi before on this rig.
I am getting thus each time at loop 5 - on stock settings and also on 21h P95 blend and 10 IBT stable.
Any body can tell me about this
http://images50.fotosik.pl/41/9e080f4378783940.jpg
didn't want to hijack but yeah keep a eye on it.. i thought it was suspisious as it sounded like the problems i was having with mine.. i have the older maximus and 3870 cards and for months i wondered why my board wasn't stable at any oc or even stock. first raid many hard drives later i put in a single without raid which allowed me to stay stable long enough to start testing things, then ram, then psu, after many days/ months of testing things i finally figured out that i was having vpu errors as the older drivers i was using wouldn't show vpu errors it would just bsod only gaming. the later drivers started showing the vpu errors and also the sound looped and i could see the errors instead of bsod.. i removed the card and borrowed a 8800gts from my bro to test and it wouldn't crash anymore. passed every test no problems at 3.6+ and played games without crashing.. i am still waiting on the ati drivers and or bios that will work with my 2 cards.. i have tried them all even modded ones, different power ram and oses.. still waiting for my fix. have to save up and get a different card to replace these 2 i guess.Quote:
Originally Posted by al-
do they use flux for soldering? could be dried up or burnt on flux..
Nice work! What was the vNB and any other critical settings for that?Quote:
Originally Posted by Beleno
Probably due to SPI not liking vista, try hyperpi and check the 'audio, solve incompatibility with vista' checkbox.Quote:
Originally Posted by seban
@cryptik, i never even tried this board.after i opened it and saw that it was warped and clearly used, i just started making calls to asus to rectify the problem.
Right click select properties and run in XP compat and admin mode.Quote:
Originally Posted by seban
hi guys n gals.
my question isnt for the ocing on this board, i have an ample oc from 2.4 to 3.6 on my q6600(air) and am slowly working on the memory as we speak. my problem is sound, where do we get drivers, the ones asus provide say my hardware isnt supported by the software and ends the install, i tried creative but cant find the card and googling doesnt seem to get me anywhere, just wondering if anyone has the same problems and may know of a solution...thanks in advance
peace
fatman
I'm almost there guys! Unfortunally, an C1 stepping plus a fairly high VID isn't a very good combination...:mad::down:The fact I'm all on air doesn't help either :(
I'll get it stable, though! Wish me luck ^^:D
http://i723.photobucket.com/albums/w...az/bad_CPU.jpg
Guys I need help!
Can anyone give me a reason NOT to get the Striker II Formula?
I havent had a NForce chipset since my A8N SLI Deluxe/FX-60 and would like to give it a go, unless there are issues with the board otherwise I'm getting a Rampage.
Anyone know of issue wich would make this a no go???
Would either run a Q6700 or a E8400, both are collecting dust on my desk.
Thanks
Is anyone else experiencing higher load Vcore than idle Vcore when Load Line Calibration is enabled? Or is this an specific issue with my board? :shrug:
No it's a common thing, mine will overvolt as wellQuote:
Originally Posted by Musho
Okay, I couldn't get my CPU stable at the previous Vcore, so I decided to up the Vcore one more notch :p: It looks like I've got a new 24/7 setup! I'll switch to Watercooling in a few days, to make sure my chip won't die anytime soon :rolleyes::D
:up::yepp::clap:
http://i723.photobucket.com/albums/w...o-az/notch.jpg
It is hyperPi and I cant run it in compatibility mode - it will not start at all. As admin did not help.Quote:
Originally Posted by Grnfinger
It is hyper PI I am affriad and option for sound is checked.Quote:
Originally Posted by CryptiK
But basically if PI does not pass test the error message does not look like this?
Let it run for 12h Blend and for real stability 24h :) (and it still can be unstable in 3d)Quote:
Originally Posted by Musho
I'll let it run blend tomorrow. It's still priming away small FFTs at the moment, and going strong ;)Quote:
Originally Posted by seban
So I run 4 instances of Super PI and they passed no problem - so It looks like HyperPI does not want to cooperate with my sys - which is kind of crap as it is lot of hassle to run 4 instances and set affinities for each.
3h of small FTT should be more then enough. Go for blend it will tell you is your vtt and vnb stable.Quote:
Originally Posted by Musho
This is decryption of errors - which I can confirm is working for me like a charm.
1024 - increase VTT or vNB
896 or 448 - vNB
768, 512, 332 - VTT
Thanks, that sounds really helpfull to try and pin down any stability issues that might arrive when I try blend. Is it possible to tell when the ram is the link causing the problems, though? Or should I run memtest first? I'm pushing my ram quite a bit too, it's running at 1140mhz, at PL8 at stronger, with static read control enabled ;)Quote:
Originally Posted by seban
Anyway, here's the result of the small FFT test :yepp: I'm off to try the stability with some games, as I'm dieing to play now :D I'll run blend soon. I'll keep you guys updated.
http://i723.photobucket.com/albums/w...-az/stable.jpg
need help OCing my new rig (Q9550)
got an E0 Q9550 from ebuyer, and a maximus II formula, and watercooeld it.
under OCCT my max FSB i can get to pass is 467, but game crash within seconds....
these are my voltages:
http://i277.photobucket.com/albums/k...Capture-18.png
(btw temps under laod never go above 50 on any core)
I wouldn't bother with Striker. Rampage is a better option.Quote:
Originally Posted by Grnfinger
Get those voltages off of auto! The maximum safe CPU PLL voltage is considered 1.6V. 1.7V is already quite extreme. Yours is at almost 1,8V :eek: Also, I wouldn't go over 1,45 VTT. You're getting close to this value, as you're already at 1,44v. I am using just 1,3v at 8,5x475fsb. Put your CPU and NB clock skews on normal (not auto!) and test if you gain any stability. Also try out (100ps delay on CPU with normal on NB) and (normal on CPU with 100ps delay on NB). Out of these 3 possible combinations, pick the most stable. Set your VTT to 1,4ish. You can back this voltage off later. vNB at 1,45 is indeed a good start, but you should try to get that one down later one, once you get your setup stable. Now you should SLOWLY increase your FSB. You shouldn't have to increase any voltages to get it stable. (you might need up to 1,45v Vcore to get 4ghz stable, though)Quote:
Originally Posted by shaffaaf27
Set your NB GTL reference voltage on auto and set CPU GTL reference voltage on auto on all 4 options. While you're working up on the FSB you'll notice cores failing in prime95. Take note of which cores are failing. If core 0 or 1 fail, you'll have to play with CPU GTL reference 0 and CPU GTL reference 2. If core 2 or 3 fail, you'll have to play with CPU GTL reference 1 and 3. Note: GTL reference numbers do NOT represent cores. 2 cores on the same die use the same GTL voltage. Both dies have 2 different GTL voltages. One for the address bus, and one for the data bus. For now, always keep GTL 0 and 2 the same. And keep GTL 1 and 3 the same value too. So you'll have to figure out 2 values. 1 value for both GTL 0 and 2. And one value for both GTL 1 and 3. You can finetune them seperately later. Getting cores stable with GTL is purely trial and error. Every single CPU responds differently to different GTLs, and even different boards have different amounts of vDrop on the GTL lanes. Just trying anything ranging from -70 to +70 till you get all cores stable again, and start moving up again on the FSB, till cores start to fail again. Then you have to tune GTLs again, etc etc
Edit: Use small FFTs to test GTL reference stability in prime95.
When to adjust CPU PLL and what is ts influence on other voltages if there is any?
If ram passes 3+ of MemTest 86 2.1 then it should not be a problem.Quote:
Originally Posted by Musho
My CPU was stable small FTP at 4ghz at 1.3 but was failing Blend instantly :). Anyway u will get there sooner or later.
How many reasons do you want? Its nvidia chipset,that alone is enough.Quote:
Originally Posted by Grnfinger
It's an older outdated chipset to my knowledge, a 790i would be a better choice if you wanted an nvidia chipset as from what I have read it appears to do good FSB and has digital PWM. A Rampage Formula would IMO (based on what I have read) be the best choice from that list of possibilities you have. Have a read through Striker II threads and RF threads and just see what appeals to you most. I've never used either board so I can't give you any direct reasons to go with one over the other.Quote:
Originally Posted by Grnfinger
Yes, as Grnfinger said this is normal behaviour for my board too.Quote:
Originally Posted by Musho
I have never seen that error message come up, I usually get a hard lock or 'not exact in round' error, but that's not to say it's not a normal message, I just have not personally seen it before. If superpi runs and hyperpi doesn't just go with what works, although somewhat odd - I use XP but I've read in multiple places that SPI doesnt often play nice with vista and hyperpi (and checking the sound box) was the solution to that, your situation is the opposite.Quote:
Originally Posted by seban
[QUOTE=CryptiK;3515451I have never seen that error message come up, I usually get a hard lock or 'not exact in round' error, but that's not to say it's not a normal message, I just have not personally seen it before. If superpi runs and hyperpi doesn't just go with what works, although somewhat odd - I use XP but I've read in multiple places that SPI doesnt often play nice with vista and hyperpi (and checking the sound box) was the solution to that, your situation is the opposite.[/QUOTE]
Hey CryptiK :up:
Hyperpi 0.99 is the one that doesn't start at all...0.98 runs fine on VISTA 64 disabling audio...I tried all to make 0.99 but doesn't
Cheers
Sergio
Hey guys
I'm running this mems fine on 1,9vdimm and 1200Mhz but PL10, i was trying to get them lower but i'm getting all kinds of bios blocks...No post and most times i have to turn of the PSU switch and back...
I tried more NB (till 1,45-1,5) and more vdimm till 2,05 but seems so unstable...
Anyone here that's using that volts on the NB ? or experiencing this ?
I'm lowering Mhz now to 1125 on PL8 and seems much better, will keep testing but i'm kinda shocked as more NB and vdimm seems to be worst...this gskill boot at 1,8 on 1200 on PL10, not stable tough
I tried NB GTL refs and skews but no luck just keep that bios block
Cheers any help i appreciatte
Muchas Felicidades Para Todos en estas Fiestas :up: (Merry Xmas to all !!)
Sergio
Hey thanks for the tip will check 0.98 !Quote:
Originally Posted by andressergio
Mary Xmass 2 u2 ;)
Damn, very low Vcore you used there! Give me an E0 :p:Quote:
Originally Posted by seban
I played GTA IV for 30 minutes and then it BSOD'd. I tried blend and it failed instantly :D Turned out 1,32 vNB wasn't quite up the job at running a quad at 475 FSB, with all 4 ram slots occupied, running at 1140mhz, PL8, stronger, static read control enabled. 1,41 vNB is looking good so far, although NB temps are reaching over 50 degrees on stock cooling now. I guess I'll just order Swiftechs NB-max waterblock too, together with my CPU cooling parts.
Thanks for your help mate :up:
Quote:
Originally Posted by JoeBar
Quote:
Originally Posted by N3it1
Quote:
Originally Posted by CryptiK
Thanks for teh input:up:, after spending the day researching Striker II I agree the Rampage is a solid choice.....
It will work mate :up:Quote:
Originally Posted by seban
Saludos !
Sergio
It seems many of you guys have been claiming perfect stability with Prime95 Small FFT with the Q9650 however I don't see anyone here noticing that this stability test is pale in the wake of Linx/IntelBurnTest (same thing - Linpack libraries, different GUI). Try running the latter for say 50 loops and I bet some of these overclocks will fall over with either an error or BSOD.
The MF2 is a very picky board to setup at 450MHz FSB and 1200MHz DDR2 I find, especially as I don't consider the PC stable unless it passes Linx. Getting it to pass others (i.e. such as Prime95 FFT) is easy in comparison.
Would be nice to see others confirm this as right now I'm thinking the P45 chipset timings/clock phases must be to blame.
BIOS 1802 here.
50 loops of IBT is crazy. Your only going to kill hardware doing that.
10 loops 20 tops, then a nice prime run will more than suffice anyone.
BTW I have passed 15 loops of IBT and failed Prime in under a hour so to say IBT is the great and all powerfull stress test is rather bold.
PPl's version of stable vary, 10 loops IBT and 8-12 hours Prime and its rock solid imo
This board is easy once you learn how YOUR hardware behaves with the bios tweaks
http://i237.photobucket.com/albums/f...r/MII_4GHZ.jpg
http://i237.photobucket.com/albums/f...table_1130.jpg
Hi My Friends
http://valid.canardpc.com/cache/screenshot/466748.png
Valid Link
But Dont Try To Find Max FSB It's Hard To OC Quad On This Board More Than AMD :D
How much vcore that needed ?Quote:
Originally Posted by ViViD
Great clock :up:
QFT! :up:Quote:
Originally Posted by Grnfinger
When you get your Rampage we will surely miss your input and help here my friend. :clap: :clap:
Oh I'm still keeping the MIIF, it will be nice to finally compare the 2 chipsets myself.Quote:
Originally Posted by SiGfever
You wont get rid of me that easy:D
:ROTF: :ROTF: :ROTF: :ROTF:Quote:
Originally Posted by Grnfinger
That's good to hear... :)Quote:
Originally Posted by Grnfinger
+1 :)Quote:
Originally Posted by JoeBar
Can any body give me a quick tour on how and when to adjust CPU PLL? are there any symptoms that tell when to do it? Or can it help in not increasing vtt ? :D
PS Probably as most of the time I will discover it myself but maybe some body will point me right direction?
PS2. Christmas is almost here and I am still 4ghz not stable.
Grnfinger, very nice 06 score there, those 3870's in crossfire pack a punch.
***********************
I managed to do a couple more runs, and got a bit more efficiency out of it.
SPI32M @ 500 FSB, 600 MHz, PL8
http://i266.photobucket.com/albums/i...mLSCenable.jpg
Everest @ 500 FSB, 600 MHz, PL8 (same settings as SPI32M run)
http://i266.photobucket.com/albums/i...mmLSCenabl.jpg
SPI32M @ 534 FSB, 640 MHz, PL8
http://i266.photobucket.com/albums/i...mLSCenable.jpg
Everest @ 534 FSB, 640 MHz, PL8 (same settings as SPI32M run)
http://i266.photobucket.com/albums/i...mmLSCenabl.jpg
Very impressive scores CryptiK :up:Quote:
Originally Posted by CryptiK
Use 1.645V On Cpu Voltage It's Too :shocked: :DQuote:
Originally Posted by carepolice
Tanx M8
jeje nice results my friend !!! :up:Quote:
Originally Posted by CryptiK
Saludos Amigo !
Sergio