Hi
You don't need to use modded driver AND modded HAL. One or the other (when used inconjunction with a registry DST string inserter or modded BIOS).
Andy
Printable View
Hi
You don't need to use modded driver AND modded HAL. One or the other (when used inconjunction with a registry DST string inserter or modded BIOS).
Andy
here my best results...
560 MHz Primestable FSB @ vNB 1,44V
http://www.abload.de/thumb/560fsb_primestable75wg.jpg
540 MHz Primestable FSB @ vNB 1,30V
http://www.abload.de/thumb/540fsb_primestablewzyn.jpg
Hi, I'm posting here because you guys are clearly pc experts : - )
I have broken my PS2 keyboard port on my MIIF whilst changing keyboards, the port has recessed into the metal housing and I now get "Keyboard Interface Error" on bootup. My machine seems to be runniing fine with a usb keyboard but I really need some reassurance that it will not cause a short in the future and kill my components.
I cannot afford a new MB at the moment and good replacement Core 2 MB's are becoming hard to find in stock, now.
So what do you experts say, is it nothing to worry about ?
Many thanks,
boo
The fact that the board has detected the issue isn't a bad thing, and you could assume that the circuitry is probably quite isolated from a lot of the more important parts of the board, but the board has detected a short in the socket. It might have shut the socket off but without a voltmeter it would be hard to tell.
If it was me, I would probably try and pull the socket out again until it didn't give the error (or if possible, un-solder it completely), shorts aren't a good thing over long-term because things can heat up or burn. Keep it running, and take a look around the area of the socket. Check for burns, strange smells, excessive heat. If the board starts to become unstable, I would suggest keeping it turned off until you remove the socket entirely.
Thankyou, great advice, very much appreciated!
My brother used a screw to grasp the connector and pull it back in to place as far as possible.
No error messages at bootup now, the PS2 port is ruined but that is very cheap compared to a solder repair job or a new motherboard.
Hopefully job done : )
Thankyou!
boo
I bet Asus would replace that board for you just by telling them you keep getting a keyboard interface post error on bootup. ROG products get the best replacement service from Asus, at least thats my experience with them. I recently had to replace my MIIF and when I got the replacement I discovered my troubleshooting was in error and nothing was wrong with the old board. Asus did not care that I was wrong and only wanted to make sure everything was working properly even though I was in error and not the board. Not sure if you would get the same good service from them though. But it is my bet that you can live without the PS2 port.
Hi, thankyou for your reply, I did think about contacting Asus support but the whole process seems like so much hassle to me, unfortunately that's how my mind works, I would most likely just buy a new one (if I had the funds).
Yes I agree no working PS2 port should not be a problem, what a bummer of a thing to happen tho, I imagine everyone breaks their computer or other hardware at some point lol : -)
Many thanks.
I RMA'd mine around 2 months ago - I took care to pack all included items in the box like it was new (I even put the plastic stickies back on the mosfet sinks). 2 weeks later I got a crushed, taped up MIIF box with just a motherboard and a severely scrunched up manual! No sound card, no i/o backplate, cables, CD or LCD poster. I sent it back right away, and after 2 more weeks I finally got a new board. I was astounded!Quote:
Originally Posted by SkOrPn
It clearly states in the RMA procedure to ONLY send the mobo back and nothing more. you are not suppose to include your original box or accessories, just the mobo and a standard ups, usps or fedex box. Just take care in packing it so it arrives un-damaged. Asus does not send new retail items back to its customers, so you actually got lucky. Below is the RMA email you get when dealing with their RMA service..Quote:
Originally Posted by In-Fluence
Quote:
**Please do not respond to this email. For any inquiries, please contact the Customer Service Department at http://livesupport.asus.com or 812-282-2787 Option 2**
Dear Rod,
Your RMA# is xxxxxxxxx
Please complete the attached forms, sign, and fax back to (510) 608-4511 or email xship_rma@asus.com. Please allow up to 24 business hours for processing and before ship out.
**IMPORTANT:
- Any incomplete forms will cause a delay in processing time and All fields with a “*” must be completed.
- Serial # should be taken from actual product.
o Located: top of purple parallel port, actual side of PCI / Power Connector, underneath side of product. Format will be 10 or 12 digits and will start with a number, excluding 0 and 3rd digit will be M or Z
o If any discrepancies please contact Customer Service Team immediately by phone
o If s/n# is altered or missing on physical product, ASUS manufacturer’s warranty is void. Please see http://support.asus.com for more detail on your warranty.
- Part# must be listed on forms. Part # will be located near the edge of the board on a white / green sticker approximately 20 digits alpha-numeric including dashes. Example: C56B123-02347-60-MBL100-A02 or 90-MBL100-G0UAYZ
- Revision # is located near PCI slots in small white print next to model name imprinted on physical product
- Defective products must be received within 14 days from received date; otherwise credit card will automatically be charged and no refunds will be given.
You must write your RMA# on the outside of package. (If you do not write the RMA# on the outside of your package product will be rejected at the time of delivery).
Please do not include any accessories such as manuals, software, CPU, memory, retention mechanisms or clips, cables, Non-ASUS Products or your original box. If you send any of the above mentioned items they will not be returned or replaced.
Physical damage is not covered under your ASUS manufacturers warranty and automatically VOIDS your warranty. ASUS reserves the right to reject any product received with physical damage.
- Damage can include any shape or form of corner / edge damage / markings with pens and/or markers
- Broken traces or burns due to improper installation / removal of other components
- Broken CPU socket or bent pins
ASUS - USA is a warranty repair service center. Please contact place of purchase for credit, refunds, upgrades / model changes, or new retail box replacement. ASUS does not provide these services under any circumstances.
RMA Return Shipping Address:
ASUS Computers International
Attn: Xship Dept, RMA #xxxxxxxxxxx
800 Corporate Way
Fremont, CA 94539
Thank you,
ASUS Customer Service Team
812-282-2787 or 510-739-3777
Customer Service: Option 2
Technical Support: Option 3
My abit ix38-quadgt died and i change it with this one. I'd like to ask if there is a way to change the dram vtt, because i can't find it but pc probe 2 monitors it and it's giving different values depending on my overclock, so I assume there is somewhere an option and i left it on auto. Or does it change by itself when i change the ddr voltage? What's the maximum safe value for everyday use for dram vtt?
I went via the retailer and followed their instructions. They were annoyed with Asus though.Quote:
Originally Posted by SkOrPn
Could you post your settings for these please?? :)Quote:
Originally Posted by GaBBa-Gandalf
I have an E8400 paired with M2F
Previously I had to rma my board as I saw that the NB was at 80~90'C ish even when at stock 3.00ghz
Even after a washer mod.......still getting the same temperature reading
Well now I just got the board back form RMA
But I'm still getting the same temperature reading on my NB
[S]I'm currently running 3.00ghz @
vcore=1.1v
vnb=1.1
vtt=1.1
vpll=1.5
vdimm=1.9v[/S]
Anybody care to help? Will post a SS soon enough
:UPDATES:
The board has been behaving these past few days
Temperatures drop.....but lingers around 70-80ish
Uploaded a screen shot too
http://img.techpowerup.org/091101/M2F-E84-3GHZ-1066.jpg
dro, you should replace the stock thermal paste if you haven't done so yet.
time to update my status, this IS on air cooling, wish my board was better cause my chip has room to go
currently at 3.9ghz (8*486)
vcore 1.393v in bios 1.31v under orthos load
vNB 1.55v
VTT 1.40v
PLL 1.55v
pic
http://i37.photobucket.com/albums/e9...thos39ghz2.jpg
Yes Dro, definitely take off that heatsink assembly and give it a clean with some alcohol or dedicated TIM cleaner. Lay the assembly flat out on the table too and check to see if there is any twisting. If you get it right, you can gently bend the areas to provide more contact with desired area. Replace the heat paste with something like arctic MX3 for best performance, and when you screw it back on, start with the NB screws first, and work your way outwards. If pumping 1.4v through it, I often reach the 60's but never 70°C. If all that fails, I would say you have a duff sensor. Good luck!
when replacing the thermal paste, don't forget the 2nd part of the NB block, the piece that unscrews from the larger assembly.
i used IC D7 paste and didn't see NB temps over 52C. my pic above is the temps with aftermarket heatsinks, my NB is a thermalright hr5-sli ifx, the mosfet to the left of the socket is thermalright hr9 type 2 and above the socket is Enzotech 81 something or other, SB is also an Enzotech but i may change that to a thermalright also.
It's strange how my NB temps seem to be so much higher, I can only put it down to running a 65nm Quad (as opposed to the majority who are running 45nm), it also seems that I generally need more NB volts too. It was the same on my P5B-D. On both MIIF boards the lowest NB Idle temp I saw was around 37-38°C, and that was with the NB naked against a waterblock, but as you might have seen a few pages back, without the heatsink assembly, some temps improved slightly, and other just increased beyond comfortable.
Once I get my new bits in, I'll probably go for seperate sinks on the SB and mosfets, and put the waterblock back on the NB. Hopefully a good system flush, followed by installing an extra Rad, exhaust fan and using better coolant will yeild better results. I've been stuck with the last of this huuuge tube of AS5 paste for a good few years now, so I hope my MX2 and MX3 will also improve matters. The memory and CPU I hope to receive by the end of the week will almost definitely will :)
hey guys
thanks for the feedback
actually this is my 2nd M2F that i test
the 1st set was a loaner from a good pal mine
then i decided to buy one for myself
this didnt occur on the loaned set......but only to mine
last time before sending mine out for RMA
i removed every piece of the heatpipe assembly
i tried everything from washer mod to changing TIM
same results.......
thats when i decided to sent it in for RMA
upon receiving the board from RMA
everything looked new
i remember i kinda flex the board at the top mosfet area and there was no sign of flex when it came back
so i assumed they must have fix the board
unfortunately......same problem still exist
i wouldn't mind and i'd totally understand if i'm getting 70-80'c at 1.4v or sumting
but the thing is im running 1.1vcore, 1.1vnb, 1.1vtt & 1.5vpll
Definitely sounds like another duff board mate, I would get it shipped back if the following fails:
Does the temperature drop if you push down on the northbridge area of the heatsink whilst on?
Just a long shot, but have you tried double-checking the temperature in everest? I've never really trusted the Asus probe software - it's functional, but a resource hog and buggy. Does bios show high NB temps too?
agreed, last time i used Asus Probe (4 yrs ago), it didn't read true half the time.
Asus DO say on their site to NOT send all of the accessories and additional stuff, JUST the board :p:Quote:
Originally Posted by In-Fluence
I know, the RETAILER asked me to send everything back. I was just following their instructions. What they did from that point on wasn't my problem :)
Well if the retailer said to send everything back to asus, then their stupid. Just caused you 2 more weeks of pain :(
I shipped the item to the retailer. Tbh I think they did botch it a bit as they ended up sending me a brand new one which came from another online store. I would assume they probably sent the complete kit back to Asus and they sent back an appaulingly packed board directly to me. I've done it this way before without any major hassle. My guess is the retailer realised Asus' procedure for sending back boards, but wanted to provide a good service and got a new one for me. Either way they came through good, and I borrowed work's P5Q-D until I got the replacement so there wasn't anything lost really. In fact, it was a good opportunity to see what the P5Q-D could do :)
I'm about to try out your bios, gonna have a go with the reapers first, then try the dominator gt's if things go well.
I'm getting good results so far :up: Experimenting with low voltages and so far 1.35v NB seems enough for 465FSB PL8, RAM @ 1166MHz, ain't too bad so far. This mBIOS is almost like magical pixies dancing inside my mobo, the vanilla 2202 just didn't let me OC with any stability - period.
are you saying that your modded bios is helping lower NB volts or you just happen to have a board that does well at low volts?Quote:
Originally Posted by Ket
From my experience with the modded BIOS, its helping lower volts for me. With vanilla 2202 I needed 1.4v NB to even attempt that.. nevermind get it stable. Currently I'm sitting on 475FSB PL8, deffo not too shabby for minimal work :up:
I have the bios 1901 also I have the g-skill trident 2x2 1066mhz, all that it he bears in oc without moving 1.8v?
Maximus2 formula
E8600
2x2 trident 1066mhz 1.8v wath chipset it has ?
Stay on 1.8v, your memory should be good with that voltage. I'm not sure I understand what you were asking.
For chipset voltage, it depends what performace level and clock twister you want to run as well. Try 10 and 'lighter', then raise them. Try and stay around 1.3-1.4v for northbridge.
Did I understand your question correctly?
I would like to know how well the tridents work. I'm considering buying some different memory. Will you try to overclock it? ;)
Hey fellas, I have a hard time getting my q6600@3.8GHz.
I had asus p5k and biostar tpower i45 before and my q6600 would run stable at that freq with 1.46V (same as I used on m2f), but now I seem pretty lost in all the options of this MB. It needs roughly 1.5V to get to win without bsod.
BIOS: 2104
Could you help me out with this maybe?.
Options I currently use are below
PHP Code:Extreme Tweaker
Ai Overclock Tuner : Manual
OC From CPU Level Up : AUTO
CPU Ratio Control : Manual
- Ratio CMOS Setting : 9
FSB Frequency : 423
CPU Clock Skew : Normal
North Bridge Clock Skew : Normal
FSB Strap to North Bridge : auto
DRAM Frequency: DDR2-1016
Dram Clock Skew CH1 A1 : Auto
DRAM Clock Skew CH1 A2 : Auto
Dram Clock Skew CH1 B1 : Auto
Dram Clock Skew CH1 B2 : Auto
DRAM Timing Control: Manual
CAS# Latency : 5
RAS# to CAS# Delay : 5
RAS# Precharge : 5
RAS# ActivateTime : 15
RAS# to RAS# Delay : auto
Row Refresh Cycle Time : auto
Write Recovery Time : auto
Read to Precharge Time : auto
Read to Write Delay (S/D) : Auto
Write to Read Delay (S) : Auto
Write to Read Delay (D) : Auto
Read to Read Delay (S) : Auto
Read to Read Delay (D) : Auto
Write to Write Delay (S) : Auto
Write to Write Delay (D) : Auto
Write to PRE Delay : Auto
Read to PRE Delay : Auto
PRE to PRE Delay : Auto
ALL PRE to ACT Delay : Auto
ALL PRE to REF Delay : Auto
DRAM Static Read Control: Disabled
Dram Read Training : Disabled
MEM OC Charger : auto
Ai Clock Twister : auto
Transaction Booster : manual
perf level 12
pull in all disabled
PCIE Frequency : 100
CPU Voltage : 1.46875
CPU PLL Voltage : 1.51325
FSBT : 1.259
DRAM Voltage : 2.15775
North Bridge Voltage : 1.35175
South Bridge Voltage 1.5 : 1.53975
South Bridge Voltage 1.1 : 1.13975
CPU GTL Reference 0 : AUTO
CPU GTL Reference 1 : AUTO
CPU GTL Reference 2 : AUTO
CPU GTL Reference 3 : AUTO
North Bridge GTL Reference : AUTO
DDR2 Channel A REF Voltage : AUTO
DDR2 Channel B REF Voltage : AUTO
North Bridge DDR Reference : AUTO
Load Line Calabration : Enabled
CPU Sread Spectrum : Disabled
PCIE Spread Spectrum : Disabled
Trust me, I wasted months trying to get my Q6600 stable when overclocked. The only 2 things I feel are worth suggesting are to try adjusting CPU/NB skews to something like 100 for CPU and 200, 300 or 400 for NB. This helped me stabilise a lot at higher fsb, but test your own values and find the most stable point. I also found it was a lot easier to get 450x8 stable than 410x9 for some reason, so try to find out your max fsb, if you can boot 470-490 easily then it might be worth trying the x8 multiplier. You will probably need higher vtt and NB for this though. You'll also need to tune GTLs over 450fsb. Good luck with that, keep us posted ;)
On another note, I'm having a lot of trouble getting ~96-64k FFT stable. I've tried raising pll, vtt, nb voltages and adjusting gtls, but it just doesn't want to play ball. It always resets when hitting 96k and occasionally fails the 80k prior to that. Another overnight bleand test proved stable everywhere else until it hit the 96k range. At this point temps are at their highest across the board, but don't go far over 50°C which is still lots cooler than with my Q6600 before so I don't think its overheating (Q9650 hits 62-63°).
Anyone else experienced this and know how to stabilise?
not sure anyone saw this sticky yet?they are focusing on the
newer chipsets but say they will accept input on all ROG products
http://www.xtremesystems.org/forums/...d.php?t=238770
I've remembered it wroooong godamit :D
I had 3.6GHz@1.46V (drop was about 0.03), so i don't actually have a problem... :)
Works fine :)
On Maximus 2 Formula and setfsb what code should i use?
This far the ICS9LPRS918BKL is worked pretty good.
SETFSB does not yet support P45, it freezes mines. (look in "read me")
OK.Quote:
Originally Posted by humeyboy
Its weird that i have tried ICS9LPRS918BKL and JKL and some 919 ended works just fine for me.
ICS9LPRS918BKL this is the one that works for all
the P5Q series it should work with the MaxII.
It is possible to use the Heatpipe cpu and Mosfet Nb of the Asus Rampage extremes?..
Possible, never tried. However they are essentially identical in design, I'd be extremely surprised if there was any temp difference at all assuming you could even use it. You should be able to screw on the extra NB fins from the REX in place of the stock 'pin fin' NB block.....not sure if this would help at all though the main issue is the crappy block & 3 layers of paste. The basic design is absolutely terrible for both HPA's.
Did a bit of a rebuild today and pimped my rig. Nice to be able to monitor coolant temps now. Also got rid of the heatsink assembly. NB temps weren't reassuring enogh for me. Now I feel I can give this thing some volts and see how it fares :)
Here's my post in the w/c gallery thread:
http://www.xtremesystems.org/forums/...postcount=1298
And the kind of load temps I'm looking at:
http://i194.photobucket.com/albums/z...wcmodtemps.jpg
Unfortunately after removing my waterblock I noticed it wasn't making fulll contact with one corner of the CPU, I think it's surface is really uneven, but really don't want to lap it yet. Temps are still quite even on all cores though.
please please please guys, dont listen to this animal...what he is doing is to overclocking his computer, damaging it and then he sell it, he doesn't know anything about overclocking.Quote:
Originally Posted by Mikefra2008
Here in chile, mikefra is a liar and a scammer, after doing this, he sell his computer...this is the only place where he can get some help with his computer
sorry my english, if you can, read this
http://www.overclockers.cl/foros/ind...owtopic=249604
http://www.facebook.com/group.php?v=...d=175764822034
http://www.tarreo.com/foro/index.php?showtopic=528498
http://www.madboxpc.com/foro/Asus-Ra...a-t109517.html
http://www.madboxpc.com/foro/Denunci...w-t111141.html
http://www.chw.net/foro/overclocking...e-mikefra.html
bye
good news for you de5tr0yer btw :) I guess I must have just had quite a naff q6600, yours seems to fly along nicely. Are you gonna try and push higher?
I've spent the past couple of days tweaking my board with this q9650 and I gotta say it does suit it very nicely. I'm running some good speeds at good volts with plenty of headroom for both temperature and voltage. So far I've managed 4.32GHz P95 stable (one complete blend loop at 480x9) with a measily 1.3v :cool:
Been at 490x9 for a day or so, but still not quite got it right. It's failing around the 320-256kfft mark, changing voltages hasn't helped, but still more to explore - gtls next - I'm sure this will get me close and I'll discover what needs adjusting to help compensate. I'm also suspecting skews to 300/300 might be worth a try as it helped me to do a 1M PI bench at 509x9, alas, I've still not broken the 10s barrier :(
One major setback is the rather high improbability that I'll manage anything over 500fsb stable which is a shame, I'm sure this CPU could do a lot more.
Any recommendations from this point?
http://i194.photobucket.com/albums/z...vcoretuned.png
I was removing HSF from my board, the stupid reverse thread screw that feeds down through the board to a normal thread not became loose, preventing me from removing the HSF.
My only option was to lever in a mini visegrip to try and hold the screw to allow me to get the nut of and remove the HSF.
Anyways, the vicegrips slipped and i damaged a capacitor, didnt look too bad but when i fired it up the two small compoents circled blew.
They also need to be replaced, but I'm waiting on some info so i can identify them.
I have sourced a repacement capacitor and have a decent soldering station.
I have also successfully recapped motherboards before.
http://i50.photobucket.com/albums/f3...h_DSC00009.jpg
http://i50.photobucket.com/albums/f3...h_DSC00011.jpg
If anyone can help (Im asking around) with identifying the components in the red area I would be very gratefull indeed.
A close up picture showing the markings on the 2 small components (resistors i think?). Thanks :up:
Sorry to say mate, but I wouldn't hold a lot of hope. Your best bet might be to try and get another broken M2F and swap bits from that, but those parts are tiny and although I worked in a SMT factory, these parts are very hard to identify. I would also say that its likely other components are now damaged because you switched it on after you broke it. I'd go for a new board to be on the safe side, damage around that area could mean bad things for your CPU/RAM
You could try to rma it. I had the same situation with my Maximus Formula board:
I removed heatpipe and installed alternative coolers and at the first start a small piece was damaged. So I reinstalled the old pipe and sent it to the shop. A few days later I got a mail from the shop that they saw that the heatpipe was removed and that they will send it to Asus. 2 weeks later I got my board back from Asus, repaired!
Really lucky I guess. But it`s worth a try (that`s very bad english, isn`t it? But I don`t know it better, sorry.)!
Noob works !!!Quote:
Originally Posted by bhzrd_rs4
http://valid.canardpc.com/show_oc.php?id=850009 :up:
Tried to do a superpi 32m at that speed, but I crashed 7,5 minutes into the test. :(
Do note the pretty high NB stress. I'm running the 1:1 RAM divider because my ram sucks.
Can boot into windows with same setup with ram at 1225mhz, but it is less stable. Crashes 3 minutes into 32M superpi.
I'm also running all 4 slots occupied.
Same as I stated in other thread means next to nothing unless hours of Prime and a few runs of IntelBurnTest. ;)
I have been in /windows running normal everyday Programs/Apps at over 500FSB but then next restart it will not work, thats Asus for you.
My MIIF is sensitive to NB temps once latency starts to get down under 60ns..or if pushing for FSB over 515mhz. DICE'd it will run 575FSB + with tighter latency..but air cooled limit for me, is ~520mhz FSB with ~57ns latency.Quote:
Originally Posted by humeyboy
Your on a Dual Core m8, I had 575FSB with my E8600 with lowish voltages (8GB Memory).Quote:
Originally Posted by CL3P20
A Quad Core is a different kettle of fish though. :)
Hello
I have changed out my board for a foxconn black ops.
It now sits in its box and I have no use for it.
It is a great board, I just could not pass up the good deal from newegg for the x48.
If I was to sell this asus board, what would I get for it?
Where would be a good place to sell it?
I also have a Q6700 to sell to.
Thanks
Dear all!!!
i've got a problem with this mainboad.. i have a q9650 and i can set all the cpu pll value without any problem. There is a little overvolt but it's ok. When i try the same think with a e8600 cpu q22a435 it seems to have cpu pll locked @ 2.106. every value i set i get 2.106 in bios monitor? any idea?
thanks
vamp
You have pll voltage set at 2.1V? And CPU is still working? :eek:Quote:
Originally Posted by vampire2k5
Or is it just me who learned that the pll V must stay as low as possible?
i set it at 1.51v and now from bios i read 2.2v cpu is alive for now and under oc works pretty well with very low temp. I hope in a sensor problem of the board. the strange thing is that with my q9650 i have no problem with cpu pll.Quote:
Originally Posted by De5tr0yer
Then it is not cpu PLL Voltage, check the names again.Quote:
Originally Posted by vampire2k5
But if your cpu works, then surely it isn't 2.2V ;)
i`m sure about it.. it`s cpu pll voltage .. with my q9650 if i set it @ 1.51 volt i have a little overvolt but it`s ok. with the e8600 it`s going crazy and every voltage i set it always show me 2.106 or 2.2v ...Quote:
Originally Posted by De5tr0yer
I`m writing both to intel and asus support...
Vamp
You could check it again with a tool like Everest or something to see if it displays another value.
If you are on an old bios version, you could try a update.
Have you tried removing the CPU, powering up without one in, then putting your dualie in again? This clears the CMOS, and I assume might help it prep for another cpu... I had a similar problem recently where after swapping my GFX card from the 8x PCIe slot to the 16x one, my CPU kept hitting 100°C and shutting off, after checking voltages all were actually normal, even after loading a pre-saved profile, but CPU was deinitely reaching that temp judging by how quickly the liquid heated up. Doing the aforementioned solved it though. It didn't half scre me though, but no real damage done I think...
when i moved from q9650 to e8600 it told me that a new cpu was in use. I also tried ALL the bios from 13** to 2202 but nothing happens. Now i will try this cpu with a rampage formula..Quote:
Originally Posted by In-Fluence
tnx
Vamp
CMOS Clear!?
Quote:
Originally Posted by vampire2k5
How did you update bios? If you've done it from win, try from dos.
And ofc, clear cmos before upgrading bios ;)
Got a bit of an odd problem here guys...
I've got a Q6600 LinX stable @ 3.8GHz w/ 1.45v on this board but I cannot get 3.9GHz for love nor money! 1.5v fails, 1.55v fails. 1.6v fails! But it doesn't just lock or crash, anything over 1.488 volts seems to just reboot the system as soon as it gets 100% load.
It's a new M2F with a 1.25v VID Q6600 and a new Corsair 850HX PSU so I know it's not the power requirements that are failing it...
Has anyone got any ideas as to why it would just power down from a certain voltage onward? Are there any jumpers I need to move to allow a certain voltage limited to be lifted (I couldn't see anything documented)?
I had the same issue with my P5B, and it only slightly improved with this board, but did enable me to boot at 4GHz with a Q6600. The only thing I feel might be worth checking is the 'TM enable' feature under CPU settings - make sure it's disabled if you havent already as this can cause the PC to shut down / reboot if temps hit TJmax (normally 71°C on these chips). Failing that, CPU/NB skews might be worth a try - post your settings, there might be something we're missing..
CPU TM disable hasn't changed it. :( I'm quickly trying my older Thermaltake Toughpower 850w to see if it's the PSU.Quote:
Originally Posted by In-Fluence
It is literally anything above 1.45v does it. I might try with lower clocks to see if I can isolate it as a voltage problem.
EDIT: Toughpower exhibits the same behaviour. Back to drawing board...
Gunna try 3.6GHz with 1.5v...see if it'll push it over the edge like that...
Ok, it seems to be certain FSB/vCore combo.
9x400 1.35v vCore = stable.
9x400 1.5v vCore = stable.
9x425 1.45v vCore = stable.
9x425 1.5v vCore = stable.
9x435 1.5v vCore = shutdown.
8x450 1.35v vCore = stable.
8x450 1.5v vCore = shutdown.
8x475 1.45v vCore = stable.
8x475 1.5v vCore = shutdown.
My basic voltages for 520MHz FSB Prime stable with 8GB ram are:
vPLL = 1.5v
vFSBT = 1.4v
vDIMM = 2.3v (within ram specs so it's safe)
vNB = 1.4v
vSB 1.5v = 1.5v
vSB 1.1v = 1.1v
I have tried bumping all the voltages up to the first yellow value (I forget the values for each one) and keeping the vCore @ 1.5v and it still reboots so it doesn't seem to be a vPLL/vFSBT/vNB voltage problem.
Would loosening the Transaction Booster level to 11 or 12 make a difference?
I'd be stunned if you were completely stable at 520MHz FSB, I assume that's a typo?
From my experience, I think it's quite a common trend for Q6600's and Asus boards to shut down with high vcore. My P5B did it, and P5Q, and MIIF, although I got furthest with the latter. Might be the same for other brands but couldn't say for myself.
Completely stable @ 520MHz. :D 500MHz can Prime Blend indefinitely and I've had about 2-3 hours out of 520MHz before I got bored... :p:Quote:
Originally Posted by In-Fluence
I've never had an Asus board do it before which is what is bugging me. :(
Please post your 500FSB settings. :)
Also is your mobo a Rev A2 or A3 NB ?
vPLL = 1.5vQuote:
Originally Posted by humeyboy
vFSBT = 1.4v
vDIMM = 2.3v (within ram specs so it's safe)
vNB = 1.4v
vSB 1.5v = 1.5v
vSB 1.1v = 1.1v
Transaction Booster @ 10
DRAM Static Disabled
Clock Twister @ Strong.
It's an A3 Northbridge. This Primed for an entire night. I fought with it for ages and eventually went back to basics, lowest volts, overclock, stable, overclock more, fail, more volts, stable, overclock more... I got to 485MHz with 1.3v vNB and vFSBT but from 485 to 520MHz needed 1.4v for both.
OK 1st thanks for info but its not really any good to anyone.
Need to know if its 1:1 and GTL Values etc.
I gather by your Memory Voltage its not 1:1 (which is harder on NB) unless that's PC6400 your overclocking.
I cannot go for non 1:1 for 5000FSB as my Memory will not do 1200mhz on this Mobo (supposedly it will on other Mobos as its the older Rev and clocks well).
All the Memory settings that can be enabled or disabled should not matter if 1:1 if its at least PC8500 as at 1000mhz it will be underclocked.
Also due to me currently doing some work on PC I have only 4GB installed so it again should be easier even on a Rev A2 NB.
I tried your above settings be it a different NB and CPU but it would have been a good starting point IMO.
I cannot even get to Windows, chucks it after Bios posts.
I have Primed at 500FSB before but it was not for long and no settings were stable for any 2 reboots and even tried it with only 1 Memory Module so 2GB and even a 1GB PC6400 Module I know does 1000MHZ in other Mobos and it still isn't stable.
Like Humeyboy said, detailed o/c settings would be ideal. There's a template available if you go into the intel forums, then click on the paperclip by this thread's name :)
Would like to see some screenshots too, as I'm sure most people on this thread would as I think this far over 500FSB stable is unheard of on this board. Could you double check CPU-Z and check the number of cores?
I could get 500fsb stable but only with 2 cores enabled, the other two disappeared somewhere beyond 488 and there was a gaping fsb hole until ~500. It didn't play ball with that chip at all for me. Q9650 is lots better in this board now, but still not that impressive, which is why I'm gonna give a UD3P a whirl like so many others.
Sorry, I was just posting the settings I had changed to achieve the clocks.Quote:
Originally Posted by humeyboy
GTL settings were AUTO, ram was 400 strap, performance level 10, clock twister @ strong, dram static disabled and ram @ 1:1 so @ 1000MHz.
All 4 cores. I don't have my 520 settings to hand but these are my Prime blend 12hours stable for 500/510MHz.Quote:
Extreme Tweaker
Ai Overclock Tuner : Manual
OC From CPU Level Up : AUTO
CPU Ratio Control : Manual
- Ratio CMOS Setting : 6
FSB Frequency : 500 / 510
CPU Clock Skew : Normal
North Bridge Clock Skew : Normal
FSB Strap to North Bridge : 400
DRAM Frequency: DDR2-1000 / 1020 (510MHz)
Dram Clock Skew CH1 A1 : Auto
DRAM Clock Skew CH1 A2 : Auto
Dram Clock Skew CH1 B1 : Auto
Dram Clock Skew CH1 B2 : Auto
DRAM Timing Control: Manual
CAS# Latency : 5
RAS# to CAS# Delay : 5
RAS# Precharge : 5
RAS# ActivateTime : 15
RAS# to RAS# Delay : auto
Row Refresh Cycle Time : auto
Write Recovery Time : auto
Read to Precharge Time : auto
Read to Write Delay (S/D) : Auto
Write to Read Delay (S) : Auto
Write to Read Delay (D) : Auto
Read to Read Delay (S) : Auto
Read to Read Delay (D) : Auto
Write to Write Delay (S) : Auto
Write to Write Delay (D) : Auto
Write to PRE Delay : Auto
Read to PRE Delay : Auto
PRE to PRE Delay : Auto
ALL PRE to ACT Delay : Auto
ALL PRE to REF Delay : Auto
DRAM Static Read Control: Disabled
Dram Read Training : Disabled
MEM OC Charger : auto
Ai Clock Twister : strong
Transaction Booster : manual
perf level 10
pull in all disabled
PCIE Frequency : 100
CPU Voltage : 1.25
CPU PLL Voltage : 1.5
FSBT : 1.4
DRAM Voltage : 2.3
North Bridge Voltage : 1.4 / 1.5 (510MHz)
South Bridge Voltage 1.5 : 1.5
South Bridge Voltage 1.1 : 1.1
CPU GTL Reference 0 : AUTO
CPU GTL Reference 1 : AUTO
CPU GTL Reference 2 : AUTO
CPU GTL Reference 3 : AUTO
North Bridge GTL Reference : AUTO
DDR2 Channel A REF Voltage : AUTO
DDR2 Channel B REF Voltage : AUTO
North Bridge DDR Reference : AUTO
Load Line Calabration : Enabled
CPU Sread Spectrum : Disabled
PCIE Spread Spectrum : Disabled
I don't have any screenies atm but I can get a couple of quick LinX runs tomorrow if people want? I don't see why people would disbelieve what I've said, I thought I'd been around long enough and put enough time and effort into the Rampage Formula thread to not need to provide proof for my clocks. :(
Got 520MHz w/ 8GB Priming for you guys now.
dont take offense,were not doubting your skills but the MIIF itself.Quote:
Originally Posted by kup
That's cool, I know what ya mean, sometimes very out of line results crop up and they need to be verified. :up:Quote:
Originally Posted by radaja
I'm about 2hours into a 520MHz Prime Blend session. It's got a touch more voltage than for 510MHz but it's stable and Priming. I'll either post pics in about 10 hours on my lunch break or tomorrow evening after work. :cool:
hell im happy with 2-3 hrs prime blend.thats quite the accomplishment with a quad and the MIIF
I woke up this morning to a locked screen after about 7 hours (according to start time and frozen clock) so have started again for you. My only screenie was just under 2hours in so I'll post that plus any more I have this evening.Quote:
Originally Posted by radaja
Seems like you're doing extrordinarily well Kup. Like others have said, it's not that we don't trust you, but so you can confirm the virtually unbeleivable! The amount of time I've wasted trying to get my board stable at 460+fsb with my old quad, and ~500fsb stable with my current chip is silly, and I'm just speaking for myself! Massive kudos for getting as far as you have, would be great to see higher clocks. Does it still behave if you up the multi?
Cheers dude. Sorry for coming in here all headstrong, I didn't realise it was such a high fsb until you guys said. 500 was a breeze and the 510 and 520 were a bit of a struggle which is why I thought lots of people would have done it too.
I usually test with a quick 3run blast on LinX and if it passes that I'll try for at least 6hours Prime x64 Blend or Small for cpu.
Funny enough, with only 4GB installed I experience more instability than with 8GB...lol.
Will be able to try that for myself shortly. Turns out my RMA'd sticks have been sitting in our local sorting office for a week and I've had no 'failed delivery' card... Royal mail are doing themselves no favours at the moment! :ROTF:Quote:
Funny enough, with only 4GB installed I experience more instability than with 8GB...lol.
What revision is your board? 1.03G?
Not sure off the top of my head but I'll have a look when I get home tonight.
As promised...
http://img707.imageshack.us/img707/6...hz5hour.th.jpg
This crashed after five hours but this is during the day so it's warmer. I came home to find my curtain had fallen down (lol) and was caught in the fan on my heatsink...oops! :rofl:
It behaves @ 500MHz up to 3.5GHz and then I can't go any further because I go past my CPU's limit (bloody thing is 3.825GHz stable with only 1.45v but hits a solid wall right there as documented above ^^^).Quote:
Originally Posted by In-Fluence
those are insanley high volts on the vtt/NB/pll
but nice job:up:
I don't run it like that for 24/7 operation. :D :eek: I'll work on lowering those volts. :up:Quote:
Originally Posted by radaja
Any ideas why I would hit a wall with overclocking, as I commented on above?
not really sure why,but maybe just try getting 7x500@3500 stable with decent voltages.
does 7.5x500@3750 work?
Q6600 here so no half multipliers. :(Quote:
Originally Posted by radaja
I've flashed Ket's mBios 1901v2 and will retest everything again. I'm sure I had it running past my 'wall' on the 1901 bios before I flashed it but I don't have anything documented in my notebook. :(
EDIT: And this board is a rev 1.02g.
whats the chipset rev?you can find it in cpuz under the motherboard tab,its either A2 or A3
oh i didnt know that about the Q6600.just stick to 7 then
It's an A3 chipset.
They are very high volts. I bet you would be able to get that 520MHz setting stable and lower a load of volts after tuning your NB and CPU GTLs. Very impressive indeed though, well done... I've not tried such high NB volts myself but with it sitting naked against a waterblock and it not hitting more than 40°C with over 1.5v, I'm starting to wonder if a bigger bump would make all the difference...
Ok, I'm struggling with the GTL values. What should they add up to? CPU GTL equation should add up to...? And NB GTL equation should up to? I've used the settings presented in the GTL thread for 65nm CPUs (+50,+10,+50,+10) and it lowered both 3.6GHz and 3.8GHz required vCore for LinX x64 stability (3.6GHz @ 1.35v and 3.8GHz @ 1.4375v) but I'm unsure what the NB GTL should add up to, is it supposed to be whatever the vNB is?
I too have not got a real answer on NB GTL's but I cannot run Minus only Auto or Positive.
I am at 475FSB on 1.27v NB & VTT and NB seems best at +40.
I can get 485FSB stable with Ket's Bios but the gain in voltage although not too high is not worth 10MHZ IMO.
I want next 500FSB or I simply will stay at 475FSB.
Today I wasted all my time on once again trying to clock this POS to stable 490+FSB and no matter how much voltage or what tweaks its no go for long if it even boots (ain't the same any 2 boots).
If I were you I would be thankful and clock down to 5000FSB and lower voltages (yours are crazy high).
The voltages are nice and low again now, that was only a bench/test run.Quote:
Originally Posted by humeyboy
I've found a nice guide on GTL settings here - http://www.overclock.net/intel-mothe...ml#post7161320
I'm a bit confused as to which I tweak to what for a 65nm CPU, do I tweak .667 to .635 or .635 to .667 for the CPU GTL?
Just found out what I set was correct for my NB.
Treat it the same as the CPU GTL's (weird as the NB is not the same nm as the CPU I have).
kup, those are ridiculous high volts on pll / vtt / nb.
You are for sure degrading some components. I wouldn't run prime for hours with those volts.
Anyway, look at simps's guide on this thread, he had a 65nm quad too, and he got it 24/7 500FSB stable with reasonable volts.
http://img80.imageshack.us/img80/5244/new50013wu3.jpg
The volts are much lower now. That was just for 520MHz.
I couldn't agree more with Marcio:) It seems like you're one of the lucky ones who got a good CPU & board combo and it could be that yours will behave the same as Simp's setup did. It's a very thorough guide, worth trying in your case I think :)
Anyone got a link to his post? I can't use the search feature at work. :(Quote:
Originally Posted by In-Fluence
Cheers Influence. I've just had a try but my 500MHz voltages are already lower. :( I'll keep trying for more.
This Q6600 is pissing me off now, it literally stops dead @ somewhere between 3.7GHz and 3.9GHz with any voltage over 1.464v...no crashes, no lockups, just power down. :(
I have once again Google'd some keywords and found a few sites concerning this Mobo and OC'ing.
You have to take some of the info as BS and other info as a "trial and error thing", but there is also good info out there.
Now what I did not tell you the other day was that although its true I was doing some PC work the real reason I'm on 4GB for now is due to me dropping 1 of my Modules onto a hard floor and now it will not even post. :(
I have found 2 settings (I have messed with them before but there is too many variables) and now I am 100% stable @ 490FSB, 99% stable @ 495FSB (still tweaking/testing).
When I get 495FSB 100% stable and I try lower some voltages and tweak GTL's more and do a proper long Prime/Orthos and then I will then use these settings to try and get 500FSB stable as I will only need to raise a few of many settings I hope.
I do not see my 8GB being an issue as this Mobo was made for 16GB and I have had the exact same stable or failed settings with 1-4 Modules (2GB-8GB).