Where do you find this option in the BIOS? Turbo Power LimitQuote:
Originally Posted by Vandre
Printable View
Where do you find this option in the BIOS? Turbo Power LimitQuote:
Originally Posted by Vandre
In the AI Overclock Tuner screen, two steps below the speedstep option, only visible when Speedstep is turned on, multiplier set to auto and turbo set to on. Perhaps it needs to be set to Manual mode as well, can't remember right now.
Thanks for posting that! :clap:Quote:
Originally Posted by Vandre
This whole time I'm thinking this will never be fixed and lo and behold it works!
Once speedstep is turned on with auto multi that option shows up. I set it to disable and put AI back to manual, set the multiplier, turned off speed step (it hides the option again but it stays off!) and reboot. Now the multiplier stays locked. I just tested with a 980 at 133x33 and it stayed at 4.4GHz though a run of Cinebench and LinX. :eek:
Whee! GTX480 SLI working on this board too. :)
hey guys I need some help
I finished my water cooling I just need to overclock my cpu now
can somebody give me the best way to get to 4ghz (I mean things like how high my bclock needs to be)
I don't need a super mega over clock I will be satisfied with 4 ghz
specifications:
cpu:i7 920 D0 (good batch)
motherboard:asus p6t7WS supercomputer
Ram:corsair dominator 1866 6gb
watercooling:apogee XT
HW labs sr1 420
temps:idle 30/32
thanks in advance
ftp://ftp.asus.com.tw/pub/asus/mb/so...-ASUS-0811.zip
[ 0811 ]
----------------------
1. Patched system hanged after overclocking failed with i7 980X or changing CPU from
Gulftown to i7 980X.
2. Set default uclk frequency to 2133MHz for i7 970.
3. Change String "CPU turbo power limit" to "High TDP Turbo Mode"
thanks. ASUS FTW-) heheQuote:
Originally Posted by Fire4Wire
I had perfectly working 0804.Quote:
Originally Posted by KingOfsorroW
BUT 0811 is making problems again.
I can't make it work, tried all combinations with on/off.
Reverted back to 0804 and everything works again.
Can anyone else test this ?
Cheers
i wanna try out this board
can it do 230blk+?
i read the thread got me worried i wont be able to hit 4.8 cause of throttling
If there were any throttling, this thred would've been up and running.Quote:
Originally Posted by zsamz_
There is no "actual" NF200 voltage adjustment on this board is there? At 110MHz PCI-E it's crapping out.
nice to see no throthling :)
cpu 1.36v vdimm 1.66v:D
Rev.2 of the motherboard has been released but no info on it anywhere on Asus's site.
Memorable difference so far is a 4 pin molex EZ_PLUG in the top right hand corner by the memory sockets.
I know just got one as an RMA on the one I sent back with a burnt out 24pin socket which pulled too much power when running quad fermi.
http://i232.photobucket.com/albums/e...e/P1100085.jpg
That's nice to know. I wonder what that molex connector is used for. Does it mean that the current versions don't supply enough power to pcie or is it meant for better cpu voltage support? ... interesting...
Either way it should take a load off the 24 pin, I'll soon find out :)
This dropping multi is driving me crazy... im running bios 0811 with no end in sight... this is a retail 980x...
during peak testing, the multi drops back to 20... between tests when the cpu temp/load drops off, it goes back up to 30...
im dying to get this thing stable at 4.5 without the turbo/multi crap going on... this is under a thermal right ultra 120!!! i doubt it could sustain those temps if the heat keeps coming at solid 4.5 though....
see screen shots...
bennie flash back to 804
ill give it a try this afternoon and report back...
didnt do anything.. multi still drops... anything over 25 multi, it drops back to 25... right now im at 4300 @ 1.36 volts passing max linpack on 5 runs... 172x25... temps max out at 75ish...
did you use the turning off turbo power limit trickQuote:
Originally Posted by benniebeeker
yes... :(
i am at 4.4 now and going... still tweaking... hitting 80-82 while running linpack... temps may be my wall on air...
i got my 920 workin perfectly with 804
with any other bios i get multi drop
ahhhh.... makes sense... im using the 980x... 804 bios is still droping for this chip... doesn't matter, really... its not my rig... ;)
try the trick i had to do it twice before it workedQuote:
Originally Posted by benniebeeker
Hello Zsamz, I am using 0811 bios and have multi drop to 25 on my 980x retail.Quote:
Originally Posted by zsamz_
Is there a way to eliminate it using this bios?
Thanks in advance.
try 804Quote:
Originally Posted by vmanuelgm
read this
http://www.xtremesystems.org/forums/...&postcount=489
[ 0904 ]
----------------------
1. revise CPU offset temperature for ESC 1000 system.
Mine arrived today, and it's slightly diffrent. Theres a molex plug next to the 24 pin connector, which does not appear in other pictures i have seen. I suspect it's a modification to increase power supply to the board, however, i have not seen this plug on any others i have looked at. Does anyone else have this white molex plug on the board? I will have it together in the next couple of days, and report back on any other diffrences i observe. I'll check also the throttling situation on this model, and report back on what happens with this revision, hopefully it will be fixed.
My mistake. Someone on the previous pages has already posted this change. I thought is was something new that had not been seen on previous versions.
I added adjustable multipliers and turbo TDP/TDC limits to RealTemp 3.60 and I was wondering if this is useful at all to the dropping multiplier problem.
If the method posted by illuminatiASUS doesn't work for you then can you send me a PM so we can do some testing. Being able to adjust turbo TDP/TDC on the fly has made a big difference to some of the Intel Core i mobile processors that are a lot more limited for bios options.
RealTemp 3.60
http://www.techpowerup.com/downloads...Temp_3.60.html
Has anyone figured out where the voltage read points on this board are, specifically for VDIMM? There's a good picture of the P6T7 "naked" here and I have a nagging suspicion that the memory voltage points are next to the three MOSFETs beside the DIMM slots...which are covered by the cooling assembly :doh:
Can anyone help with a RAID issue i am having. I have 4 HD's. All are attached to the intel ICH10 RAID controller, on channels 0.1.2.3. 2 of the drives are western digital 300gb's, on channels 0 and 1, the other 2 are 500gb maxtors on channels 2 and 3. I have windows loaded on the channel 0 WD drive. I am trying to set up channel 2 and 3 as a seperate pair of RAID discs, but everytime i combine the hd's as a RAID, it seems to write off the data on the HD attached to channel 1 as well. It also will not let me select the HD on channel 0 as the initial boot drive after the channel 2 and 3 have been configured into RAID O. Does anyone have an idea what im doing wrong, once i have configured the RAID, it wont even boot into windows. Is it impossible on this board to raid 2 of 4 attached HD's and leave the other 2 as standard discs outside of the RAID array? Anyone had similiar problems they may have found a solution too? Im not at all happy with this board after having a really tough time trying to get this to work.
It had the latest bios at the time 0811 which I left on there.Quote:
Originally Posted by KaminoReal
Good news is no melted 24pin socket anymore, bad news is the 1500W PSU wasn't replaced under warranty, I just got what the supplier thought it was worth :(
But this one has been stable for months with a 970@4.2 and 4 x GTX470's@700Mhz so all is good so far. :up:
I have this board and I never posted before, but I will since I'm having a little trouble.
First off the board somehow refuses to boot with the 1394 firewire heather plugged in, it doesn't matter if there is a device or not, Just pluging the header and board not boot.
I have used the board without firewire for some time but what happens now is that when I shut down and come back the next day the board won't boot.......
I have tried cpr and all, I have a xeon w3570/ 12gb 1600mhz no overclock.
To get the system to startup I have to short the pins to clear bios/cmos.
I don't know before I had a i7 950 on it no problems, I also had the xeon w3570 on it no problems, the only change I made was I added a noctoua dh14, I the reconfigured the voltage for the ram (which on auto it was using 1.8v.) Factory specs are 1600mhz@1.65v 8-8-8/24. So I keyed that in and I noticed that made my system a little unstable again no overclock since I never saw a need for me to do it.
So I started looking for a way to stabilze the ram since that's the only thing I played around with. I after spending hours and still having issues for hours/days I found out they work perfect @ 1600mhz 1.6~1.65v at timings of 9-9-9/24, so I kept it at that..
So now I leave everything on auto or whatever I do to it doesn't matter. When I shut down and come back the next day.........press power.......board won't boot......I check for green motherboard power led.....its on.....I pull the side off..... jump pins......press power.....put pins back.....system powers on........
Damn I don't know if the ram did something....
Its patriot viper ii extreme I paid over $500 for the 12gb kit which is 3 sticks of 4gbs..
I tested the same ram on a rampage iii extreme which has the i7 950 at it refuses to run @ 1600mhz....its doesn't matter what I do to it...help...
Andreezy
Probably, you have a bad ram? Or, it is incompatible with the board. There is list of certified memory modules that are sure to work with this board on the asus site. Take a look at it.
test your ram on a different machine to make sure the modules actually work. Or test them one by one on your computer using memtest, maybe just one of them gives errors. Or you may buy a cheap set of memory modules and see if they would run on your system...
Good luck. )
i tested each of the 3 modules and there was one that had errors, this same one with errors has the heat spreaders loose, i had to press them back in place, they still look more loose than the other 2 modules.Quote:
Originally Posted by KingOfsorroW
i then tried to jiggle/move the heat spreader on the 2 good modules and they are solid, on the 1 module with bad/failed memtest they seem loose.
memtest reports that the memory ###### was unable to properly copy/hold data......cant rem exactly what it said, i just took it out.
i purchased a set of OCZ Platinum 6GB (OCZ3P2000LV6GK), so installed the new 6GB kit.
i just noticed after i installed the 3 new modules that this p6t7ws mobo does 1600->1866->2133->2400mhz on the ram.
i just :rofl: then i think i set them ram @1866 set cpu multis to 30*133(auto) and cpu to 1.25v, qpi/dram to 1.35v, everything else on auto.
i used the computer for a while made sure everything was stable, hoping it turns on tomorrow.
i shut down and now im typing from my laptop.
i tried turning the board on today, no cigar so i had to pop open a redbull and zipOnit while taking out the board....
so now i am sad, 1 board down and 1 memory module down.....:shrug:
i am back with my p6t7 fully functional, i called asus tech support to try andQuote:
Originally Posted by Andreezy
have them help me troubleshoot my problem (possible rma), a very
experience tech must have taken the call.
when i explained my problem with the board, i was told that it seems
to be a grounding/bios issue and to try and avoid an rma since it
would just be returned if it wasnt the bios and i loose time and shipping.
i then looked at the system running looking for a cause and noticed the
HDD/Power led never lid up. i though right of way of connectors possibly being the cause.
well then after i put everything back together (the right way)........:yepp::yepp:
now im running the board O'C to 4.0Ghz and thats 20 loops of interlburntest with all 8 cores loaded on max settings.
i am very pleased to say that i am very happy how things turn'd out.
except i had to rma my nice patriot ram:(
http://i121.photobucket.com/albums/o...TestTurboV.jpg
http://i121.photobucket.com/albums/o...ntagescore.jpg
Congrats :up:
thanx...now im new to OC, very versed technical bg though hope my voltages arent bad.Quote:
Originally Posted by KingOfsorroW
man that was some great advice you gave me also:up:Quote:
Originally Posted by KingOfsorroW
thanx to you i tested the memory 1 by 1 and was ready when i called for support.:up::up:
thanx :up::up::up:
This is what happends when you put too much load on the motherboard that doesn't have an additional onboard molex connector:
http://img524.imageshack.us/img524/594/dscn1127b.th.jpg
Uploaded with ImageShack.us
http://img25.imageshack.us/img25/7837/dscn1147n.th.jpg
Uploaded with ImageShack.us
So stay away this motherboard if you intend to build a really powerful system. Or look for a vesion with a molex connector. It is clear now why it was added in the second revision of it. :mad:
wow that looks very nice and toasty, i will keep that in mind.:shocked:Quote:
Originally Posted by KingOfsorroW
Thats only going to affect you if you have three or more beefy gfx cards. e.g. Fermi
I have first gen p6t7.
It has 6core i7-980x on 4.3ghz - HT on,
air cooling (prolimatech megahalem)
quadro fx5800, 2 x areca 1680ix-12, 10gb lan, highpoint 4322, 24 internal wd re3 1tb, and 12 external wd re4.
working mostly on full load 100% for at least 16 hours per day.
for the last 10 months.
Not a single crash.
RHEL 5.3 64bit and Windows 7 64bit
I should add that ASUS contacted me later I received a really great support from them, unfortunately it fell on deph ears of the shop where I bouhgt it in Moscow and this case is still going on, and unfortunately I made an order for a new motherbpard from evga by that time, which arrived dead...obviously. Now managing two rmas for both boards...
The p6t7 was really trouble-free for all the 1,5years of use. Except for some short period of time when it was unclear what bios setting should be set to in order to avoid the cpu throttling. And until I plugged in the 3rd fermi card of course... Asus still the best in my opnion.
never had any problems with asus, great products and support.:up::up:Quote:
Originally Posted by KingOfsorroW
KingOfsorroW remember i told you so, those evga boards you have to sing a hail marry before you get a working one.
http://i121.photobucket.com/albums/o...2-20211620.jpg
Yeah, singing a hail merry is what I do now ))
Hey guys
I have been over clocking my 920 D0 (3845B223) on the p6t7 ws supercomputer board, and I got to 4,134 Mhz stable but now on 4,2 GHz Prime95 crashes.
here is a BIOS template:
CPU Ratio: 21
Speedstep: disabled
BCLK: 200
PCIE frequency: 100
DRAM frequency: DDR3-1603MHz
UCLK frequency: auto
QPI link data rate: auto
CPU voltage: 1.30625
CPU PLL voltage: auto
QPI/DRAM core voltage: auto
IOH voltage: auto
IOH PCIE voltage: auto
ICH voltage: auto
ICH PCIE voltage: auto
DRAM bus voltage: auto
DRAM data REF voltage on CHa: auto
DRAM CTRL REFvoltage on CHA: auto
DRAM DATA REF voltage on CHB: auto
DRAM CRTL REF voltage on CHB: auto
DRAM DATA REF voltage on CHC: auto
DRAM CRTL REF voltage on CHB: auto
LOAD line calibration: auto
CPU differential amplitude: auto
CPU clock skew: auto
CPU spread spectrum: auto
IOH clock skew
PCIE spread spectrum: auto
I am a bit of a noob when it comes to over clocking so my question is, what should be done?
specs of my system are in my sig
It would be awesome if you guys could help me get this thing stable :)
pleeease? :worship:
Try enabling Load line calibration.
Trying that now, thank you :)Quote:
Originally Posted by Shiranui Gen-An
Edit: I can get it to 21X199 with Load Line enabled. 200 just seems to be a magic line, any more tips?
I am getting better temps with Load Line enabled BTW :)
21x200BCLK shouldn't be hard at all with a 3845B D0 chip. Try setting QPI/DRAM voltage (which is better known as VTT, I don't know why Asus uses that stupid term for it) to 1.30~1.35v.
I have core voltage at 1.320 and qpi/dram at 1.35. prime 95 still crashes instantly.
Give it some IOH voltage and try 1.2qpi/dram with 1.5v dram if you can.Quote:
Originally Posted by melle.k
IOH at something like 1.24?Quote:
Originally Posted by Roadhog
And by dram do you mean dram bus voltage?
(By now you might be wondering what a noob like me does whit a mobo like this :rofl:)
I appreciate all the help guys :up:
Try this...Quote:
Originally Posted by melle.k
Also, make sure your memory timings are being set right in the bios. Not just the 9-9-9-24 whatever ones, but all the rest of bank 1.Quote:
CPU Ratio: 21
Speedstep: disabled
BCLK: 201
PCIE frequency: 100
DRAM frequency: DDR3-1603MHz
UCLK frequency: lowest setting
QPI link data rate: lowest setting
CPU voltage: 1.35
CPU PLL voltage: 1.82
QPI/DRAM core voltage: 1.2
IOH voltage: 1.25
IOH PCIE voltage: lowest setting
ICH voltage: lowest setting
ICH PCIE voltage: lowest setting
DRAM bus voltage: 1.65 or whatever your memory is rated at
DRAM data REF voltage on CHa: auto
DRAM CTRL REFvoltage on CHA: auto
DRAM DATA REF voltage on CHB: auto
DRAM CRTL REF voltage on CHB: auto
DRAM DATA REF voltage on CHC: auto
DRAM CRTL REF voltage on CHB: auto
LOAD line calibration: enabled
CPU differential amplitude: auto
CPU clock skew: auto
CPU spread spectrum: auto
IOH clock skew: auto
PCIE spread spectrum: auto
I copied everything you said to my bios, but one thing is not possible.The DRAM bus voltage cant go at 1.65 (which is the rating of my RAM kit) only 1.64 or 1.66 and with 1.66 I get a warning that I might damage my CPU. Which of the rating shall I choose?
And about the memory timing, I have no clue about what I should do :cow:
Edit: the voltages did the trick though, now running smooth on 4,217Mhz :D
Thanks a million for the help Roadhog and Shiranui Gen-An :bows:
1.64 and 1.66 doesn't matter. Whatever one you want.
Ah ok, thanks again :)Quote:
Originally Posted by Roadhog
See if you can run the memory at 1.5v.Quote:
Originally Posted by melle.k
I just got a blue screen again after 1 hour of Prime 95, set the DRAM bus voltage at 1.66 (was 1.64) now and raised the QPI a tiny bit. hope this will keep running. Would running the memory at 1.5v help keep stable?Quote:
Originally Posted by Roadhog
No, but you're supposed to keep the memory within .5v of the qpi. You might need more CPU voltage.Quote:
Originally Posted by melle.k
So if I understand every thing correctly, the DRAM bus voltage needs to be within .5v of the QPI/DRAM core voltage? and I need to crank the CPU voltage from 1,35 to a bit higher?Quote:
Originally Posted by Roadhog
thats just a myth lolQuote:
Originally Posted by melle.k
intels way of scarin us
i really doubt if intel even knows
bump your IOH PCIE to 1.55-1.6
you might wanna bump the qpi pll to 1.3-1.36 for higher ram clocks
pcie frequency 102
set cpu to 1.4 and blk to 210 for 4.4ghz dont be afraid as long as temps dont get too high
if you on air you can turn ht off for lower temps
that sounds good :D but I don't have a QPI PLL here in my BIOS and I just started in the OC world so I don't know which one you mean.Quote:
Originally Posted by zsamz_
QPI/DRAM:cool:Quote:
Originally Posted by melle.k
and what rev is you ram
thanks :)Quote:
Originally Posted by zsamz_
its rated for 1866 but my bios says 1683 now
is it really?Quote:
Originally Posted by zsamz_
testing at 4,4 now, multi goes back to 20 when temps reach 70 though.
Yeah, mine does that also so I just settled for 4ghz.Quote:
Originally Posted by melle.k
use bios 804Quote:
Originally Posted by melle.k
and use the trick to stop the throtling on page 19-20
multi will saty @ 21x:yepp:
Do you risk melting the 24 pin connector if you have a revision 1 board?Quote:
Originally Posted by zsamz_
nopeQuote:
Originally Posted by Roadhog
its multiple powery hungry cards doin it
i did tri-sli 3 gtx285 a while ago no problems
guess i shouldn't try it with my 6970's then lol.Quote:
Originally Posted by zsamz_
I've had it at 4.5ghz on an older bios though. Just wasn't worth all the extra heat for negligible gains imo.
it doesnt overclock as well with 2 5970sQuote:
Originally Posted by Roadhog
but 2 6970s should be np
It seems to be working :D
Edit: nope not stable, 3Dmark 11 crashes after test 2 (unknown crash)
http://i846.photobucket.com/albums/a.../Knipsel-2.png
up volts to 1.42 then go up by 0.02 till its stableQuote:
Originally Posted by melle.k
ennable llc
I did that until I was up to 1.45 v and I only got more unstable. so now back at 1.4 v and is at least running longer then 1.45
i currently run 4,200mhz this are my settings.
i tried +4.4 before but it needs a big jump from ~1.35v to above ~1.45v to start becoming stable @ 4.4 and above and temps reach 83/85C so i put the idea to rest.
http://i121.photobucket.com/albums/o...GPU4200mhz.jpg
New bios is out 1001.
support for Core i7 990x
I will have 990x next week most likely so i will try to post some OC results
cheers
i am surprised as how this motherboard is obscured by the industry........
totally love this board has never let me down....
ASUS GTX 480 SLI'd 36.5K 3DMARK VANTAGE | i7 970 @ 4,230MHZ | Swiftech XT BLOCK/PUMP...ETC..ETC
and all good:yepp:
http://i121.photobucket.com/albums/o...th_36KMARK.jpg
8m.54sec 32mb Spi32 run
http://i121.photobucket.com/albums/o...e/th_spi32.jpg