What voltage do you have at 3.5? Your temps seem way high even for high vcore. My temps really start to climb over about 1.45 vcore.
Printable View
What voltage do you have at 3.5? Your temps seem way high even for high vcore. My temps really start to climb over about 1.45 vcore.
Quote:
Originally Posted by nicepun
Yeah, but I was able to run at 9x446 at 1.44v and I'm 3d06 and gaming stable at 1.47real (I have neither the patience nor the time for long periods of Prime - 5 mins and then its game time) and previous experience with C2D's would suggest it's probably got another couple of hunderd MHZs if I can give it enough volts. I'm cooling limited right now as my vapo is stock with 0 mods for the quad load - I ran my last Xeon 3060 at 1.68v and 4365mhz and this Quads low voltage behaviour matches what that chip did. We'll see in a few weeks - If I can swing a new mobo I'll need something Penryn ready anyway:D
here's my data from a previous post in this thread:Quote:
Originally Posted by [XC] 4X4N
-------------
Here's my data so far:
L724A760
VID = 1.2875V
3.2GHz w/ +50mV - 8 hrs stable
3.3GHz w/ +62.5mv - 2 hrs prime stable.
3.4Ghz w/ +125mv - 8 hrs stable - 35° idle, 70° load 8 hrs
3.5GHz w/ +175mV - 6 mins prime crash... temps 75°-80°
3.5GHz w/ +200mV - 5 mins prime stable, no crash... temps ran away so i shut down (~90°C)
3.6 w/+200mV - 1 core fails prime 1-2mins in... i don't think i'll be going past 3.6 or +200mV unless i get better cooling.
---------------
primed last night @ 3420MHz for 8+ hours w/ C1E enabled - all stable. Temps now hover around 63,64,65°C.
edit: priming right now, remotely, via VNC while at work using clkgen. 30 mins stable so far at 383 x 9 = 3447 w/+125mV :) (1.4125V in bios, 1.36V under load read by smart guardian)
:)Looks like I will find out...it will be nice to use something other than phase for a change. Can you point me to a post showing a stable 3.6 and 4.0 on air?Quote:
Originally Posted by nugzo
Its one thing to bench in the 3.6 to 4.0 range on air and something entirely different to run stable at 3.6 to 4.0 on air.
Also what week were the chips in the overclock's you were referencing and do you know the vCore and temps reported...better yet just point me to the link.
Thanks again.... :up:
http://www.xtremesystems.org/forums/...d.php?t=155921Quote:
Originally Posted by sdumper
http://www.xtremesystems.org/forums/...2&postcount=30
http://www.xtremesystems.org/forums/...0&postcount=50
http://www.xtremesystems.org/forums/...&postcount=107
http://www.xtremesystems.org/forums/...d.php?t=154758
i didn't go back through this thread, but I'm sure there are a few here also.
Im still a bit skeptical but will see for myself on Friday and thanks for all the links I owe you one!!Quote:
Originally Posted by nugzo
Let me know if you find one(3.6ghz+ stable on air). From what i have seen the high clocks on air usually are under unreal circumstances like an unusually cold room or an air conditioner pointed at an open bench setup. I'm not talking benches or 2 hours prime stable either. Talking 8 hours prime stable on air in a room that is approx 65-70F.
are you sure thats phase?Quote:
Originally Posted by sdumper
TEST SETUP:
* Q6600 G0 L727Axxx
* XFX 680i SLI Motherboard
* 2GB Corsair PC10000 DDR2 RAM
* PCI Geforce MX400 GFX Card
* 60GB IDE Drive
* Corsair HX620 PSU >> 620W
* Thermaltake Big Typhoon air cooler
* Windows XP Pro Lite
maybe i didn't read far enough, i just went by what OP posted. Anyway, good luck on yours. My first 2 sucked, my second 2 are completely stable at 3700mhz on water. 1.45v 55-60c loaded.
Anybody tried one of theses G0's on the MSI P6N diamond 680i? I'll have the MSI tomorrow and will post results.
Bah.... Damn work restrictions i can't see pics in that post from Dinos22. Does it show reasonable temps on an 8+ hour prime load? I'll check when i get home.
Must not be reading this thread then or you missed my post.Quote:
Originally Posted by sdumper
This ran & still running folding at home 24/7 stable for the past 4 days.
I've seen times when passing prime or Orthos for 24hrs that folding wouldn't run stable.
So if you want a real test use folding for a stability test, if it can fold its stable.
http://www.xtremesystems.org/forums/...&postcount=614
Also Orthos can fully stress all 4 cores if you know how to to set it up.
http://www.xtremesystems.org/forums/...&postcount=529
Here is a shot after 32 hours full load & the paste "AS5" has settled some.
Hmmmm..... OK. If a system can pass prime95 but fail folding then you should have no problem posting 8 hours of prime95 then right?
Not one of the links posted shows 8 hours prime stable on 'normal' air cooling. I don't feel so bad about my modest clock anymore. I guess what one person considers stable isn't necessarily stable for another. To each his own.
Please don't take this as an argument. I'm not trying to start anything. sdumper asked for proof of 3.6-4.0 STABLE on air cooling. I have yet to see the proof from the links provided.
Stable is an opinion. Speaking only for myself, I don't care if I can't do more than a minute of priming, as long as I don't have any crashes (random or otherwise) during gaming and my pc otherwise runs solidly.
Quote:
Originally Posted by GripS
IMHO if it is not 8 hour SP95 stable then it is a joke because your just waiting to corrupt your os
ive seen a couple of dudes who claim stable but cant post any prime 95 shots because it just isnt stable so it crashes
bench stable is diffrent though most of my benches are verry unstable but aslong as i can run the bench mark im happy
my 24/7 settings are always ~ 3 day F@H stable
To update this, I did place the processor into my brother's computer to test it. He has a P35-DS3R and an AC7Pro. It booted right up at 333x9, and 375x8, I didn't want to push it a lot farther cause I don't know what his ram can do. Both were with 1.3125V.Quote:
Originally Posted by whitebrow
I checked the settings on mine (P5B-E), I have the FSB termination voltage @ Auto, C1E disabled, CPUID disabled, Vanderpool enabled, CPU TM enabled, Execute Bit enabled, and PECI enabled. I am running 310x9 @ 1.3V in bios.
I'm going to get on the phone with Asus tomorrow to get their take. I'm hoping they're going to tell me that this is something I can fix with a Bios update, otherwise I'll be jumping on the X38 train in a few months...
Any other thoughts?
ill be jumping on that train also the asus p35 is a good mobo but sata sucks!
Well if I wanted to waste 8 hours of my folding time I would redo the test with Orthos & this time take screen shots. I don't use prime95 much, have never like its layout or the way you can't tell how long it been running as easy as Orthos. Like I said I have had it fail F@H after running more than 8hrs of prime95 or Orthos so I don't waste a screen shot on those two very often except in just to show what the temps run at after 15 or 20 mins. of there load.
I have a Xeon X3210 coming & I just moved the Q6600 over to my AB9 QuadGT for testing is another reason. I had hoped for a better OC in the GT but at this point looks like 9x390 is the max stable OC. Only more testing & maybe a backwards flash to beta 13.b04 will help get more. The GT is on final 15 bios at the moment.
Once the X3210 has been tested & I find witch CPU I like the best for the IP35 Pro IF its the Q6600 G0 then I'll put it back, then I may do the 8 hour test just so you can maybe be happy with my results. ;) Not that I really care. lol :p:
that is good!
some people mistake bench stable with F@H/Prime stable
that is odd f@h failed when p95 is 8+
did the ambient temp change? maby a littel more v core and youll have it
Quote:
Originally Posted by TheWolf
the QuadGT is running the 1066 strap all the time that is why it is limiting the overclock on the Q6600...you are getting exactly what i got on the QuadGT with my Q6600 B3....it's faster in superPI than my current IP35 Pro...now, overall system performance on IP35 Pro is now on par with the QuadGT results....i used PCMark 2005 to compare the two boards.
need help,keep get hardware failure about 3min running everest ultima.any sugg.
the only way i know of is increase Vcore...BUT...that is based on the other voltages are sufficient to support ram and chipset function.Quote:
Originally Posted by scorpion
make sure your ram is operating ina 1:1 ratio
A little more tweaking/voltage fix the problem, this was at the 3.6 OC in the above pic. I hit up to 3.990 with this cpu played with 3.8 & 3.7 but was tiered & never got a stable setting prime or Orthos wise. I'm happy for now with 3.6 & may try for more later on.Quote:
Originally Posted by billdavis
I'll have to re-check but the 1333 bus I'm sure was added for the GT in one of the later beta bios. So someone said. There is no added option in bios to manually select the 1333 bus. I asked in our mod section at the Abit Forum if the 1333 bus option could be added as selectable option/unhidden. No answer from the bios team yet or Abit. I'll have to double check were I got the 1333 bus info from. Can't seem to find the post at this time.Quote:
Originally Posted by Ace-a-Rue
Quote:
Originally Posted by scorpion
Going by your sig & memory timing I would say loosen the ram timing to 5-5-5-15 "stay at 1:1" to take it out of the picture & then test to see if it is stable.
You didn't give us much info to go on so that's about all the help I can give at this time.
I had heard this before bios 14, but here is the info.
BIOS release V14 for AB9 QuadGT
Release information:
[ENGLISH]
1. Support FSB1333 CPUs
2. BIOS compiled date: 06/21/2007
I didn't do much testing with 14 because of some reported problem with that bios. Hmm maybe I should since the problem was to do with ICH8R AHCI function & fixed in 15. You would think the 1333 was carried over into the 15 bios, but could have been part of the problem with 14 & removed from 15. hmmm :cool: Got me thinking... :) ;) :eek:
BIOS release V15 for AB9 QuadGT
Release information:
[ENGLISH]
1. The ICH8R AHCI function now works as intended.
2. BIOS compiled date: 07/18/2007
I actually backflashed from bios 15 to 14 and it didn't improve anything. Just food for thought.
Quote:
Originally Posted by Gunlance
Thanks for the input. I'll probably still test it. But I seem to have better OC among other thing with 13.Bxx versions. So I will retest with it soon.
might be 1:1...on the IP35 Pro the 1333 bus is armed when i use either 1:1 or 1:1.2Quote:
Originally Posted by TheWolf
Wouldn't happen to have a link to that bios would you? I definately want to give it a shot. The max stable I've foudn thus far with 15 is 400x9 with 1.5 set in bios.Quote:
Originally Posted by TheWolf
that is a 1333 fsb CPU not a 1333 memory strap!..big difference!...a 1066 fsb CPU working with a 1333 memory strap will give you a higher FSB...that is not the case with the QuadGT which was designed to operate on the 1066 memory strap...maybe a "G0" cpu you could get a higher FSB because of the lower watts it operates at...a "G0" is still a 1066 fsb CPUQuote:
Originally Posted by TheWolf
Here is a link to all the beta bios for all the new board by Abit.Quote:
Originally Posted by Gunlance
Yea I mixing up the two, sorry. The N/B Strap CPU as: 1333 is what I posted about in the mods section a while back.Quote:
Originally Posted by Ace-a-Rue
Still no answer & still no selectable item in the bios. May be a chipset limitation.
Quote:
Originally Posted by TheWolf
that is strange; listing newer bios below an older bios!
The post is edited & new are added as they come out. Why would you start at 10 & count backwards when 1 is were the count starts? ;)Quote:
Originally Posted by Ace-a-Rue
Quad Prime - in place FFT
3.6GHz
P5K Deluxe
1.5V BIOS set, Damper enabled
D-Tek Fuzion + homemade loop; temperature 23° C
IDLE
http://img.photobucket.com/albums/v3...prime_idle.jpg
LOAD
http://img.photobucket.com/albums/v31/canny/prime.jpg
Here's mine.....8 hour shot :)
Swifty watercooled.
http://i16.photobucket.com/albums/b3..._A8_3.8Ghz.jpg
Ouch on water you were still hitting 57c :(Quote:
Originally Posted by bito
I keep hearing folks say these chips to 3.6/4.0 Orthos stable on air. Do you have so/so WC system or are people exaggerating on the merits of this chip on air?
Hmm regarding Orthos stressing 4 cores...thats the first I have heard that and Im hardly a noob ;)Quote:
Originally Posted by TheWolf
Anyways it would be nice to try this out so will do a little checking and see if I can set it up to stress 4 cores.
@bito...nice work man...3.8 is respectable for sure!Quote:
Originally Posted by sdumper
dont know...maybe people are cranking the ac to cool the room down to 70 degree's or below, my computer room is the hottest room in the house, older home, and lousy ac, and it stays about 79 degrees. if ya crank up the volts on these chips, the temps get up there!
so the people on air running 3.6~4.0...dont know how they did it, or they just have a fantastic cpu..imho :shrug:
For sure my watercooling setup is not the best and could be improved upon to give lower temps at the expense of noise. I do not have a/c to help.Quote:
Originally Posted by sdumper
But i have not seen too many prime stable runs at the high speeds touted as being easily achieved by the G0 Quads on air or otherwise.....mainly just SuperPi shots. I guess it still depends on the quality of the chip itself and the other components in the build.
@icon57.....thank you :)
I've come to question the temps of the air clocking guys who are hitting high OCs. I've got a PA120.3, a lapped IHS, and a DDC-2 with petras top JUST for my quad. I STILL hit the mid 60s range when pushing the volts. So either they have the greatest quads around, or my quad is a reconditioned space heater :)
why list the old one first...it would seem better to list the most current and then older versions in descending order.Quote:
Originally Posted by TheWolf
Quote:
Originally Posted by G H Z
i'm a little surprised that you need that much vcore for 3.6 Ghz overclock...i'm at 3.8Ghz and only using 1.45v actual vcore at idle conditions...mine drops down to about 1.41v under load
what kind of cooler are you using?
anyone experiencing any wierdness with coretemp and their quads?
i just noticed that it seems my cores swapped around. I assume this must be a glitch in the software, as i don't see how the hotter cores would change core #'s.
also, do other peoples vcore do the same thing with vdroop? where by upping the voltage a decent amount in the bios yeilds the same vcore under load. @ VID + 125mV, my vcore was aprox 1.36X... @ +137.5mV it goes to 1.376V under load, and @ +150mV it stays the same, at 1.376V.
http://img250.imageshack.us/img250/326/myquadgr1.th.jpg
I need the exact same settings to reach 3.6 Same as G H ZQuote:
Originally Posted by Ace-a-Rue
DL Orthos unpak in a folder, then make a second folder & unpak a sceond copy. Place a Icon on the desk top for each "if you like" open both. Now before starting each open task manager locate ORTHOS right click on it & go to Set Affinity, click, you will see the second pic. below, uncheck CPU1 & CPU3 click ok, then do the same with the other run of ORTHOS but this time uncheck CPU0 & CPU2. Close task mgr & start one run & wait about 5 second for it to start working & then start the second run. That's it.Quote:
Originally Posted by sdumper
Boy some prople are just to hard to please. :rolleyes: The post in this forum doesn't start at 1000 & count backwards & who knows there maybe more than 1000 post. yada yada yada! :rolleyes: :p:Quote:
Originally Posted by Ace-a-Rue
the difference could be that i run everest stability stress program and that puts all four cores to 100% while stressing FPU, Cache and system memory...if it passes with that, i am happy.Quote:
Originally Posted by Gunlance
Quote:
Originally Posted by Gunlance
i hear ya!
i know my exos2 is not the greatest wc setup, but it does the trick.
on load, with a high oc i hit the 60's no prob, and if i start cranking up volts it will go into the 70's on full load. and my GO will do 3.2 on 1.4v, if i want higher Mhz, i need to up the volt to 1.5, and above
so, i have to either keep my ac blowing so cold, and/or add some more fans for more airflow into my case....i dont see either of those things happening...(well maybe some carefully placed fans)
Quote:
Originally Posted by TheWolf
man!...you must be the guy who is updating abit's bios page.:) ..so defensive!:shrug:
Upto 3.60GHz that is still under max Vcore (1.35 according to Intel) of 1.34V load taken by DMM :D
http://img515.imageshack.us/img515/4...e360ghzfe2.jpg
Nope but I could be, but I stiill can't see why some would wine about starting at A & going to Z, like its going to kill you to look down a list a little ways. :rolleyes:Quote:
Originally Posted by Ace-a-Rue
This is my effort with a L720B013
[IMG][img=http://img405.imageshack.us/img405/2385/luckydh8.png][/IMG]
the link would not show the image
Quote:
Originally Posted by DFI pit bull
fixed the link for you
http://img405.imageshack.us/img405/2385/luckydh8.png
how much vcore are you talking about?Quote:
Originally Posted by Gunlance
Quote:
Originally Posted by DFI pit bull
WOOT! WOOT...NICE!:up:
Awesome thanks man!!!Quote:
Originally Posted by TheWolf
For how long?
Thanks Wingz.
I think i'll host on my brothers website from now on.
I really love this new chip, have had it at 4.52 ghz but did'nt get a screenshot.
just started playing with mine, batch L726A951
1h prime shot @ 3800 /1.5 (1.53 set in bios)
http://img227.imageshack.us/img227/1018/3800jr1.th.jpg
At roughly 1.52 in bios(1.45v - 1.46v in windows under load) I'm hitting 63 on hottest cores and 61 on the other 2.Quote:
Originally Posted by Ace-a-Rue
Can someone with a quad clocked to to 3.6ghz using 9x400 with 1:1 memory do me a favor and post up a superpi mod 1m time? I wanna see if it really is a strap issue with the quad gt. If it is I'll be lookin for a new mobo to get this thing higher with less vcore.
I just got my Abit IP35-Pro and started out of the gate 600MHz higher than my P5N32-E SLI was capable of. I just had a single worker thread in Prime95 go down, but the rest are running strong in an in place test. Anyone know what that could mean?
http://i158.photobucket.com/albums/t...CT_CPUZ3.3.jpg
It means you are not stable. Probably need to bump up the vcore a little.
EDIT: Gunlance, here is a quick 1m for ya :) Also, a prime shot for temps. Ambient is around 78.
http://img503.imageshack.us/img503/4...001mwt0.th.jpghttp://img159.imageshack.us/img159/6...0prmxo2.th.jpg
^^ Thanks.
I can do that too....but I'm not going to for 8 hours :DQuote:
Originally Posted by bito
1.525V BIOS set
Blend
http://img.photobucket.com/albums/v3...ime_3.8GHz.jpg
I probably don't. I am a 3D bencher, this is the first time I've run Prime in years and didn't bother to check for lowest passable vcore. This test was to check the capacity of this loop and if you ask me it's very good for a quiet setup - 6" x 8" heatercore with two 53CFM 120mm fans non-ducted.Quote:
Originally Posted by Ace-a-Rue
Post your results and we'll compare ;)Quote:
Originally Posted by Ace-a-Rue
Blend isn't going to push it that hard. Try small ffts, that'll warm it up some.
This is on the DFI P965-S. (you didn't specify chipset). I just ran this for your request. I can't run prime95 at this voltage without going past 70C.Quote:
Originally Posted by Gunlance
http://img3.freeimagehosting.net/upl...3dc3a0d257.jpg
I'm been playing around with mine some, a L721A814 G0.
I'm at 3.3Ghz ( 366 X 9) Memory at 549Mhz 5-4-4-12.
http://img210.imageshack.us/img210/3...hstablelw6.jpg
It also feels like I have a good 200-300 mhz left on my current set up, maybe up to 3.6-3.7 24/7 stable after I lap my tuniq tower.
My 8hr run was small fft's for max. heat load.....:)Quote:
Originally Posted by fornowagain
Hey!
anyone here with a 680i and Q6600? I upgraded my bios to P30, and intead of stabilize my system, it did worst, now, I can not boot vista, if im at 2800, the best i can do is 2700, before i flash my bios, I was at 3400 at 1.45 on bios, and the system rebooted after 10 minutes of Prime on all 4 cores!
Any Help appreciated
PS: could it be the PSU? I have a 550W (2x12v - 20A)
Thanks!
THANKS! :up:Quote:
Originally Posted by GripS
That vcore being reported is that in bios or under load? (1.45v)
Thanks again.
BIOS. Load is 1.42-1.43.
My does she Goooooo L719B271 1.43vcore No tweaks yet, just put her in and let her rip, more to come:D
http://img210.imageshack.us/img210/2228/quad6de4.jpg
Wow! Can you post the same screenshot where it show's correct voltage? Cooling?
If true it's definitely very impressive running on P965 chipset. :up:
Today I pushed it further
http://img410.imageshack.us/img410/5173/46bw6.jpg
love this chip, i think it will go even higher with this volts -> 1.53V (1.504 after vdroop) / 1.55V in bios ..
http://img213.imageshack.us/img213/4371/38003hmj0.jpg
hmm, CPU-Z 1.41.0 - new beta?
Looks like SS to me based off his sig...Quote:
Originally Posted by GripS
Heres mine that I got yesterday from ClubIT. I can prolly push it higher with more volts as I have the Zalman 9700NT.:shocked:
Can you try dual orthos and show TAT temps?
the voltage was 1.43 as measured with Dmm, board is fully voltmodded, Cooling Single Stage evap temp unload -40C, will have a go this weekend, today it was fit and quick test, don't know why cpuz cantwork the voltageQuote:
Originally Posted by GripS
Quote:
Originally Posted by BigRigDriver
dude, you do realize you are only stressing 2 cores from what i see in your ss.
you will have to use 2 instances of orthos, or use prime95.
btw what week is it?
FYI, in place generates the most heat ;)Quote:
Originally Posted by bito
mine is realy nicely scaling up, but will not try with higher volts, at least for now..
http://img184.imageshack.us/img184/8179/3850fn0.jpg
just got my L726A343 I bought from Ebay. Very happy so far! Didn't tested a lot , just 1hr of orthos ... tested at 3200mhz (356 x 8) , auto settings for vcore (default vid is 1.2875v)... big vdrop: 1.264 to 1.232v , Giga 965 ds3... on air , intel HSF !! this cpu appears to be a keeper.
need to buy better ram , the maximum this corsair VR ram run is about 390mhz.
:)
Hey guyz...I broke my 680i....I went to switch for faster RAM & the darn thing wont post in all but the last dimm:bananal:
In place generates the most overall system heat. Small fft definitely generates more CPU heat.Quote:
Originally Posted by G H Z
My first aggressive OC since they moved from mobo jumpers to BIOS:
http://i9.tinypic.com/67d0kra.jpg
I did this in about 15 minutes. I have NOT tested for stability. Just testing to see if it will post and the temps. So I guess the REAL temps are
CPU: 57*C
Cores: 51*C,51*C, 50*C, 50*C
I am using a triple rad w/ not the best fans (waiting for san ace's to arrive) and a D-tek fuzion.
Quote:
Originally Posted by periquitos
forgot to mention that the maximum temp in coretemp was 61C , while my room is at 25C.
OFFTOPIC: do you know guys , will I need to reinstall the OS and all the APPS after upgrading from a dual to a quad core? I was testing in 3dsmax8 and only exact 50% cpu was working while rendering. Tested in After Effects 7 with the same results. These softwares are working like they was with the dual core cpu !!!
not much software is coded for quads yet...
True but you know what thats the same thing we were saying when dual cores came out and some folks were still running single cores :)Quote:
Originally Posted by icon57
I believe that 3dsmax is , just don't know whats happening!
make sure all 4 cores are enabled in bios. Sometimes with my 680i when switching between dual and quad core, sometimes cores 3 and 4 will be disabled in bios.Quote:
Originally Posted by periquitos
not much no, BUT, 3dsmax and maya for instance can handle ALOT more than 4 coresQuote:
Originally Posted by icon57
Quote:
Originally Posted by BigRigDriver
hmm at 1.45 and also a clubit purchase week 23 at 3.6 folding all 4 cores im hitting 67c.
Oh yeah and this is with a high speed fan blowing across a V1 with AC set at 70f...:down:
If your running air that's pretty damn good.Quote:
Originally Posted by sdumper
here is my best overclock 3.9 Ghz and everest stable...voltages are a bit high for my liking...with way less voltage for vcore, cpu vtt, mch and ichio 1.5v, i can live nicely at 3.84Ghz.
currently outside of the case...probably the temps will increase about 3C for cpu temp after putting inside the case...but...on the other hand, the vcore will be 0.04v less.
Awesome Ace. I need to get me a P35 board
why?
Im using a p5k deluxe and it can go high but try stressing 4 cores ... and no hype guys please....I mean stressing 4 cores at 1.45 on air cooling not water and a modest oc like 3.6 or so.
Im a little angry right now because i sold a perfectly good B3 because i bought into all the G0 hype and am very disappointed. I wanted a solid 24x7 air folder at 3.6
While you are only full loading the CPU at 50% with one ortho :DQuote:
Originally Posted by BigRigDriver
try prime95 25.3, it supports multi-threads :D