Cheers Dude!... im benching now at 4ghz, however.....my ram is running single channel...im running g-skill ripjaw ddr3 1600mhz ram....any ideas why???
cheers
gordy!
Printable View
Cheers Dude!... im benching now at 4ghz, however.....my ram is running single channel...im running g-skill ripjaw ddr3 1600mhz ram....any ideas why???
cheers
gordy!
hey guys, not sure where to post but if anybody can help that would be great.
i have a gigabyte p55-ud2 motherboard, and have corsair 4gb ddr3 1600mhz ram. but i dont think its supported by this motherboard. i cant return the ram now, the ram is showing up as 1066mhz now, is there a way to make it work at 1600?
the official site says it doesnt support it but other sites say it does...
Overclock it. On my mb there isn't enough memory dividers to make it run at rated speed. The only way of doing so is buy raising the bclk, which in turn will increase the memory speed. Not sure about your mb, but I'd assume that it's similar.Quote:
Originally Posted by theob42
yes but not sure how to do that...
Stupid question but do you have them both in the red slots? The ram slots alternate red and black, each color is a channel pair. Anyway, mine run dual channel just fine in the red slots. :up:Quote:
Originally Posted by ninjagordy
Mike is right, you're not seeing the 1600 speed because you need to OC to reach those speeds. Look under your RAM settings there will be different speeds listed. In the past you had to press ctrl+f1 to get into the tweaking options on giga boards.Quote:
Originally Posted by mikecdm
guys, just a question: what is the SAFE voltages for:
vCC, vTT, vPLL and IMC? thanks
I've been really happy with the MSI P55 GD65, with a couple of major caveats......
1. I have a BFG 850W PSU powering the system, and until doing some research could not figure out why the OCGenie refused to work at all. The system would boot, power down a few times, then refuse to boot. Evidently, my PSU is one of the handful that refuse to work at all with the GD65.
2. The GD65 suffers from horrendous VDroop. If you set the BIOS to 1.4V with Loadline Calibration "Disabled" (I say "Disabled because the BIOS is screwed up and "Disabled is actually "Enabled") the voltage drops to 1.31-1.32 before a BSOD. With the Loadline Calibration "Enabled" (or "Disabled" in the BIOS and the VCore set to 1.33V, the voltage climbs to 1.42-1.456V. I've tried everything from 1.33 to 1.4V and the VCore still spike tremendously. In my opinion, this is an unsafe fluctuation.
Anyone else seeing similar results?
All of you p55 OCing nerds should .. Check your sockets!
" ... The overriding concern is that we have damaged every motherboard in our possession for the overclocking P55 shootout as well as two very expensive i7/870 processors. " Check your sockets!
I read a little bit about that problem elsewhere, too...but it seems it only happens in REAL overclocking situations, with sub-zero-cooling and extreme voltages...I don't think someone with an aircooled CPU running at 4GHz with under 1,4V should get into trouble...Quote:
Originally Posted by Sam_oslo
Yes, the immediate danger seams to be threating the nerds and extreme OCing on subzero, but there is a bit more into this history. The long term effect of a high 24/7 OC seams to be under question too, as AnandTech put in this way:Quote:
Originally Posted by GrossmeisterB
"We have not had any problems with air or water cooling overclocking up to 4.3GHz, although we do have a i5/750 that has developed a few dark pads after a thousand hours or so of constant overclocking. " - Link
It is a good idea to keep an eye on how this article will be developing over time, anyways.
Drop your cpu multi down to 15xQuote:
Originally Posted by theob42
Raise your bclk to 200
Put your memory mulri to 8x
That should give you 3Ghz on the cpu and 1600Mhz memory
Check your voltages and DONT put them on "auto"
Good question. I'm now running more vcore then I care for @ 4.2 but I'm still just exploring with this combo. I'll most likely settle for 4.0 with the lowest voltages I can manage.Quote:
Originally Posted by KURTZ
http://i695.photobucket.com/albums/v...EVGAp55FTW.png
Right. This is so far caused by overpressure by LN2 pots and such. Not really worried at all.Quote:
Originally Posted by Sam_oslo
hey guys,
right, i re-installed my ram, and cooler today, got my temps to a comfortable level now but the single channel scenario still persists.
is there any way iv switched something of on the board??
maxiumus III gene, 2 x 2gb ripjaw 1600mhz ddr3 and an i5.... (thats my rig)
cant understand it...tried every socket and still single channel!!!
all help appreciated...
:up:
I've putted my system together and played a little with it.
First of all I tried the OC Genie button on my GD65, it will set the CPU at around 3300Mhz with an 200Mhz baseclock.
Then I googled a Core i5 overclocking guide and used the values mentioned there (1.4v vcore 1.4v VTT 1.65v DDR and 1.9v PCH)
I turned down the mem ratio and the QPI ratio.
After setting the CPU to 3,7Ghz I went into Windows. After some Superpi runs I started playing with the baseclock buttons. Here is my final Superpi result at settings above.
http://i37.tinypic.com/v6rqch.jpg
How much Vcore and VTT is safe to use 24/7?
hey dude... i personaly wouldnt go much higher than what your at now, im sure one of the pro's will give u an exact figure. hows the load temps at that speed??
cheers dude
Gordy :up:
I am suffering a massive vdroop on my P55-GD65. When I set the vcore to 1.46v in Bios, it's idle around 1.44v and when I stress the CPU with Prime95 it will drop all the way down to 1.34v according CPU-Z. Is such a drop normal and are there any solutions? I can't get my PC prime stable @ 4Ghz this way. Currently it max lays around 3750Mhz.
I posted this the other day...... switch the Loadline Calibration to "Disabled", and drop your VCore to 1.34-1.35 and see if that helps.Quote:
I've been really happy with the MSI P55 GD65, with a couple of major caveats......
1. I have a BFG 850W PSU powering the system, and until doing some research could not figure out why the OCGenie refused to work at all. The system would boot, power down a few times, then refuse to boot. Evidently, my PSU is one of the handful that refuse to work at all with the GD65.
2. The GD65 suffers from horrendous VDroop. If you set the BIOS to 1.4V with Loadline Calibration "Disabled" (I say "Disabled because the BIOS is screwed up and "Disabled is actually "Enabled") the voltage drops to 1.31-1.32 before a BSOD. With the Loadline Calibration "Enabled" (or "Disabled" in the BIOS and the VCore set to 1.33V, the voltage climbs to 1.42-1.456V. I've tried everything from 1.33 to 1.4V and the VCore still spike tremendously. In my opinion, this is an unsafe fluctuation.
I like the board, but it does a really poor job of managing the voltage fluctuations.
I'm actually having some issues too, I don't know if its because of my motherboard or processor.. apparently my i5 750 isn't prime stable unless I set my Vcore + Vtt to Auto.. I've managed to get 4ghz.. ran prime 95 small fft's for 10hrs without error.. I messed around with manual voltages and one core fails after around 5hrs..
Download Everest and Real Temp and use them to get an idea of where your voltages and temperatures are. I am pretty sure you are ok with the voltages set to "Auto", but lets figure out whats going on.Quote:
Originally Posted by Yamcha67
Well the temps are under 70C (small fft load) , and CPUZ + Everest say the voltages are 1.43V.. but for me cpuz and everest don't seem to detect the voltages correctly, because when I had my core volt at 1.35V in bios it would read it at 1.4ish.. and when I had my volts at 1.4 cpuz/everest would read it at 1.5.. I also have load line calibration disabled.. should I have it enabled?
Do you mean to tell me that none of you geniuses considered just taking out some of the bend in the lockdown arm to give the plate just a tad more tension?Quote:
Originally Posted by Sam_oslo
I would have tried that after the first mishap. Then I would have at least known if that was/wasn't the problem.
Looks like most of them were loose or misinstalled and their could have been some arcing here and there.
Hi guys!
So this is a "tip" thread also, so i have a question.
Its 8-pin (cpu power) necessary for a i5-750 cpu?
The situation :
cpu : i5-750
mobo: asus p7p55d
psu: arctic cooling fusion 550
This psu has 2x 12v rails, each 17A, and only has 4-pin for cpu power.
The motherboard has 8-pin, but 4-pins are covered by black plastic "thing".
Is it safe to have combination mentioned above and only 4-pin plugged to motherboard for powering cpu?
Also, i found this
http://img113.imageshack.us/img113/663/steker.jpg
Are there any limitations when i only have a 4-pin? How about OCing? What is the pwr consumption of i5-750?
Can someone try runing on 4-pin instead of 8-pin? (im not responsible for any damaged hardware, personal injuries etc. :up: )
edit: oh i almost forgot! how about this 4to8-pin http://www.plinkusa.net/web4-to-8.htm ?
If you have it set to "Disabled" then your voltages should not be dropping below what you have set in the BIOS, its should rise above what you have set. (Its a BIOS flaw, Loadline Calibration enabled should act like VDRoop and raise the voltage under load, instead of sagging).Quote:
Originally Posted by Yamcha67
Right now, until I find a way to measure those voltages accurately, or hear from MSI, my advice is to disable loadline calibration and run your system with the lower voltages. At 4GHz, I've found 1.33V VCore, 1,275 VTT to be the minimum stable, on air cooling it gives me temps in the mid 60s under load, and mid 30s at idle.
i5 750, GB P55M-UD4, 2x2GB Gskill Ripjaw 2000's, running 200x20 and 2Ghz dram stable with the following bios settings:
Vcore: 1,3V
LLC: enabled
VTT: 1,27V
Vdimm: controlled by XMP profile (=1,65V)
Timings: 9-9-9-27 1T as per XMP profile
QPI mult: 36x
Aftermarket Air cooler (Noctua C12P topdown), temps are in the high 60's crunching WCG.
Initially I thought I got a crappy chip but I guess it's decent enough? These Lynnfields all seem to need so much Vcore due to their on-die PCIe controllers :/
Whats a safe VTT? is 1.3 OK? I am unsure what could be holding me back, I am prime stable at 3.8
vcore 1.27 BIOS / 1.30 Windows FULL LOAD.
RAM within spec.
VTT 1.25
I seem to just be completely unable to hit 4Ghz stable, even with 1.37 vcore, has my CPU just hit the MAX? or is there something I can try to see if I can get to 4Ghz.
Thanks,
Temps are fine also, well within spec at about 58-60 FULL load using my Corsair H50.
Specs Below.
I disabled Loadline calibration on my GD65 board, so the voltage doesn't drop under load anymore, but rises somewhat.
I used the following settings:
Vcore: 1.4v (Load: 1.42v)
Vddr: 1.65v
Vtt: 1.4v
PCH: 1.9v
I used QPI divider of 16 and mem 3. Intel EIST / Turbo and C1E were turned off.
I didn't manage tot get 4Ghz stable at these settings. Even 3.85Ghz was not prime stable, generating errors after a few seconds.
I'm running 3750mhz now @ 1.28vcore and 1.2vtt and it works just fine, no errors so far in Prime.
Does anyone knows what I can check to get my CPU to around 4Ghz ?
here is my setups volts
Vcore: 1.408
LLC: enabled
VTT: 1,37V
Vdimm: 1,65V
Timings: 9-9-9-24 1T
QPI link: 3210
cooled by triple rad
full loads at around 59-61c
Quote:
Originally Posted by ArCElM
Quote:
Originally Posted by NoepZor
I had the same problem, I think we made a mistake getting an MSI board :P, anyway I was able to get 4ghz by setting the Core & VTT Voltage to Auto.. prime stable for 12hrs.. I've tried setting voltages manually and it seems one core always fails, I don't know why =/.. Auto seems to work for me tho.. you should give it a go.. hope it works for you..
v interesting, I get exactly that, one core always fails.
So VVT AND VCORE at AUTO? ive never heard of that before.
@ArCElM
Disable all your powersaving features and turbo mode...
Enable Load Line Callibration, set Vcore to 1.325V and VTT to 1.350V
Try that and tell me how it went...
Hmm I knew it was motherboard, I will also try enabling load line calibration, maybe I can get over 4ghz out of my i5
Quote:
I disabled Loadline calibration on my GD65 board, so the voltage doesn't drop under load anymore, but rises somewhat.
I'm telling you, the loadline calibration on this board is wrong, "Disabled" is enabled, and "Enabled" disables it. If you set it to "Enabled" at that low voltage, its going to sag to about 1.25-1.26 and cause a BSOD. I think that 1.35 VTT is too high, try what I explained earlier....set the VCore to 1.33V and VTT to 1.3, with Loadline Calibration DISABLED, and you should have luck.Quote:
@ArCElM
Disable all your powersaving features and turbo mode...
Enable Load Line Callibration, set Vcore to 1.325V and VTT to 1.350V
Try that and tell me how it went...
On my MSI P55 GD65, I tried enabling Load Line Calibration with Vcore + Vtt set to auto what happens is prime fails and CPUZ reads the voltage at 1.36V, where as with Load Line Calibration Disabled CPUZ reads it 1.43V and 12hrs prime stable...
Does it only make a difference with manual volts?
Yes, its the same on "Auto" or manually setting the voltage. Loadline Calibration is just another name for VDroop. When its enabled, its supposed to keep the VCore from sagging under load.....unfortunately, its configured backwards in the BIOS and is confusing a lot of people.
Whats equally troubling is the voltage control on this board (MSI GD-65), it either sags terribly, or the voltage spike when the Loadline Calibration is actually doing its job is in my opinion, very unsafe. I'm not sure that its as simple as a new BIOS revision to fix the issue either.
I did find out this @ 4000Mhz:
When I test my system @ Prime95 blend mode: Errors after a few seconds
When I test it @ Prime95 small FFT (fits in cache, almost no mem used): 1,5 hours no errors (then I stopped the test).
I think my memory or motherboard/memcontroller is resisting. When I lower the CPU speed by the multiplier top 2400Mhz, then I also get errors in Prime95 blend mode.
My mem is running @ 1200Mhz 9-9-9-24 @ 1.65v, which is within its JEDEC specs (1333 @ 8-8-8-21 @ 1.5v is possible). Maybe it is incompatible with my motherboard/cpu somehow. @ 187 bclock and around 1122Mhz DDR it's working just fine.
My chip is sweet as on decent volts upto 3.6/3.8 but 3.9 and 4.0 is just not happening.
Currently at 3.8, vcore 1.27 BIOS, 1.304 FULL LOAD.
I think my chip has just hit its max tbh and I can;t complain about 3.8 on decent vcore.
if you dont need all 4 cores try 2 cores
clocks higher and much more bandwidth ;)
I was doing a similar overclock on my X3440 and thought I would give it a try on my i5.
1.375v vcore set in BIOS with vdroop (LLC disabled) 181x21 turbo enabled (4 threads) and 181x24 (2 threads) with C-State enabled. In effect you get a 4.3Ghz dual core and a 3.8Ghz Quad. If you want to try this type of overclock I would recommend leaving C1E disabled at first to dial in your voltages for stability. With C1E enabled your voltages can drop quite low (0.988v) and will spool up on demand. You do need more vcore to stabilize the x24 multi then is required to run 4 cores at x21 multi. In my overclock at 3.8 I'm showing 1.352v/1.364v load in CPUZ. My board overvolts with C1E enabled and will show as high as 1.399v/1.411v in CPUZ while running SuperPi 1M and 32M.
http://i695.photobucket.com/albums/v...i575043ghz.png
Hello I am overcloked cooler air with a Core Contac Freezer. I'm thinking about buying the Dragon F1... Dry Ice
4.476 Ghz Air.
http://www.pctunerup.com/up/results/...906I54.476.JPG
Greetings
How much is max voltage for 24/7 on H2O?
I don't know what to set on:
PCH voltage?
Voltage on CHA/CHB?
IMC voltage?
QPI fre?
CC Skew?
I have Asus P7P55D, I5@4250MHz vcore 1.416V, PLL 1.8V, IMC 1.225V, other on AUTO (prime stable).
Max BCLK on this settings is 217 .
What can i set to get more BCLK?
Max validation:
http://valid.canardpc.com/show_oc.php?id=792231
24 multi? How?Quote:
Hello I am overcloked cooler air with a Core Contac Freezer. I'm thinking about buying the Dragon F1... Dry Ice
4.476 Ghz Air.
Turbo and C State "enabled" should give you (x24) multi for two threads. http://www.xtremesystems.org/forums/...&postcount=139Quote:
Originally Posted by cole2109
The 24X multiplier that CPU-Z reports can be a little misleading. If you could eliminate all background processes, you would be able to get a constant 24X multiplier but in reality, you can't do this. There will always be background processes that become active and will drop your multiplier on a Core i5-750 down to 21. When running any sort of load, it will cycle constantly between 24 and 21 when C-States are enabled and will drop a lot lower than that at idle.
Run a single threaded bench and in the middle of it, have a look at RealTemp. It displays the average multiplier on your hardest working core and takes into account the constant multiplier cycling that's going on inside your CPU.
At idle when CPU-Z is able to show a 24X multi, the multiplier on the hardest working thread is nowhere near that high.
http://www.fileden.com/files/2008/3/...alTempBeta.zip
Hello
disable the load line had calibbration.
cpu vtt 1.275v
cpu v core 1.418v
environment temperature 25 to 26Cē 12 am - 3 am I overclocked :( in day 30Cē 33Cē
low apologize- bad for my English.
Air
4.472 Ghz / 213 fsb x 21 / Memory 2130 Mhz / CAS 8 - 8 - 8 - 24 / SPI 2MB 20.750 Seg
http://www.pctunerup.com/up/results/...0.750SEGI5.JPG
Air
4.506 Ghz / 188 fsb x 24 / Memory 1880 Mhz / CAS 8 - 7 - 7 - 20 / SPI 2MB 20.812 Seg
http://www.pctunerup.com/up/results/...4.506GHZI5.JPG
Vinas :up:
Makes no sense to post your 8500 overclock TBH. Also please show us your load temps on your air cooler. They must be pretty high for those volts I'm thinking.Quote:
Originally Posted by CARLOS*A1
I dunno I think its just me maybe, and these forums are called 'XTREME SYSTEMS' but..
How come everybody is so happy to run above the max vcore?
1.4 vcore is the absolute max or you will damage your chip.
but I always recommend 1.35 and under for safety for the i5.
Does nobody actually know 1.4 is the MAX for these chips?
Why are people still using vcores from 2 chip generations ago and applying them to chips released 2-3 years later.
The idea of a good chip, or if you want to show off how good your chip is or your overclocking knowledge, its how far can you take your chip within the vcore limit or as low vcore as poss.
Infact there should be an E-PENIS thread created, 1.4vcore (in other words load or idle, vcore stays 1.4 or under) then everybody should post how high they got there chip to clock while remaining stable for a couple of hours on prime, (or some other decent stability test) pop out a screenshot and if your feeling kind tell everybody how you did it if its impressive.
That would be a really interesting thread.
But I only speak for myself when I say I could not give a rats ass that sombody can run superpi and 4.4Ghz when they are using more than 1.4 vcore, like wow, if I wanted to use 1.5vcore I dont rly have any doubts I could probably bust out a superpi run at 4.4 or something but how the hell does that help me.. it dont, and its not some fkn acheivement to go into the BIOS and turn up vcore.
Whats of interest to me is, how good other peoples chips are, and what they can do at CORRECT voltage and if there are any tricks to getting stability at lower vcores.
For example, if I was not stable at say 3.9 1.32 vcore, would there be something else can could effect stability other than turning up vcore coz we ALL KNOW the LESS vcore the BETTER.
WHO KNOWS I think I am just fed up of seeing ppl showing great clocks then u see somet stupid like 1.47 vcore in there cpu-z shot, and think WTF? POINTLESS and UNINTERESTING...
if I see sombody with 4.1Ghz and they are on 1.36vcore im like WOW fk me NICE CHIP how did u do that? THATS impressive and something to be shared.
Wow that's quite the rant there. Nice work avoiding the swear filters also. Can't come up with another word to express your point? Anyway...just because you find attempting to reach good clocks "pointless and uninteresting" does not mean others do. There are a lot of people who overclock because they enjoy benchmarking, using every last MHz they can squeeze out of their chips.
This is Xtreme Systems, where you will find that these people are closer to the rule than to the exception. Please refrain from insulting those that (god forbid) find such things interesting. There is a place for stability and for benching with max clocks. Can't we all just get along?
@ArCElM just 4 u :D
http://www.shrani.si/t/2v/OP/3lheLrk5/1/untitledd.jpg
This is MAX stable BCLK :down:
http://www.shrani.si/t/r/Po/4SRAH4O/4250-14-.jpg
At 214 BCLK PC freez after few minutes on Prime. 215, 216, 217 freez instantly when i run prime. But games runs OK :rolleyes:
What can i do to get higher BCLK?
Here's what the datasheet says for the Core i5-700 and Core i7-800 series:
http://www.intel.com/design/corei5/documentation.htm
http://img682.imageshack.us/img682/8...i5absolute.png
Unless I'm reading this wrong, absolute max voltage seems to be higher than 1.40 volts for these new CPUs. I guess we won't know for 6 months or so what the real maximum voltage should be for 24/7 use. By then, if an i5-750 has degraded, I'm sure there will be something new to buy like a 32nm replacement.
The maximum VID is listed as 1.40 volts but the actual maximum core voltage can safely go higher than this.
Edit: For comparison, here are the specs for the previous Core 2 8000 Series.
http://img62.imageshack.us/img62/8581/absolutecore2.png
There seems to be more core voltage wiggle room with the new Core i generation of CPUs.
The Core i7-900 series is also rated at a maximum core voltage of 1.55 volts.
cole2109
LOL awsome ;) nice chip! man I wish mine would do that, it does 3.8 stable at 1.3 but after that its not happy about much without allot more vcore.
unclewebb
quality post, and informative, although I think it was best I did NOT know that, thanks though :D I read max 1.4 somwhere and thought that was it, I MIGHT be more tempted to try 1.4vcore now where as before I was playing it safe with about 1.35 tops, cooling is decent ENOUGH its the corsair H50 which is about as good as a VERY good air cooler so, we will see, I really did want 4ghz out of this, I am a bit upset I am struggling so much.
That was uncalled-for. Anger issues much? There will be no "firing back" from me, I'm not that immature. I don't have a 750 so I'll refrain from posting screenshots to clutter this thread, but here are stable results at 3.6GHz, 3.8GHz and 4.0GHz with the setup in my sig. So you see, I'm all for stability; I was simply defending those you were deriding.
Nice find from the whitepaper unclewebb. Had no idea VTT was so limited. Hope it doesn't degrade chips too much. Shamino @ the EVGA forums seems to think up to 1.5v is safe (only quoting him b/c I haven't seen anyone else give a straight answer to that question). Hope he's right; can't push bclk too high on my setup without > 1.2v.
Quote:
Originally Posted by cole2109
Someone?
What cooling & voltages are you running? How are temps with that run?
I've done a little messing around with my i5 750 as well as some stability. I have currently settled for a daily of 4.3ghz, but today I was trying to do a little benching.
http://i75.photobucket.com/albums/i2...StableLinX.jpg
http://i75.photobucket.com/albums/i2...erpi90621m.jpg
This is INSANE, man, what board are u using? may I ask how u got this? don't get me wrong, you might just have an AMAZING chip but sometimes I cant help wonder if mine could do 4.0Ghz under 1.4vcore if I had either a different board or was using some voltage tweaks I did not know about, or anything rly.Quote:
Originally Posted by mikecdm
It does 3.8 1.28vcore, 1.304 full load, no problems but then, it cba to do anything else lol
:(
WELL I do have it at 3.9 atm, but that requires 1.36 full load. 1.32 BIOS.
Cheers.
Water cooling. vcore 1.408V, PLL 1.8V, IMC 1.225V, other on AUTOQuote:
Originally Posted by hokiealumnus
Temp. 55-58 (prime)
I'm using the evga P55 FTW. There's not much to it, just a little vcore and some vtt. It needed 1.425v VTT to boot at 4.6ghz and 1.375 vtt to boot at 4.3ghz.Quote:
Originally Posted by ArCElM
For 4.0ghz it only needs 1.26v vcore to be stable. Maybe I have a decent chip?
Those are superb temps! What MB are you running?Quote:
Originally Posted by cole2109
TBH, 214bclk stable on water (or air, for that matter) isn't bad at all. You're not temp limited, but The higher numbers (anywhere from 220-230 and above) generally come with sub-ambient cooling. So, you're not temp limited per-se, but you really might be, if that makes any sense.
must be the board or a strong chip or both.
Asus P7P55DQuote:
Those are superb temps! What MB are you running?
mikecdm:Quote:
Originally Posted by mikecdm
Looks like you have a nice chip! Can you post a screen shot of your ELeet voltage tab showing your 4.0 stable overclock voltages settings? I'm having trouble getting my i5 fully prime stable and always seem to get an error on one thread/core. Is it 1.26v load while stress testing? What batch# is your i5? More please.:)
You are not the only one struggling to get your chip stable at higher clocks with decent volts. I'm having diffulculty with both my chips. I don't mind pumping up the voltages for benching but I really like maximizing stable overclocks with low volts. It seems that with the i5/i7 once you go above 3.8 the voltage really scales... not to mention how much of a bump you need for the 200Mhz increase when going from 3.6 to 3.8! Who knows mabye it is just my two chip samples. I'd like to get another P55 board to help confirm my suspicions.Quote:
Originally Posted by ArCElM
Thanks. Had a look at some bios shots from a review; just wanted to confirm IMC voltage = VTT on other boards. I'd say try increasing that a bit and see if it helps you with higher bclk. If the chip can do it with above-ambient cooling, that's probably what will help you get there. If you haven't already, it can't hurt to raise PCIe frequency a little bit as well, if IMC voltage doesn't help.Quote:
Originally Posted by cole2109
Quote:
just wanted to confirm IMC voltage = VTT on other boards. I'd say try increasing that a bit and see if it helps you with higher bclk.
But what about this?
http://www.shrani.si/f/1O/Oh/hrnCgB0/corei5absolute.png
I think you are limited by your board and its PWM.Quote:
Originally Posted by ArCElM
And don't get too bothered by Intel's specs. They don't mean too much. Just carefully watch your temps.
Depends on how much you want the clocks. I'm impressed he made it to 214 with the little voltage he's given it so far. Unfortunately it seems that we are going to need to go out of spec to get the overclocks we want. No way around it that I know of. :shrug:Quote:
Originally Posted by cole2109
I don't have any screen shots of 4.0ghz. There is only one, but I'm not sure if it was stable or not. I didn't spend too much time at 4.0ghz, I just got it stable then kept going. I have some of 4.2ghz, 4.24, and 4.3ghz. I've never bothered to mess around with the Eleet thing, so I don't have it in any of my screen shots.Quote:
Originally Posted by PolRoger
My batch# is L924B521
@mikecdm
Can u post BIOS settings for 4300Mhz (stable)?
Thanks. You should check out the ELeet. It is very handy for monitoring and overclocking right from windows and it saves time rebooting when your fine tuning your overclock.Quote:
Originally Posted by mikecdm
Thanks hokiealumnus
4.556 Ghz / 217 fsb x 21 / Memos 1736 Mhz / CAS 7 - 7 - 7 - 20
http://www.pctunerup.com/up/results/...6GHZ777-20.JPG
4.588 Ghz / 218.5 fsb x 21 / Memos 1748 Mhz / CAS 7 - 7 - 6 -19
http://www.pctunerup.com/up/results/...36_I577619.JPG
4.578 Ghz / 218 fsb x 21 / Memos 1744 Mhz / CAS 7 - 7 - 7 -20
http://www.pctunerup.com/up/results/...2540_I5872.JPG
4.557 Ghz / 217 fsb x 21 / Memos 2170 Mhz :D / CAS 9 - 9 - 9 - 26 / SPI 2MB 20.110 Seg
http://www.pctunerup.com/up/results/...0I54.56GHZ.JPG
CARLOS*A1
Nice benching runs.:up: My combo seems very bclk limited (205 to 209 max)... How about posting an Eleet voltage tab screenshot showing one of those overclocks.
hey guys which is the best motherboard?
- Maximus III Formula
-ASUS P7P55D Premium(more expensive)
-ASUS P7P55 WS SuperComputer
-EVGA P55 FTW SLI
-MSI P55-GD80
Or other?
i'd say formula because i have one, though there are socket issues as i'm sure you've heard about which this board can be affected by. if you were to buy today, skip on the formula (any asus board really) because of the socket issue and go with either the msi or gd80 (tough pick for me between the two).
i've read about socket's issues, i 'll for a revision, but i think that the performace won't change
EVGA FTW!!.
I have the P55 SLI and runs perfectly like my ex Maximus III Formula.
Here a SS with my P55 SLI.
http://img59.imageshack.us/img59/6172/20215.png
Very nicely done both of you. What voltages did you have set if you don't mind us asking?
Later at night i will post SS with configs :).Quote:
Originally Posted by hokiealumnus
Short test with i5-750 L934B467 air cooled (ac freezer xtreme)
http://www.abload.de/thumb/i5-750-4ghz-prime9o7v.png
Wow @ vcore.Quote:
Originally Posted by Chri$ch
How many have you gone thru to get that one?
You mentioned that batch in another post! Very nice... I'd like to find its brother!Quote:
Originally Posted by Chri$ch
oh dear!!! how come these i5"s take so much Vcore to get to 4ghz ? im currently running my i7 at 1.270 4.2ghz :shrug:Quote:
Originally Posted by Retro
wow!, really nice vcore but temps are to high.Quote:
Originally Posted by Chri$ch
my config setup is.
3.8ghz 21*181 @ 1.258v at load and temp's are 55 - 59° celsius degrees, at 20 - 26° ambient temp.
First revision of the CPUs, as well as the PCI-e controller is on the CPU for P55 chips.Quote:
Originally Posted by dan7777
Tool is right on. For my chip, it will do a stable 4.0GHz with only 1.3vcore. For 4.2GHz it needs a lot more, 1.44vcore.
Like Retro said, it's luck of the draw. Some people are getting really nice chips that do 4.2 to 4.4GHz, but seems like most fall between 3.6 and 4.0...
i am so confused right now i don't know what i want to build.. i was thinking amd 965c3 & gd70 but then started looking at these.. they are under 150 picked up.. or the 860 for 230 and a msi gd80.. i just don't know.. i wish i could get them all just to mess with.. the 920 is 199 but i think i want p55 not x58.. uhhhhh
Two more SS for you guys.
Sub 9 DONE!.
http://img692.imageshack.us/img692/1344/8985.png
http://img199.imageshack.us/img199/936/19922.png
Sub 20s DONE!.
Finally i got 1156 support and new NT-H1 Directly from Noctua!, thank guys for sending this.
Here in Chile i have the records for SPEED and times on Superpi and 3dmark frecuency. Cooming soon Dry ice and CFX 5770.
Regards from Chile.
Running it @ 3.6 with turbo on. Volts were left on auto, but CPU-Z was showing about 1.37V Ambients were around 22 C
15 min run
http://i621.photobucket.com/albums/t...me359/38ss.jpg
45 min run
http://i621.photobucket.com/albums/t.../3845minss.jpg
Both completed with no errors.
Edit: I keep forgetting to fix my BIOs clock, so ignore the dates. :)
http://img94.imageshack.us/img94/953...psvoltajes.png
Nice temps here, thanks to Noctua for sending NH-T1 and new mounting system for 1156 socket.
Hi guys
I have Asrock P55 Pro with bios 2.0 that come with add CPU Vcore over current protection.:cool:
I ran 19x200 with Orthos for 24 hours straight...but with stock cooler spike temp too high:down::shocked:
I am still looking for aftermarket cooler.:rolleyes:
http://i38.tinypic.com/24np9pg.jpg
I need to know which is better use cpu ratio 19, 20 or 21 with turbo?:confused:
I can't change anything above 22-24 with bios manual.:mad:
When I set all auto and I see there is 24x in Asrock OC tuner, odd :shrug: pix below.
http://i38.tinypic.com/ru5rlx.png
Anyway I did play around in bios setting below.
For 3.8 setting.
19x200 -3.8
20x190 - 3.8
21x180 - 3.8
For 4.0 setting.
19x210 - 4.0
20x200 - 4.0
21x190 - 4.0
I find 19x200 is stable for now even play games works fine, of course I will change to 20x200 with new aftermarket cooler.
Also I changed the Foxconn plate to LOTES plate myself link at http://www.hardforum.com/showthread.php?t=1465061:up: :up::clap:
pg
i hope i can get some numbers close to some of those.. i don't have any to post on a i5 yet but i am getting so close.. i need some ddr3, thinking about the trident 2x2gb 2000... i am looking at the evga p55 ftw cpu area and they have holes on it for 775 and also the i5.. does anybody know if i can i use my v2 with 775 pro mount on this board without any issues? i know i can find out by trying it but i am using it right now on my maximus se. one thing nice is i will be able to eliminate my nb loop off my cpu for this new board..
I think you can, but socket 1156 brackets seem to be making there way out now. I would just keep your eyes out for one so that you don't have to have the sink angled.Quote:
Originally Posted by loften
If one core is not stable after 4 hour prime 95 while other are, but are the chances that my I will get a bsod? Cuz from 1.35 to 1.39 the core is not stable. Here are my settings:
blk 200
multi 20
vcore 1.38
vram 1.65
ddr 3@1600 9-9-9-24
vtt 1.3
cpu pll 1.9
going to try auto
The chip needs more Vcore than what I would like, 1.36v, but at least it's Prime95 8+ hrs and LinX 20-run All memory stable.
If I drop the vcore even one step in voltage it fails LinX at some point.
http://i59.photobucket.com/albums/g3...B/1d92a9da.png
thanks,, i took a peak at it a little closer. i see what you mean by angle.. the holes for 775 are offset a little.Quote:
Originally Posted by millertime359
looks like about 25 bucks for a kit shipped if they have any in stock.. maybe i will have to try a custom one..
mine:D
L921C834
only with super pi 32mb pass
http://img202.imageshack.us/img202/3817/45ghzresize.jpg
What cooler?
forward progress..
i made aluminum back plate, 3 1/4 x 3 1/4 x 1/8 inch thick.. marked and drilled my corner holes, drilled 1/2 inch holes to clear the 3 lotes back plate screws. added plastic shim for extra security between metal.
checked for clearance on the soldered pins under the sides of the plate.. had to clear a few of the tips... got some screws and nylon washers and mounted it to my mobo. 2 nylon washers on each screw under the plate for corner spacers.
all that was needed for the fuzion v2 775 bracket was to remove the pro mount screws and file the corners of the bracket holes a tiny bit and it fits perfect.. i have about 1/4 inch gap for tightening between the mount and nuts/nylon washers under it..
i will have to put a big hole in my mobo tray for it to fit as the plate and screws are sticking out pretty far..
total cost including aluminum was more then the kit by a little but less then it with shipping and i don't have to wait.. now waiting for memory..
I got another chip in to test and this one looks a little more promising. I noticed that this one has lower VID of ~1.225v while my other chips seem closer to ~1.25v. This overclock is with vcore set to ~1.35v in BIOS "with out vdroop/LLC enabled" and the board overvolts to ~1.376v load.
http://i695.photobucket.com/albums/v...0ES4095Mhz.png