Quote:
Originally Posted by hifiking
yeah you go ahead and set your vcore to 1.775.. you might as well just walk over and throw your cpu in the garbage.
Printable View
Quote:
Originally Posted by hifiking
yeah you go ahead and set your vcore to 1.775.. you might as well just walk over and throw your cpu in the garbage.
I just did. 1.775 vcore It's reported as 1.74 in speedfan.Quote:
Originally Posted by Lestat
Memset beta for 680i. Seems to work asside from a little glitch in some fsb freq reporting.
http://rv.page.cegetel.net.perso.ceg...mSet31beta.zip
Naw don't need a new Vcore Mosfet heatsink. The base just needs to have thew bow taken out of it. If you put a straight edge on the base of mine the middle is about .040" higher than the ends lol. I just talked to one of my customers and he said his MB eVGA MB was the same way. Kinda dumb to put a heatsink on with the base so bowed out it clears most of the mosfets it was supposed to cool lol.Quote:
Originally Posted by Roger_D25
Viper
Reported is usually a long way from actually metered.Quote:
Originally Posted by hifiking
Viper
Quote:
Originally Posted by icon57
I thought that too. I can drop the volts and still boot windows. This spu is a monster. The board is limiting me right now heavily. I finally got this board to run 410FSB and I'm doing it on stock volts. 3.7Ghz no bump. It's wierd. I can boot windows and benchmark and game no problem. If i go to 1650 on the evga i stop posting immediately no matter what I set my volts to. I'm running the P21 Bios. Here are my settings.
CPU Core: 1.325V
CPU FSB: 1.2V
Memory: 2.10V
nForce SPP: 1.35V
nForce MCP: 1.525V
HT nForce SPP <-> MCP: 1.35V
the HT voltage at 1.35V and all the other voltages low finally let me get to 1600Mhz (400Mhz) + FSB.
Now I'm at the next brick wall. 410Mhz FSB
No matter which voltages or combinations of voltages I make changes to I can't get past this......
Any suggestions from anyone or is this pretty much the peak of this board. I read some great reviews and thought I'd surely hit near 500Mhz FSB.....
Quote:
Originally Posted by Lestat
:eek:
I have a test E6600 from hell...type 'A'Quote:
Originally Posted by easyrider
Max vcore out of my evga (when set to 1.8v) is around 1.71
ran the same chip at 1.75 for benching on other boards
ran the type 'A' 6600 at 1.75 for 3 months straight. So if it dies in another 6 months, it'll probably cost $100 to replace....
Been up to 4.7Ghz with this chip at times
Poor cpu....to have found such an abusive owner...:stick:
That's pretty good, I set mine to 1.8v and it only gets to 1.74 reading in BIOS.Quote:
Originally Posted by hifiking
:rolleyes: You really have NFI do you mate :nono:. You need at least 1.84vcore to get to 5.5Ghz.Quote:
Originally Posted by Lestat
Uh, did you miss my reply that responded directly to your original post??? :(Quote:
Originally Posted by pablo906
heh, maybe you've got me on your ignore list :D
I've refined the idea further anyway since then, try this.
Changing tRC to as high (loose) as possible may be required too.Quote:
Originally Posted by SLi_dog
All maxed in bios....mem. ok @ 2.45/2.5...Still with BFG and my resident ram Team Xtreem OPB ed...
http://img85.imageshack.us/img85/4532/1yo7.jpg
I can set all volt to auto, but vcore at 1.5 and get 3.4 GHz, but if I set multi to 6x, I cant raise the FSB before vcore is raised. Even though it's running at 2.4 GHz. Why is that? The only factor in high FSB on my board is vcore. Even if i set all other volts to 1.5 and 1.55 i cant raise the FSB without raising the vcore.
Here's my oc and notes so far.. Havent pushed the cpu and board much 'cause my cabi is really crammed and has bad air flow.
Specs:
EVGA 680i SLI
E6600 Watercooled
8800GTX
All Spread Spectrum are disabled.
HPET Function = Enabled
CPU Multiplier = 9x
SLI-Ready Memory = Expert
FSB - Memory clock mode = Unlinked
FSB (QDR) = 1699 mhz (at 1700+ a higher strap kicks in, ie lower performance)
MEM(DDR) = 1066 mhz
Timings: Expert 5-5-5-15-2T - Rest are Auto
VT and Speedstep are diabled.
CPU Core = 1,5650? (if i recall correct..)
CPU FSB = 1,30v (1,20 should do fine up to 410. Some say it doenst post @ +1,40v)
Memory = 2,20v
nforce SPP = Auto (i think linkboost ups it to 1,35v)
nforce MCP = 1,600v (these chips are like cpu, they might require more or less volt)
HT nforce = Auto (i think linkboost ups it to 1,45v)
FSB=424,75
CPU = 3822mhz
Mem = 1066mhz
One more thing, Linkboost is enabled and active. PCIE=125, HT=250.
hifiking,
CPU @ 3000mhz with high fsb requires more vcore than CPU @ 3000mhz low fsb.
For everthing below E6700, i would recommend keeping default multiplier. Dropping the multi and cranking fsb above 424(1699)mhz gives considerable performance drop.
Above 424mhz the strap speed is changed. Lowering multiplier also drop perfomance..
From a fellow danish overclocker (Hessi):
424,75 (1699) x 8, Super Pi 32M @ 15:39,281
430,50 (1722) x 8, Super Pi 32M @ 15:56,390
424,75 (1699) x 9, Super Pi 32M @ 14:29,360
430,50 (1722) x 9, Super Pi 32M @ 14:36,969
411,25 (1645) x 9 = 3701,25 Mhz. Super Pi 32M @ 14:39,563
462,50 (1850) x 8 = 3700,00 Mhz. Super Pi 32M @ 15:01,219
So on an E6600, 9 x 424mhz is optimal. If you drop the multiplier, you need a FSB of atleast 475mhz to compensate.
Thanks, so it's normal then.Quote:
Originally Posted by illuminati-hwt
I can't even get my ASUS P5N32E with E6600 to boot above 375fsb. This is my first Intel in over two years. I've been toying around with DFI boards on the socket 939 for too long.
I've tried adjusting voltages, and I've disabled/enabled different settings. Their has to be something huge that I am overlooking. Even at 3375mhz I could get Orthos stable, but crash whenever I tried to reboot. I thought that was pretty damn strange.
i was wondering if anyone has this problem with the motherboard shutting off on cold boot. i have been running fine on this mobo for almost a month and getting great oc. just yesterday night i turn off the comp to add some fan and when i turn it back on it wont boot. it turn on for like 2 second getting ready to boot and then shut off. the only way i can get it to boot is keep trying until i get lucky. if i do get lucky and get it to boot completely, it can restart very easy there is no problem and it can boot to window. it only happen when i turn it off. I try everything to figure what the problem. i changed voltage, cleared cmos, took it out of the case maybe due to grounding issue, took out the drive, ram, fan.....everything and still have the same problem. only thing i havent try is replacing the psu, but i really doubt that is the problem. i hope someone have an idea what is causing the problem. also i really hope i dont have to rma the board cus i can get 500fsb on it now and i would hate to get another board that doesnt oc as well.
@ Dumo, whats highest that ram can do as 1T using 2.5vdd / 4-4-3-3 on
680i chipset.....
So far...DDR2 932 4-4-4-1T @ 2.4V, faster than DDR2 1244...Quote:
Originally Posted by Dynasty
http://img67.imageshack.us/img67/558...shot204ed9.jpg
And 1300@ 5-4-3-5 + single GTX...
http://img301.imageshack.us/img301/3...shot049qe3.jpg
thanks for testing :toast: ...
killer bandwidth with 1T @ DDR2 1000 with more vdd...
Check out the growing community having problems getting any kind of stability with the RAM...
http://www.evga.com/community/messag...TOPIC_ID=21665
quote:
--------------------------------------------------------------------------------
Originally posted by gurnio2
I'm running 2G OCZ PC2-8500, and can't complete 3D Mark 06 unless I underclock to 800MHz. In fact, it locks up at the same frame each time (when the scary dragon jumps over the blimp)
--------------------------------------------------------------------------------
quote:
--------------------------------------------------------------------------------
Originally posted by Vernius
EXACTLY my problem man.
Most of my 3Dmark06 lock ups happen on that Canyon Flight test.
The dragon jumps out of the water and everything hangs up. Unless I underclock my ram.
x6800
evga 680i
a-data 1066+ 8500 Extreme
2 x 8800GTX SLI
quote
--------------------------------------------------------------------------------
posted by mda96bjh
Thats exactly my problem to!!!!!!!!!!!!!!!
I've managed to get my rated speed though now.....but i don't know if it is really giving me much of a performance boost.....
try disabling SLI memory......run the memory "linked" on the 4:5 divider...
set your timings manually & make sure the Vdimm is sufficient..
that may help
most of them seem like a bunch of n00bs that just have no clue how to run memory in a modern system
NO they are not plug and play even at stock........you have to make sure your RAM is stable before even installing windows.......................
but THIS IS NOTHING new.................i feel sorry for DFI Street, Abit USA and other forum mods and staff trying to help out all these impatient n00bs....well i don't feel sorry for Angry_Games or is he hippie Happy_Games now :rolleyes:
Quote:
Originally Posted by dinos22
well i am not a n00b.....& know for a fact that very little is plug & play.....but i can confrim the bug discussed in the post above
using both the adata 1066+ RAM & the OCZ 9200
no amount of tweaking will stabilise that system using unlinked settings or the 1:2 divider above 1000mhz.....
I did just finish a 3DM06 run with my OCZ6400 5-5-5-15 at 1000mhz 5-5-5-15. FSB at 1500. EVGA680i, P21 bios.
http://i11.photobucket.com/albums/a1...dle/HSAssy.jpg
This TC almost has a chewing gum/putty texture to it (no I am not saying it's chewing gum/putty)
http://i11.photobucket.com/albums/a1...ipsetgoped.jpg
The PCB on this chipset looks a bit damage doesn't it? You can see an outline on the PCB itself. What could have happen is unknown but that's not how a chipset on a PCB should look I tell you that much.
http://i11.photobucket.com/albums/a1...ipsetgoped.jpg
The TC looks like chewing gum/putty texture look, IMO (no I am not saying that it's chewing gum/putty). Can this really help reduce heat? The outline isn't as bad as the other chip set.
It might be the best stuff on the planet. Folks lets be honest here...does this look like this will keep your N/S Bridge chipset cool or, can this increase the heat? Please take board and remove these HS and clean the TC from the HS and chipset and apply some AS5. I suggest removing the TC with at least 90+ Rubbing Alcohol or AS5 Cleaner.