The Official OC (OverClocking) Thread!

esoteric1,
Don't know if you've read this or not but it's worth reading if you are interested in a P-35 MB.

http://www.tomshardware.com/2007/06/19/eight_p35-ddr2_motherboards_compared/


Best Regards,
theone :>)

 

esoteric1,

you're not going to get a very high OC with the vcore at only 1.36v. try 1.40 or even a little higher. you shouldn't have any heat issues with the cpu hsf you are using. even though the fsb/dram is unlinked, make sure the ram is 800mhz or less at least untill you reach the cpu OC you are trying to achieve. make sure the fsb and PCIe freq is unlocked and PCIe freq is set no higher than 115.

that cpu should OC to 3.0ghz or even 3.2ghz fairly easily.

 

Mort81

1.36v is just the voltage I run at with my 2.87 oc, the highest I've tried is 1.45v but the fsb just wont budge, the motherboard is definitely holding me back. Im just gunna get a better one and give this one to my brother who doesnt know what overclocking is.

 

esoteric1,
I'm so surprised. The 850i is supposed to be a good overclocker. When I built my 4300/965P-DS3, I started at 2.25 GHz and jumped right to 3.0 from there. I had no trouble hitting 355 on the fsb for everyday use and even ran it as high as 376 at 3.384GHz.

Clockin On, at the Speed of Light,
theone

 

esoteric1,

a buddy of mine (DocTY), has an evga 680i mobo and he loves it, even better than his asus P5W DH Deluxe (which is also a very good OC'ing mobo) and he DEFINATELY knows how to OC. he had his E6600 close to 3.8ghz on air and 4.0ghz on water. I seriously doubt the mobo is holding you back. just my .02.

 

The 6400 black processor is a 90nm windsor core with 2x 1mb L2 cache and 125w TDP.Other 90nm Athlons have a TDP of 89w so I'm guessing these cores are probably getting close to their maximum performance level.With reasonable cooling and quality components ~3.6ghz is going to be a good overclock.Remember they will heat up fairly quickly with vcore increase so good cooling is a must with these.I,m looking forward to seeing some benches from you when it's all together

The 5000 black edition is a 65nm Brisbane core and has a TDP of 65w.
There is no 2x1mb L2 cache 5000 black edition processor only 2x512mb L2 cache.These chips are clocking to 3.4ghz+ on air and need minimal vcore increase so heat is kept to a minimum.A good water cooled setup would probably see 3.5ghz+ with the right components.

As far as going 4x4 in the future,thats something to think about when I win lotto!!!

Phenom should be here this month,so I'll wait until next year when the reviews are out and the prices come down.

Regards,Crowy.

 

Maybe you just got a bad motherboard, or RAM. Sometimes people get REALLY unlucky.

 

NuckNFuts,
There is a 64x2 5000+ Windsor with the same amount of L2 cache. It's built on the 90nm platform and is an 89 Watt chip! See here:
http://www.newegg.com/Product/Product.aspx?Item=N82E16819103030

Clockin On at the Speed of Light,
theone :>)

 

Russ,Thats the locked multiplier 90nm windsor x2 5000.
It still has 2 x 512mhz L2 cache and won't overclock anywhere near as good as the x2 5000 black cpu.

Regards,Crowy.

 

Crowy,

I thought his original inquiry was about a different 5000+, that might run cooler than the "Black Edition"! The "Black Edition" is 64 Watts and the Windsor is 89 Watts. I thought that's what he asked! Could be wrong though. LOL!!

Clockin On at the Speed of Light,
theone :>)

 

Russ,he thought there was a 2x1mb L2 cache x2 5000 65nm black edition(which there isn't).
Only the black x2 6400 has the 2x1mb L2 cache.



The black 5000 x2 cpu's also have the new G2 stepping(the 90nm windsor are F3 stepping) which consume less power and will overclock higher on stock vcore.

Regards,.Crowy

 

I think I got bad results because I got the 122-CK-NF68-TR version and not the 122-CK-NF68-A1 and possibly on top of that I just got unlucky and got a bad board since it doesnt even hit 1333 and its rated for that, wish I RMA'd it while I still could, no longer have original packaging thanks to recylce prone parents. But maybe I'll just tell you guys how i went about ocing and hopefully i just screwed up.

CPU Thermal control - disabled
C1E Enhanced holt state - disabled
Intel Speedstep - disabled
CPU Spread Spectrum - disabled
HT Spread Spectrum - disabled
PCIe Spread Spectrum SPP - disabled
PCIe Spread Spectrum MCP - disabled
Sata Spread Spectrum - disabled

memory - unlinked

nforce SPP 1.5v
fsb eventually at 1.5v

ran into trouble around 1280 fsb
increased vcore from 1.36v to 1.45v but no luck

dropped multi to 6x and still wouldnt work
tried to get over possible fsb "hole", increased fsb by 40 at a time going all the way to 1700, had to clear CMOS everytime past 1320

tried using different 7x and 8x, no luck

cried myself to sleep, lol

 

esoteric1,

I was only able to run my E4300 at 9x or 6x. Wouldn't work at all at 7x or 8x. I was able to get the fsb to just over 1500 and at 3.2GHz it was running at 1470MHz.

When I replaced the 4300 with the 6750, I was able to raise the fsb to 444 (1775MHz) for 3.384GHz. The problem could well be the CPU example you have. See my PM!

Clockin On at the Speed of Light,
theone

 

The new MB has arrived! Imagine my disappointment when it was Rev. 1.0. I sent an e-mail which was answered right away. They explained that the board I received had been a re-built rev. 1.0, but it incorporated all the changes that are in the Rev. 2.0 with the exception of the additional 4 USB ports, and there are 4 headers on the MB for an additional 8 ports. She also advised me that the specs have been changed for revision 1.0 and are identical to the Rev. 2.0, with the exception of the USBs and the printer and com port location. I've been asked to try it before they will go any further, so that's what I'm about to do! Hopefully it will not work like the old Rev. 1.0 and start re-booting over and over.

Wish me luck!

Clockin On at the Speed of Light,
theone :>) (<:

 

theonejrs,

I noticed no restart issue with Rev. 1.xx of P35C-DS3R, no different with that of the P35-DQ6 which only got picky during either high OC above 417FSB and resuming from S4 after a long sleap/Hibernate session.

This P35C is fine to FSB404 with ACPI function for each S3 & S4 (Stand By Hibernate). Above 450FSB S3 becomes a hit or miss and usually crashes and restarts like DQ6 and requiring you to re set manual OC of CPU and voltages, PCI-E etc. S4 resume appears to be fine up to 505 and iffy above 515FSB.

Let me know what kind of HIGH FSB you can get stable on this mobo (GA-P35C-DS3R). I'm leveling out at around 515, and just moving to 516 becomes unstable (at either 6x or 7x for my E6750.

 

"P4 Prescott Overclocks"

How many of us jumped on the Prescott Bandwagon in early '04 for the new fun high FSB overclocks on an Intel sytem. It seams like it came and went so quick never really had a chance to call one rig a "quote daily user" as upgrading was happening so fast and often.

I started on the P4 3.4E on ASUS P4C800-E DLX (!875P), moved to P4 540 on ASUS P5AD2 Deluxe, "very" quicklt the 550J on new revised P5AD2-E Premium.

I think they were fun for the time since we had not had such fun overclocking an Intel CPU on Intel chipset for so long. This was the beginning of Intel becoming the performance choice for overclocking again. And now look where we are. Will it remain this way for a lil' longer? Only time, or AMD, will tell.

Can I get some cool results from the forum regulars and newcomers as well?

Rollin' Back The Clock!

 

NuckNFuts,
I'm up and running. First I went to 2.66, then 3.50 and then 3.6GHz. What voltage setting are you using for the CPU? I'm at 450x8 3.6GHz (1800 fsb) with no problems yet. Cpuz reports the CPU voltage as 1.44v but I have it set to 1.482v in the setup. I set the MCH to +2 and everything else to +1.

Any suggestions?

EDIT
I can get it to run at 3.55 perfectly, and it runs at 3.6, but I can't get it quite stable at that speed. I went up to 1.52v on the CPU but it just shutdown on the boot up. I'm 100% certain that it's not the memory as I've had it much higher before. The biggest difference was in the memory bandwidth. It's a lot fatter! The rest of the benches are a little better than they were with the old MB.









Still, it looks pretty good!

EDIT AGAIN!
Hit my first snag! Couldn't get the sound to work! I installed the SB Audiology card I was using before. Seemed to be working ok except that sometimes the computer would freeze up. Just lock solid. I finally noticed that it only happened if I used the Audio. I took out the SB card and decided to find out what the mystery was behind the HD Audio not working. The drivers were loaded but no sound. I opened the support DVD and discovered a folder marked Audio. Inside was an MS patch file which I ran! Presto instant sound! Simple enough. No more crashing either and the sound is superb!

Clockin On at the Speed of Light,
Russ

 

theonejrs, I dont have to tell you every CPU is different, but it is the same for Mobos too. I saw this a lot with my old P4C800-E's

You see now this mobo has a large vdrop with no CPU-PLL, FSB Termination, or GTL-Ref control so we just have to tweak around it.

We also do not get to see what the afective overvoltages are or become.

I never had to set BIOS any higher then 1.47500 to get stable at 450x8. It did take close to 1.51v to get stable at 475x8 as 3.80 with vdrop bringing it daown to 1.46v loaded. I have the 3 month stable rig running as 3.50GHz as 500x7 w/ vcore of 1.43125v (BIOS) reported 1.39v CPU-Z/Everest, and drop to 1.36v 100% loaded. Mobo overvoltages (MCH, FSB, & PCI-E) need to be at least +0.2v from 450 ~ 505FSB to keep mobo stable using on Intel RAID controler ICH9R0. DRAM was/is set as needed ro whatever I use at the time for the DDR1000 in 1:1. Currently Crucial Tracers @ 4-3-4-9v w/ +0.3v. Since we also get no DRAM ref control, it bounces up as high to 2.34v but drop to as lil as 2.24v. 2.34v is fine since they are air cooled w/ tri-fan set from Doninators.

What other "Extreme Tweaker" settings have you adjusted?

I did have to go full +.03v on all mobo volts to keep stable at 515FSB w/ 7x. I ran mine for testing with tight timings w/ MCH & FSB to +0.3v and PCI-E to +0.2v and DRAM to +0.3v.

ASUS MAXIMUS FORMULA SE I finally got some time to play on my 2 week new system and it is doing nice so far. Just wanna finish tweaking to see how much further I do get for the money spent. I do wish I waited for the P5E WS Pro as I had loving experience with my past work station boards from ASUS since my recent P5WDG2-WS Pro.

Clock On!

 

NuckNFuts,

Enlighten me! What's the advantage to running a lower multiplier? Would I be better off running a x7 multiplier with a 515MHz fsb to try and get to 3.6GHz? That would put me on a 2057MHz fsb! My big problem with the voltage! The bios will not allow me to set it higher than 1.50v. If I go 1 step above 1.50v the bios resets. At 1.50v, Everest reports 1.46v. Do you recommend that I up the voltages to +2 or +3?

I think it says a lot that the processor runs out of steam at exactly the same settings as it did on the 965P. It makes me more inclined that I've about reached my limit with this chip. This MB would run at 3.6GHz but the 965P wouldn't. I just can't get it stable, although I haven't pushed all the voltages up. Everythings set to +1 except the MCH which is set to +2!

Let me know what you think, please!

Best Regards,
Russ

 

theonejrs, overvolting the mobo to +0.2v is a start (especially the FSB since this is what needs more juice to get above 425. 400 at +.1 or "AUTO" is fine but you're gona need that +.2 for stability under load.

What RAM and timings have you tested with this same RAM to refference with? A vcore of 1.47500v set in BIOS should give you something close to 1.42v in CPU-Z and 1.4v loaded. I would try your 3.6GHz as 450x8 since I didnt find this mobo to be so stable over 515FSB.

So interesting to see it was just the mobo for me. Cause moving the same CPU to the X38 ASUS returned the same CPU results but with overall added smoothness.

It may be the CPU as well. I want an E6850 and wonder if I will get the same as early rev. But looks like almost all the E6850's do 4GHz so not so risky I guess. 4.2GHz on the otherhand is gonna be hit or miss from CPU to CPU.

Good Luck! Keep us osted!