ranking by Mhz is just a see who has the highest OC.
the whole idea of super pi is the speed at which it completes.

again...
just trying to get my head around how all of this is supposed to work.

1. super pi 1M and 32M measure speed of processing.

2. OCCT and Orthos Dual Prime measure OC stability.

3. the Suicide screenshot measures the highest OC that will boot to windows.

are these assumptions correct?

This is starting to become a pain in the arse... So nrg, you want to have seperate categories for dual/quadruple instances and single instances? That sounds fair enough to me I'm not sure about the times though...

[Edit @ Rehit's edit] SuperPI is also a measurement of stability, just like 3Dmark06 is for GPUs. It spits out a time/score, but it still tests the stability of your OC.

i certainly did not want to cause any kind of trouble.
i am just trying to understand the processes and the logic behind them.
people have to understand that some of us are noobs and are trying to learn from those who are not.

i understand that super PI tests the stabilty...
but how many stability tests do we actually need.
it makes sense that some tests would just measure raw speed.

Check the first page...

yes, but it's also a form of stability test... 32M is harder than 1M, 4M is easier than 8M, etc... dual cores is harder than one of two, four cores is harder than one... that's the point of running SuperPi on all cores from an overclocking point of view

say if your CPU did 4.3ghz suicide, but your prime95 24+ hour was like 3.9ghz... you could show that the CPU is stable enough to run SPi 1M at 4.2ghz or something... even though the best time is at 3.95ghz...

meh... basically you've come into this thread and asked everyone to change (over a year's worth?) their results because your Quad core gets hot when all four cores are doing something... just seems a bit odd lol

We should have the super pi done by the MHZ So the highest clock to pass both 1m and 32m


and then a single pass for the 1m and 32m Fastest time is the best.

and keep the rest the same. as running more than one instance of super pi is hard to judge as *not* every one starts them all at the same time bla bla bla

................................................................................

Anyone else?!

i never meant for anyone to change anything.
i was just trying to understand how things work.
i have tried to point out along the way that i am not trying to make waves or cause an arguement.
i will be more than happy to play by whatever rules are in place.
noobs have to learn at some point, and understanding how and why things work is fundamental to learning.

Rehit, there's nothing wrong with asking questions etc. You do not need to set the affinity each time, your windows will take care of that (does for me every time...), so it's not such a big pain. I split the categories now, because I'm sick of everyone b**ching about this and that.

*goes back to study*

The reason for running multiple instances( At least in my mind) is to prove a measure of stability as well as the ability to make the run. There have been many times I could run a single and have it pass but if multiple threads are run, each with the affinity set to a separate processor the ability to make that run is more difficult. Hence it keeps the wham bam get into windows and get a Pi screenie before the computer bsod's or just reboots crowd out of it. Im sure I can pull off a 12 minute 32m pi score but If I can run all four cores at that same speed then It carries more weight to me. But I am a big proponent of long term stability. Just my 2 pennies worth.

Ok here you guys go . Single instance of Pi at 4.1GHz. That was with my Mushkin 9200 1:1 . Game on.

Gotcha down Son Is that air or water?

I hope this was enough changing the categories, this is beginning to be a little bit annoying

Go ahead and put me down for water and the E6850...

Mine was on Air. Tuniq Tower. Nice one there roadrunner. I thought I might have an edge but it looks like that was short lived.
Once I get back from CES i guess i will have to put the 6750 under water and see if I can pull another 100 or so MHZ out of it.

Alrighty, fixed that up and added road Time for bed lol

Just a little better.

That E6750 you got is a beast.

Why yes ...Yes it is. Anything over 4.120 doesn't seem to want to boot. The board it is on is different from the first screen shot so I am creeping it up and figuring out what it wants. But enough for now.

Hey Makaveli, where the hell are you with your Q6600 anyways?

thought this was quite nice nearly 100% on fsb
just installed vista ultimate x64 sp1 the combination is quite beutiful
.....Click to view attachment

JUst did this, hope it's ok.

Yep, that's fine

here we go with a little bit of a OC on my E6850 with my watercooling setup



\

Here is a suicide of 4049MHz





more to come soon, well when I get a day off

3920MHz 30min OCCT Stable

more to come

*sigh* I seem to be having issues getting to 3.3GHz, regardless of CPU voltage and FSB/multi. King, what are your voltage settings?

I cant seem to get 4000MHz stable at all...500x8 and 450x9 are not able to stay stable 3920MHz is the highest so far, keep getting a BSOD with a memory error, lucky me

i think that gives me 3rd place

Alright... lemme do the noob explanation j/k

For 3.3 I have 1.408 Vcore set (too much, but I don't have time to tweak it these days, I'm sure I could pull 3.6 outta this one!), and left all the other voltages stock NB helps a hell of a lot, but be sure to add a fan on there ASAP, that mofo gets REALLY hot!

I haven't rebooted this rig in like 2 weeks, so I have no certain values for my settings, but did you turn all the spread spectrums off? Also, what is the 5V rail of that PCP&C like? Cause same as my SLI-DR Expert, this Blood Iron seems to be pulling the most from the 5V line.

@ mojo, I can add OCCT, but you need to get the newer version of SuperPI SuperPI 1.5 Mod by XS or something, the link is in the first post

@ firky, gotcha mate

damn...alright ill get that and re run with the 3920MHz

here is the dual instances...

The Silencer 750 has 30A on the 5V rail.

I've tried upping the vCore, it doesn't make any difference. 3.2 is completely stable, 3.3 hardlocks Windows within seconds of starting Prime95. I've tried different FSB/multi combinations to make sure it wasn't an FSB hole, no dice. I tried bumping all of the voltages (except clockgen) and it doesn't make a difference. I turned off C1E, spread spectrum, etc.

I know the problem isn't my temps. My hottest cores are idling at 27C (3 GHz with WAY too much voltage). Load temps are only 36C after a few minutes of Prime. I don't know what my temps are/were at 3.2 though. I don't have time to restart and check because I have to go to work.

I was mainly curious about your SB and SB Core (CPU PLL) voltages. I can figure out/test the rest easily enough.

Sounds like you may have not been to lucky with the chip you got. Some OC well some do not....

Here is a single instance of super pi just over 4GHz

lol that was my new highest clock till i beat my self





Getting fast"er"

Alright, that should suffice My Noisetaker II 600W has 34, but I believe 30 will work fine

Anyways, as a matter of fact, I just rebooted and had a look at my BIOS. NB voltage was 1.41, SB at 1.55 and the other was at 1.207 or something like that... Not really high voltages there IMO. Did you try the other RAM slots? At least for the 939 DFIs, different RAM slots for different ICs, some would seriously limit your OC potential, some wouldn't even boot if inserted into the wrong slots.

@ Firky / mojo, added

Yeah... reminds me of my old P4. Everyone else was getting 4+ GHz and I had to fight to get the dang thing to 3.6. I'm not conceding defeat yet though. It's just disappointing because none of the CPUs I've had OCed better than average. My P4, my E6600 that wasn't stable of above 3.55, and now the Q6700 that won't go beyond 3.3.

Out of 4 quads I got one decent one....

thanks for the add, I will probably be trying some more things tonight when I get out of work (1:30am EST)
here is a little better pi time

Lol, thanks for the add sounds like some random fool on myspace Luckily I never had one of those accounts

thanks king I just ment that I appreciate you taking the time to keep the socket 775 comp up and running

You're welcome Mojo I try my best

My Orthos 24 hour entry.

Ahh yes... Phase change is neat indeed

kitfit1 what kind of temps are you getting with that Mach2 GT?

Nerm, where the hell are you with that X2 4200+ in the 939 competition anyways?

I haven't done anything with it since the compressor on the chilly1 blew up. It has been running at stock speeds for nearly a year

I am currently trying to convince the wife to let me build a whole new rig.

Stock speeds?! Are you kidding me?! Not even an air overclock? NOOOOOOOOOOOOOOOOOOB

Well since I got Crysis for Christmas I am going to have to OC the hell out of it again. lol

Yea, you probably have to, but that's always fun, isn't it?

So, I spent some time fooling with this mofo again This is with increasing the volts from +200mv (millivolt methinks?) to +225mv So a minimal increase, I also bumped up the chipset voltage a wee bit. It's been folding like this for a while now, I played some CSS, seems alright. I'll test 24 hours Prime later

1M @ 3.6 (400x9) Single Instance (older screenshot):


1M @ 3.5 (438x8):


32M @ 3.5 (438x8):


OCCT @ 3.5 (438x8):

Come on up the clock some more! I notice my Quadruple runs with the Quad @ 4104 never got posted. http://forums.overclockersclub.com/index.p...6496&st=840