Finally decided to Overclock E6600

[Tigjaw]

Hi guys,
After a year or so I've finally plucked up the bravery to overclock my E6600. Now with so many new CPUs taking the shine off my little conroe, like a balding man buying a porche I delved into the field of overclocking.

Following a few beginners guides I started with a very modest overclock from 2.4GHz to 2.7GHz.
Basically I just want to know if I've done everything correctly and if my temps are okay (at the bottom). Also my rig is in my sig but just for convenience sake:
Conroe E6600 2.4Ghz, Gigabit DS3
X1800XT 256MB, Gskill 2Gb PC2-6400
160GB WDC, SOHW-832S DVD/RW
Enermax 620W, T.T. Big Typhoon.

These are the steps I took:
disabled CIA2
disabled C1E
disabled EIST
disabled Virtualization Technology
disabled CPUID
disabled No-Execute Memory Protect
set PCI Express (PCI-e) frequency to 100 mhz

-Set Graphics Booster to Auto
-Set CPU Host Clock Control to Enabled
-Set CPU Host Frequency to 300.

-Set System Memory Multiplier to 2.00
-Set DRAM Timing Selectable to Manual
-Set CAS Latency to 5
-Set RAS to CAS to 5
-Set RAS Precharge to 5
-Set tRAS to 15.

-Set DDR2 OverVoltage Control to +0.2v.
-Set MCH Overvoltage to +0.1.
-Left vCORE at 1.325.

The above instructions are supplied in an Overclocking 101 for the Gigabyte DS3 article courtesy of TomsHardware. (can I link it?).

Anyway, here are my temps:

So basically as it's my first overclock and to me the temps look slightly high at idle (40c) I was just hoping for some feedback on anything here.
Just to confirm, I did run an orthos test for 13hours overnight and everything passed.

I'm hoping to pump it up to 3GHz tonight; would it be a bad idea to go straight from 300MHz*9 to 340MHz? and what temp increase should I expect for that?

Thanks in advance,
Regards,
Tig

P.S: sorry for the small pics, the site I uploaded the image to shrunk it. Everything is still visible, it just takes a bit of squinting.
P.P.S: You may have noticed I posted a thread about my Vcore saying 'fail' but that was at stock ie. before this overclock. Since I set the Vcore manually the vcore now says 'ok'.

 

[shadowscotland]

My E6600 would hit around 3.2 without any voltage changes on an overvolting IP35pro.

My E6600 clock really well with only 1.35v in bios - make sure you ram is at the right voltage/CL - CPU-z should help check on the memory tabs.

340 should be fine as should 360 or even 370 at stock volts - 400*8 worked well for me too (that way ram is running 1:1) with some extra juice to the NB

At around 3.2-3.4 they need extra volts (and produced significatly more heat) most max out around 3.6 on good air/water cooling.

From 2.4-3.2 (1.35v) my chip gained 9c with the stock cooler using orthos.

Excerlent chip for non-overvolting overclocks, runs cool, have loads of cashe - great little chip

 

[Tigjaw]

My E6600 would hit around 3.2 without any voltage changes on an overvolting IP35pro.

My E6600 clock really well with only 1.35v in bios - make sure you ram is at the right voltage/CL - CPU-z should help check on the memory tabs.

340 should be fine as should 360 or even 370 at stock volts - 400*8 worked well for me too (that way ram is running 1:1) with some extra juice to the NB

At around 3.2-3.4 they need extra volts (and produced significatly more heat) most max out around 3.6 on good air/water cooling.

From 2.4-3.2 (1.35v) my chip gained 9c with the stock cooler using orthos.

Excerlent chip for non-overvolting overclocks, runs cool, have loads of cashe - great little chip

Thanks shadow; there's some nice benchmarks I can work with; I'll give 340 a go tonight and run orthos overnight again.

You say that you gained 9c from 2.4-3.2GHz with the stock cooler; does that mean with my Big Typhoon I should have a temp less than what I've shown in my screenshots? It went from 37c at stock to 41-43c at this overclock; that's quite a jump for a good 3rd party fan isn't it? (Big Typhoon).

 

[Matt_Wilkinson]

look this may be forum but i garuntee you e6600 is good. no need overclock as it runs fat anyway. why change it into a e8400 when it is faster to start with. i know the clock speeds indicate otherwise but i am pretty sure that in RL performance a dual core is better than a quad for gaming.

 

[Tigjaw]

look this may be forum but i garuntee you e6600 is good. no need overclock as it runs fat anyway. why change it into a e8400 when it is faster to start with. i know the clock speeds indicate otherwise but i am pretty sure that in RL performance a dual core is better than a quad for gaming.

I know E6600 is great at stock anyway in fact I'm not that tempted to change for a while yet considering gaming peformance for example isn't increased that much by some of the newer processors and games don't even use four cores yet. However; this is in turn the reason I decided to overclock my E6600, I don't want to replace it but I would like to give it a bit more juice and I'll definately be content to stop at 3GHz.

 

[shadowscotland]

You say that you gained 9c from 2.4-3.2GHz with the stock cooler; does that mean with my Big Typhoon I should have a temp less than what I've shown in my screenshots? It went from 37c at stock to 41-43c at this overclock; that's quite a jump for a good 3rd party fan isn't it? (Big Typhoon).

Firstly the stock cooler ramped up the fan - so actual gain if fan was constant would have been higher.
And secondly the 9c rise was measured without changing any voltages. (unlike your results - voltage affects temps more that core speed)

Is your BT setup for airflow or low noise? (aka fans and speed/volts)

 

[Tigjaw]

Firstly the stock cooler ramped up the fan - so actual gain if fan was constant would have been higher.
And secondly the 9c rise was measured without changing any voltages. (unlike your results - voltage affects temps more that core speed)

Is your BT setup for airflow or low noise? (aka fans and speed/volts)

I think it's at max RPM; I've looked at several programs; I-Cool, TaT, CoreTemp and so on and I can't find any settings that ramp up the fan speed.

I in fact made a mistake in my first post when I said I put vcore up to 1.35; I didn't actually touch the vcore I just left it at 1.325v.

Since my last post I've overclocked to 3GHz with idling at ~44c and ~56c at full load (EDIT ~59c right now in prime). Each time I overclocked I left the vcore alone and the computer loaded fine but prime failed every time. (Prime with 2 workers and it was only worker #1 that failed each time).
To combat this I upped the vcore, still failed, upped again, still failed; I've now got the vcore to 1.35v (which is about 6 or 7 notches above stock?) and it has ran prime for about 9 hours without errors or warnings.

Maybe my E6600 just isn't up to the usual overclocking standards of E6600.

Could there be another reason why I've had to up the vcore so much? Like yourself, most people have said they didn't have to touch the vcore even at 3.2GHz.

 

[gurusan]

look this may be forum but i garuntee you e6600 is good. no need overclock as it runs fat anyway. why change it into a e8400 when it is faster to start with. i know the clock speeds indicate otherwise but i am pretty sure that in RL performance a dual core is better than a quad for gaming.

The E8400 is faster than an E6600 stock for stock, also overclocks higher and is faster clock for clock...and they are both dual cores :/

Every single one of your posts has been total misinformation. Go away and learn something before attempting to contribute something to these forums again...

 

[rypt]

look this may be forum but i garuntee you e6600 is good. no need overclock as it runs fat anyway. why change it into a e8400 when it is faster to start with. i know the clock speeds indicate otherwise but i am pretty sure that in RL performance a dual core is better than a quad for gaming.

Firstly ... CLAIMED!

Secondly, please please learn how to communicate in "proper" English.
I've even highlighted the errors.

THIRDLY GET A CLUE
E8400 is a dual core just like the E6600 (hence the E rather than Q in the name)
The E8400 is a good 15% quicker clock for clock than the E6600

This is the 2nd or 3rd thread of you sprouting BS, go troll somewhere else.

Applies perfectly to the above quoted poster

 

[shadowscotland]

Maybe my E6600 just isn't up to the usual overclocking standards of E6600.

Could there be another reason why I've had to up the vcore so much? Like yourself, most people have said they didn't have to touch the vcore even at 3.2GHz.

Firstly the auto setting for Vcore is different from one board to another.
Also as I said in my first post my Abit overvolts - so 1.35v in bios would be higher in cpuz/uguru/easytune etc.

1.35v is not high - it's around the 'auto' or 'stock' voltage of some boards.

As to the fan - what it connected to - molex/3 pin mobo header/4 pin mobo header.
If it's the first two most likely it's at 12v (max rmp)
If it's the cpu fan header (4pin) then it should change speed dependant on temps

a four option is it's a 3pin fan in the 4pin cpu fan header - again this will be max rmp

 

[garylouden]

I have had my E6600 since release. I found the best and easiest way to OC mine was to set all voltages to auto (except RAM) and then clock to my desired speed, check the temperatures and then start lowering voltages to see how low I can get it. The reason for this is because my 680i seems to be inaccurate when dealing with the voltages.

I think for me to hit 3.4Ghz is 1.4ish (at work cant check) and my temps seem to be fine, 40-45 idle and 60-65 during orthos using a Akasa Freezer 7 Pro.

 

[Gunda]

while leaving volts on auto is easy it can does tend to lead to higher volts than needed to reach a stable clock.

You may need to up your volts slightly on NB to remain stable at the higher FSB speeds.

 

[Tigjaw]

Firstly the auto setting for Vcore is different from one board to another.
Also as I said in my first post my Abit overvolts - so 1.35v in bios would be higher in cpuz/uguru/easytune etc.

1.35v is not high - it's around the 'auto' or 'stock' voltage of some boards.

As to the fan - what it connected to - molex/3 pin mobo header/4 pin mobo header.
If it's the first two most likely it's at 12v (max rmp)
If it's the cpu fan header (4pin) then it should change speed dependant on temps

a four option is it's a 3pin fan in the 4pin cpu fan header - again this will be max rmp

Firstly thanks for the replies guys; secondly Shadow, it's the fourth option:
''a four option is it's a 3pin fan in the 4pin cpu fan header - again this will be max rmp''

So I guess I'll just be dealing with the temps, I know they're not bad to be honest, I just hoped they'd be ~4c lower than they actually are at idle and load.

Prime has now been running for ~15 hours on both workers at 3GHz with 1.35v at temps of 58/56 on full load.
EDIT: By the way, my BIOS says that 1.325 is the normal vcore for my processor.

Forgive me as I'm still not knowledgable enough all this hard-techy stuff yet, but would increasing the DDR overvoltage allow me to reduce the vcore?

And if not, is there any other optimising I could do here to get the best performance possible?

One thing I found out last night was, out of these combinations:
Multiplier 9 * 300 = 2700MHz
Multiplier 8 * 337 = 2700MHz
the one with the lower multiplier would be faster despite the identical clockspeed. But how far does this go?
Would Multiplier 8*337 (2.7GHZ) be more (powerful) than, equal to or less (powerful) than 9*3.33(3GHz)?

Thanks again for your wisdom

 

[shadowscotland]

Lower multi stresses the NB - so you might need more voltage to keep it stable.
But higher FSB affect everything so 8*337 is faster the 9*300

FSB affect ram speed directly, but over the rated speed you might need to losen timeings and up volts. under rated speed tighter timing and less volts. Some ram will happily run signifiganly above stock timeings.

From my own testing running ram at CL4 with a 3.0Ghz = CL5 + 3.1Ghz (using same multi)

now you got some of the basics fleshed out might be worth a re-read of the sticky and advanced OC sticky