Discussion 2 identical MB + 2 identical CPUs + 2 identical heatsinks + 2 identical PSUs + 2 identical CASES + identical RAM = Huge difference in OC

[Retro pc User]

I assembled 2 identical pcs:

• CPU : NOW i m testing [Q9550] but I have tried many [QX9650][QX9770] [Q6600][E5200]
• MB : Asus P5Q-SE2 (https://www.asus.com/it/Motherboards/P5Q_SE2/ )
• RAM : DDR2 800MHz
• Case : Chieftech 2008
• heatsink : Artic cooling (cheap)
• PSU : Bequiet

PC1: is perfect at 3900 MHz with the Q9550 (459x8.5) and holds up> 4000 with the Extreme processors

PC2: Not stable even at 3400 MHz with the Q9550 (400x8.5).

it's the same MB, I've tried every way, but the difference is amazing! Which MB P45 does not even hold 430 of FSB!

 

[Afro_ninja199]

no such thing as two identical CPUs

 

[CountMike]

It's called "Silicone lottery".

 

[Retro pc User]

no such thing as two identical CPUs

It's called "Silicone lottery".

I think the odd bit here is the MB: a P45 Motherboard that already gives problems at 420 FSB

 

[alexbirdie]

Winter 2012/2013 I built 2 PCs for my son and me with 3770k.

I was a kind of lucky, got a good 3770k, did 4.5 Ghz at about 1.18V ( for daily use 4.3 Ghz at 1.14V).
My son's 3770k needed 1.14 V for 4.0 Ghz. When rising to 4.2 ( about 1.18v) it becomes very hot and therefore stayed at 4.0 for daily use.

Both with same MB and cooler.

As some already wrote: silicone lottery.

No 2 pieces are identical.

 

[drea.drechsler]

no such thing as two identical CPUs

I think same could be said of memory...and even motherboards. VRM FET's are subject to the 'silicon' lottery as much as anything else, as the capacitors and inductors are to a 'electrolyte' and 'ferrite' lottery. So at the limit some samples may have more stable power output than others. That's why mfr's design margin into the system in the first place...and that margin is what overclocking is aimed at exploiting. Some just have more than others.