4-4-4-12@667 vs 5-5-5-15@800. Which wins?

[xilog]

So, I just discovered that my Corsair 5400C4 that's rated at 667 will actually run at 800 but only if I relax the timings to 5-5-5-15 and up the voltage to 1.95. Is this really any better than stock 667 using 4-4-4-12?

 

[Darryl_1983]

I believe faster speeds is better than tighter timings with newer cpu's.

 

[C4EVA]

Run some benchmark programs (SuperPI, WinRAR, ...etc ...etc) and see which settings produce the quicker results. I would have thought the extra 133Mhz would win in pretty much all tests tbh.

 

[Jokester]

It'll be much the same to be honest.

 

[Wiggins]

You may see a few more FPS on games but on a day to day usage point of view, like jokester says, you will see very little difference at all.

I ran some 533mhz memory oc'd to 660mhz with 4-4-4-10 1T timings and them compared them to Corsair XMS2 at 800mhz 4-4-4-12 2T and the Corsair was actually beaten in most memory benchmarks. I only saw a difference when I upped the voltage of the Corsair and had them at 900mhz 4-4-4-12 2T.

 

[HangTime]

Much of a muchness really.
Why not go for a halfway house, keep the timings at 4-4-4-12 and bump up the mhz a bit into the 700s.

 

[Banjo]

I'm not sure whether the difference will really be that major, but the faster speed + looser timings option will give better bandwidth overall.

Banjo

 

[Richdog]

Timongs make little difference on Intels nowadays... unlike the AMD's of old which were more sensitive than peppermint spray on the old chap.

Just keep it at 5-5-5 and go for the highest speed/bandwidth you cna muster.

 

[Talrinys]

Smack up that voltage a bit and get it higher