Check out my setup and give me advice

[RavenGuard]

Ok. First off, here's my hardware:

Gigabyte P35-DS3L
500GB 7200RPM SATA
2GB Mushkin DDR2 rated for 1066mhz @ 5 5 4 12
Core 2 Q6600
HD 2900 XT
Enermax Liberty 620 watt



Here is how I have it configured currently. Honest to God, I read a whole pile of crap and gave this a try. Please let me know anything wrong with my setup and give me some tips.

Core 2 Q6600 is set to 2.8Ghz (on stock cooling, will go higher once I get cooler) using a 9x multiplier with FSB 311mhz.

2GB DDR2 is at 620mhz @ 3 1 1 4

HD 2900 XT running at 828/900


This is entirely stable. Before this goes to the critics, I have some random info. Firstly, when I first made the build the RAM was set to 1066mhz at it's rated timings. It worked fine, but Vista gave my RAM a lower score than my hard drive. After this OC and even with the RAM timings at 4 4 4 10 vista told me it was performing at a 5.9, the max possible score. Now the timings are even tighter, so I figured this would be the best setup.

Is the low clock + awesome timings > high clock + crappy timings?

Thanks.

 

[kenrippy]

i would bet (guessing here) that high clock + crappy timings is better. in the real world, cpu clock speed is what matters, ram speed & timings are for benchies and nothing else.

 

[RavenGuard]

Originally posted by: kenrippy
i would bet (guessing here) that high clock + crappy timings is better. in the real world, cpu clock speed is what matters, ram speed & timings are for benchies and nothing else.

Sorry you must have misunderstood, I was referring to ONLY the memory clock in my question. My CPU is at 2.8Ghz.

 

[kenrippy]

i understood you, sorry i can't actually answer that with confidence.

you could always run some benchmarks to find out for sure though (sandra etc.)

 

[myocardia]

Ken's right, you won't be able to tell the difference either way, unless you're doing benchmarks. With most software, it makes no difference whether the RAM is fast & loose, or slow & tight. With some video encoding apps (but not all), slower & tighter gives a slight advantage. With most games, faster with looser timings will give you slightly higher FPS's. Read this thread for more info: memory bandwidth.

 

[RavenGuard]

So if I ran the memory at 1066 5 5 4 12 vs 620 3 1 1 4 I wouldn't see any difference?

 

[myocardia]

Originally posted by: RavenGuard
So if I ran the memory at 1066 5 5 4 12 vs 620 3 1 1 4 I wouldn't see any difference?

I have no idea. Why don't you run a few benchmarks, and let us know. You're the first person I've seen with RAM that will run 620 DDR2 @ 3-1-1-4. I know when I tested out some video transcoding, with 3-3-3-8 timings @ 533 DDR2, vs 1066 @ 5-5-5-15, the tighter timings provided a very slight advantage, but not enough to really care about either way.

 

[RavenGuard]

Originally posted by: myocardia

Originally posted by: RavenGuard
So if I ran the memory at 1066 5 5 4 12 vs 620 3 1 1 4 I wouldn't see any difference?

I have no idea. Why don't you run a few benchmarks, and let us know. You're the first person I've seen with RAM that will run 620 DDR2 @ 3-1-1-4. I know when I tested out some video transcoding, with 3-3-3-8 timings @ 533 DDR2, vs 1066 @ 5-5-5-15, the tighter timings provided a very slight advantage, but not enough to really care about either way.

Can anyone provide for me links to programs I could use to benchmark?