cpu-z ram settings and OC

[ktec]

Here is a current screenshot of my ram settings.

From what I understand JEDEC is what is COULD be.
So my real question is when going into my bios, should I just lower the cas, ras, etc to the Jedec settings or is there more to it? Which profile should I run?
Also, how can I optimize windows7 to fully make use of this 8gb ram?

 

[Cromewell]

Have you set the RAM to manual settings now? It doesn't match any of the SPD settings (which is fine, but some settings are higher than they would be).

So my real question is when going into my bios, should I just lower the cas, ras, etc to the Jedec settings or is there more to it?

Nope, that's pretty much all there is to it

Which profile should I run?

You can likely run #2, but if it's unstable you'll either have to add a little bit of voltage, loosen some of the timings, or switch to using the settings from #3.

Generally, if you want safe settings take the ones for the frequency that is closest without being slower to your actual frequency. i.e. your current frequency is 573.3MHz, the #2 settings are recommended for 533MHz which is a bit slower than what you are running so you need to move to the next set which is for 609MHz

 

[ktec]

The settings it's on right now are most likely from the ASrock feature Turbo Unlock..it automatically unlocked my CPU to 4 core and overclocks cpu as well as ram I think.

Any way to fully utilize 8gb ram to really make things speed up?

 

[ktec]

I OC'd it to this. Left Voltage on Auto, and just upped the speed to one of 4 preset in my Bios (716Mhz) and manually lowered tRCD to 8.

My question now is tRAS went UP to 30 from 20 and tRC went up to 33 from 27 and FSB: DRAM up to 3:10 vs 3:8, so am I better off with original settings or what?
and again, is there anyway to fully utilize this increase in ram (4gb to 8gb) so windows can take full advantage?


also did a MAxmem benchmark and it seems lower than others.

 

[Cromewell]

With most of the timings you'll barely notice the difference in benchmarks. CAS and command rate are the 2 where you can get the biggest improvements, the rest are somewhat negligible.

As long as you are running 64-bit you don't really need to do anything to Windows, it can address all the RAM and applications that can use it will.

 

[ktec]

With most of the timings you'll barely notice the difference in benchmarks. CAS and command rate are the 2 where you can get the biggest improvements, the rest are somewhat negligible.

As long as you are running 64-bit you don't really need to do anything to Windows, it can address all the RAM and applications that can use it will.

Thanks, I understand that timing make a small difference, but what concerns me now is that my benchmark is way below average in all categories.

 

[Cromewell]

Synthetic benchmarks typically respond better to latency changes.

Out of curiousity, what did the score look like before?

edit: looked at it again, doesn't seem too far off. what do you think you should be getting at these settings?

 

[ktec]

look at these results..much higher than mine. ALso ddr3 and my MHz is higher but benchmark is way down
http://www.overclock.net/intel-memory/553039-maxxmem-benchmark-results-thread-12.html
for example:

 

[Cromewell]

I'm guessing your scores appear lower because it is accessing the RAM in a single thread. You are running unganged which means instead of 1x128bit bus you have 2x64bit buses. I know that sounds the same but it is different, the 2x64bit buses work independantly of each other.