Tuning Trinity(?)

[MightyMalus]

CAS# Write Latency(tCWL) at 10
CAS# Latency (tCL) at 10
RAS# to CAS# Delay (tRCD) at 10
Row Precharge Time (tRP) at 10
RAS# Active Time (tRAS) at 28
3- And really, what about all the other RAM settings?

Would like an answer for those. Which of the four "10s" are the three everyone mentions?

I'm running OOCT atm, put Turbo off for now, so max is at 3.8Ghz.

Got the iGPU at 1013Mhz and RAM still at 1863Mhz.

I'm still playing with it BUT when I reset the PC is sometimes gets stuck when it shows the ASRock logo...and I have to manually reset. What's up?

BTW, Its scoring on 3DMark11: P1620, G.S-1489, P.S-3843, C.S-1345

 

[Sleepingforest]

From MSI:

Timings: When we are talking about timings, we are talking about the speed at which the memory controller access's reads and writes from one 64bit block to another. When you see ram timings, they are generally represented like this: 9-9-9-24 or CL- tRCD-tRP-Tras... there are more advanced ram timings, but these are the most important

I'd recommend simply setting the tCWL, tCL, tRCD, and tRP identically (to 10 or 11) and the tRAS to three times that value minus 3 (so 27 or 30, respectively).

The fact that your computer is having trouble turning on tells me that your memory is probably not stable. Run Memtest and see if you can run it for 10 hours without error (I doubt you can). You may have to bump the voltage to 1.65V to run stably.

 

[MightyMalus]

Didn't give me an issue now, checked the timings and all were 10s-and 27. Running Memtest now, three windows...

Do I really gotta wait that long? It usually takes like an hour or so to go past 100% coverage...no errors yet. 10% and rolling.

(Weirdly enough, my laptop is very error prone on Memtest, yet never had an issue with using it...)

 

[Sleepingforest]

Passing 1000% is preferable and 200% should be a minimum--so 2-10 hours. The percentage isn't exactly how stable your memory is (in other words, 100% does not mean perfectly stable).

Read this (the preambles, split across multiple posts, are enough; the exact programs he recommends are a bit out-of-date). Take note of this in particular (it applies to most stress tests):

There is a very common misconception that if your machine can pass Prime95 stability testing for, say, four hours, your machine will be able to run stable, regardless of what you are doing, for four hours as well, without issue. This is simply not the case.

Prime95 often finds errors in its 16th - 20th hour of testing, a potential for instability that wasn’t found after only four hours of testing. After only four hours of Prime95, the potential for instability still exists. 24 hours is widely viewed as a sufficient time period to catch any instability that may be present, but by all means test longer if you are able.