Thoughts on Intel i7 3820 and Mushkin Redline 2133

[FAUguy]

Just to see what would happen, I started lowering my voltage down from 1.650v but kept the same frequency of 2133MHz and timings of 9-11-10-28 2T. Right now it's at 1.590v and seems stable, passed Super PI 32M and Intel Burn Test with 10 runs at Max Ram. I'll keep lowing it down a few notches and see how low I can get it before any problems pop up.

After some more testing, the Mushkin Redline 2133 is staying stable for me down to 1.575v, but the next setting down at 1.550v causes problems, such as PI 32M generating an error and Intel Burn Test not passing. So the difference between running it at 2133 with 1.575v and 1.650v is 0.075v, but I'm not sure how much less heat it would generate as it was just a big warm when ran at 1.650. Still,I guess its best to run it as the lowest voltage as long as its stable.

One other thing.....when using the older 1.30 BIOS on my ASRock Z68 Fatal1ty Pro Gen3 and the 4 sticks of 4GB Mushkin, the command rate had to be set at 2N (2T) for the system to boot at POST, as changing it to 1N would cause it not to boot. Well I tied the 1N with the new L2.04 BIOS and the system booted up just fine and hasn't shown any issues. So it looks like ASRock got it fixed in the BIOS so that 16GB of RAM can use a command rate of 1N.

 

[FAUguy]

Update...
As it turns out, with the lowers RAM voltages, IBT will fail on me after running for 30-60 min, so I'm having to raise the voltage back up. Still have to see if it will fully pass IBT at 1.630v or will have to be raise back to 1.650.

 

[FAUguy]

Update...
As it turns out, with the lowers RAM voltages, IBT will fail on me after running for 30-60 min, so I'm having to raise the voltage back up. Still have to see if it will fully pass IBT at 1.630v or will have to be raise back to 1.650.

IBT did not pass at 1.630v or 1.650v when the Command Rate was changed from 2N (2T) to 1N. So I've changed it back to 2N and will try the lower voltages again.