XSPC Memory WaterBlock & Side Plate Set Review

Testing
In an attempt to try to get some real world comparisons for what a ram block can do for you, I first extensively overclocked my ram modules on air.

The test rig consists of the following components:

Processor: Intel Core i7 3770K @ 4.8 ghz
Motherboard: Gigabyte G1.sniper 3
Video Card: Evga GTX 690 with XSPC Razer GTX690 full cover block
Memory: Samsung Green 1600mhz low clearance ram 2 X4GB
Power Supply: Enermax Max Revo 1300w
Storage: OCZ Vertex 3 120GB X 2 Raid 0 Running Windows 7 64bit
Cooling: Custom water cooling loop. 2 x 480mm Alphacool NexXxoS UT60 Full Copper
Case: Enermax Fulmo GT

On air I was able to slowly step the ram up to 2200 mhz. I ended with the following timings at 1.51 V:

CL-12
tRCD-11
tRP-11
tRAS-29
CR-1T

XSPC Memory WaterBlock & Side Plate Set

Compared to stock 1600mhz at 1.344 V:

CL-11
tRCD-11
tRP-11
tRAS-28
CR-1T

XSPC Memory WaterBlock & Side Plate Set

Not bad memory overclock at all, but some very loose timing to say the least. Now with the memory watercooled I was really excited to see what it could do!

I was of course able to get it to the same 2200mhz. Unfortunately even with gobs of voltage thrown at it there was no getting to 2400mhz. Instead I stepped down to 2133mhz and tightened the timings for my final 24/7 OC:

1.67 V
CL-9
tRCD-9
tRP-9
tRAS-28
CR-1T

XSPC Memory WaterBlock & Side Plate Set XSPC Memory WaterBlock & Side Plate Set

Although the bandwidth is lower than the air overclock, the tighter timings give it a better AIDA64 score. I also ran it through the highest memory settings in IBT to check for stability and it passed without a problem.

And now for the ageless question, is a ram block really worth anything to a watercooler? Well I was able to bump the voltage up much higher under water than air. Truly though the only thing pushing me to do that was the knowledge that the ram should be running cooler. Unfortunately memory doesn’t come with built in thermal probes to tell us it’s temperature. So there is no way to know for sure whether it helped us. So to answer the question, not really, or at best maybe.