|
Testing:
To test the EliteXStream, we are going to use the following setup:
| CPU: |
Q6600 @ 3.56GHz |
| Motherboard: |
DFI Lan Party P35-T2R |
| GPU: |
OCZ 8800GTX, 621MHz-972MHz |
| RAM: |
4x1GB Crucial Ballistix, DDR2-1066,
2.23V |
| Case: |
Danger Den Torture Rack |
| Sound: |
X-Fi XtremeMusic |
| Cooling: |
Swiftech H20-220 Compact |
| Hard Drives: |
1x500GB
Seagate Barracuda
2x150GB Raptor X |
Voltages:
First off, we are going to
measure voltage stability. Previously, I would use SpeedFan's
built in logging capability to capture the voltage readings in
realtime, while applying various loads to the system. This
time, however, I am using OCCT's tool since it provides better
resolution. OCCT also provides its own loading tool for the
CPU. For the GPU, I used the 3DMark06 test "Deep Freeze", with
8xAA, 16xAF, at 1920x1200. I randomly applied loads over the
course of 120 minutes to see if I could get any significant
variance.

Here is the 12V rail over the course of two hours. We can see
exactly when the 3DMark06 runs occurred, as we see a 0.05V drop on
the rail as soon as the benchmark goes into 3D. Ideally, we
would want to see a completely flat line, in reality you will see
some variance under different loads. .05V translates into a
0.4% variance.

The 5V rail clocks in with absolute stability across the entire run.

The 3.3V rail also shows some minor sagging at the same time the 12V
rails do. In this case, we are talking about a variance of
0.01V, so not a lot to worry about. At the beginning and end
of the run, where the system is idle, the rail rests a little higher
than under load.
| Ideal |
3.3V |
5V |
12V |
| Measured |
3.36V |
5.11V |
12.17V |
While software monitoring tools are good for tracking variances,
they are not so good at accurately measuring the rails. So, we
use a DMM to measure the rails while idle, and we find everything is
within spec.
Efficiency:
| |
Idle |
Load |
| Corsair HX620 |
310 |
575 |
| OCZ EliteXStream |
309 |
571 |
Next up, we compare the EliteXStream to my current standby, the
Corsair HX620 and see which one is more efficient.
Power usage was measured at the
plug with a Kill-A-Watt meter. Idle is defined as being on the
desktop, and load was achieved by running Prime95 V25.4 and 3DMark06
at max settings simultaneously. Here we can see that the
OCZ beats the Corsair by an hair in both Idle and Load measurements.
Conclusion:
There has been a lot of speculation as to how OCZ power supplies
might be changed once PC Power & Cooling was brought into the wings,
and it appears that an OCZ unit is still of a different breed than
PCP&C. So far, we haven't seen OCZ turn their nose up at large
fans like PCP&C does, though it is a shame that more OCZ units
aren't modular. The performance of the EliteXStream is quite
good, the 12V rail moves a little bit but is well within spec.
Power efficiency is very good, even beating out the highly regarded
Corsair HX620. Noise is nonexistant, even at full load.
Pros:
-
Great performance
-
Efficient
-
Lots of connectors
-
Quiet
Cons:
|
 |
| Performance: |
4 out of 5 | |
Innovation: |
3 out of 5 | |
Quality: |
4 out of 5 | |
Stability: |
5 out of 5 |
| Aesthetics: |
3 out of 5 | |
Software/Drivers Pack: |
N/A |
|
Overclocking: |
N/A |
|
Value: |
3 out of 5 |
|
 |
 |
|
 | Project Skill Level (5 being most difficult) | 3
out of 5 |

|