And now it is time for the fun stuff. Namely, I'll be strapping this kitty cat to a rocket and seeing if it can keep up. Oh, and I'll be load testing the Cougar as well. Now, don't go calling the cops on me, it was just a joke.

Load testing for the Cougar will be done with the usual gadgetry. As always, the SunMoon SM-268 will be the crocodile to my Steve Irwin, with two other supplementary 12V loads helping out so I can monitor this kitty in true quad 12V form. Other gear consists of a Brand 4-1850 power meter, a dual probe thermometer, a USB Instruments oscilloscope, and my good multimeter.

First up will be a set of five balanced progressive tests and two crossload tests. These will be done at room temperature with respect to the maximum ratings given by the label. These tests will be done again later in the hot box to see how the unit does when heated, but before that happens I'll do my usual low load test and power on spike suppression test.

Now that the initial testing at room temp has been done, there are a few things to mention about the above table. It's a nice color scheme, isn't it? All those shades of gray really work well for the site. And those borders, look how even they are. I should be talking about other things, you say? Well... ok then.

Let's take a look at efficiency first, because it's the best thing about the numbers I can see up there. Not only has it met Silver like a good little kitty, it has decided to go bringing the awesome in test two with an astounding 92.6% figure. Folks, I've never seen that before, to my knowledge. Even the Gold rated Seasonic X-650 didn't pull that number off, nor did that other Gold unit I tested, the OCZ Z Series 850W. That's fantastic, HEC... I wonder if you could get Gold out of this thing too without too much tweaking.

Sadly, the unit descends from the lofty heights of efficiency into some more mundane numbers as we look at the voltage readings. The 12V stability is not the best I have ever seen, but it is still very good at around 3% on the worst of them. And the two VRM supplied rails? The ones that were supposed to give us "most stable performance, and perfect regulation?" Uh, not so much. Both were outside 4% regulation. I've seen much better from units that didn't use the VRM approach.

Moving on to the low load tests, where I examine units using a load level that the folks at 80 Plus don't test for, we get almost 81% efficiency from this unit at a 9% load level. While this doesn't sound all that spectacular, consider this - most SMPS designs do not do well at low loads. Efficiency tends to drop off the cliff the lower the load. This unit is actually doing better than average here.

Overshoot Transient Testing - Cougar S700
VSB On	VSB to Full 12V	Off to Full 12V

In these tests, I loaded the Cougar to full power. Then, I flipped the power switch on while getting the scope to watch for turn on spikes at 5VSB. After that, I moved over to the 12V1 rail and did pretty much the same thing there - first with the unit powering up from standby and then from a completely powered off condition. The resulting turn on spikes can be seen in the above scope shots.

Now, ATX spec tells us that we cannot get spikes that go higher than 10% over mean value, or go into the negative side of the scope shots. How did this unit do? Well, if you look at the above shots you will see that the 5VSB rail's spike only hit 4V while the 12V1 rail stayed below 8V. That is better than average on both monitored rails, but not the best I've ever seen.