ASUS Rampage II Gene mATX LGA1366 Motherboard Review

Heat & Acoustical Testing

We are finally going to get to look at north bridge temperatures of the X58 chipset under load in Windows. The ASUS Rampage II Gene is the first X58 motherboard I have had the ability to look at that allows for temperature monitoring in Windows. This was recently made a big deal with the EVGA X58 3X SLI Classified as there was a lot of concern by EVGA users that the Classified ran hot north bridge temperatures due to the passive cooler. We found the same thing with the DFI X58-T3eH8, and the Classified based on PWM temperatures, but were unable to monitor north bridge temperatures on either board. With the passive cooling and north bridge temperature monitoring on the Gene, we can finally see just what is going on with the north bridge when we remove all fan support and leave the cooling to the heat sink.

First we will take a closer look at said heat sink though.

The north bridge and PWM heat sink assembly is very simple. A single heat pipe connects the two and the mass of cooling fins dissipate the heat. The thermal paste used on the north bridge - seen in the second photo above - is an interesting yellow substance. It dries hard and is quite the pain to get off. Our thermal testing done below was with the stock thermal material shown here so there really is no need to replace this material. It may not look like the norm, but it certainly seems to do its job well enough. We would have liked to see a copper plate sit on the north bridge instead of the aluminum one ASUS went with. Again, however, the thermal testing below will illustrate more than adequate cooling so we can't really demand much else.

The last photo above is showing the underside of the PWM heat sink and although there was decent contact across the MOSFETs, we can still see that the impressions are far deeper on the outsides of the heat sink compared to the middle. This means that the heat sink is bowing a little bit and lifting up in the middle. This is a common problem that plagues many PWM heat sinks. It will be interesting to see which manufacturer comes up with an innovative solution to allow more even contact across MOSFETs.

The MOSFETs lying beneath the PWM heat sink power the CPU and are identical to those that are located near the north bridge. These MOSFETs located near the north bridge are responsible for the VTT being supplied to the CPU's uncore but interestingly enough, are not cooled by anything. We would have really liked to see heat sinks on these MOSFETs but again, the testing results we are about to look at all but eliminate the need for these to be cooled as there were no stability issues, even with very little airflow over the motherboard.

The last photo above is the result of running the Gene on a styrofoam sheet. We have never quite encountered anything like this in all the years of running motherboards on styrofoam inserts, but boy did we get a little heat on the underside of the Rampage II Gene. Apparently the MOSFETs on the back side of the motherboard get hot enough to melt plastic and gave us a fresh protective plastic layer over this area. The white styrofoam insert was heavily melted and browned in this entire area. I guess it is time to get a couple more bench tables because this is definitely not something we will want to duplicate. So for those running the Gene on an open bench setup, make sure you have some room for airflow on the back of the board as it gets toasty.

The first photo above is of the setup with three fans, and the second photo on the right is the setup with a single 120mm fan pushing air through the TRUE CU. We won't be able to monitor the PWM temperatures as ASUS doesn't have that temperature reporting to the BIOS, but the north bridge does and that is what we are most interested in. It should also be interesting to see how the south bridge reacts with no fan support since it is not connected to the north bridge, unlike many other motherboard heat sink designs.

We will be using OCCT to load the system for 20 minutes which includes a minute of no load at the beginning and four minutes of idle at the end. This will let us see how quickly heat is dissipated in both setups. Let's look at the results of the north bridge temperatures first.

I think the chart is pretty telling in just how much additional airflow over the north bridge heat sink helps in cooling. With only a single fan pushing air through the TRUE, as may be the case in a normal setup, the NB climbs to over 70C at its peak and barely dissipates any of that heat after the system goes back to an idle state. With our three fan setup, the temperature barely moves and has a delta of about 5C with idle temperatures recovered within the four minute cool down period. This was completely expected, but the biggest item to note is the fact that the system had no problem running the full 20 minutes without the additional cooling. Our Classified testing of the same nature would not allow the system to run under load for more than 15 minutes. Obviously it wasn't the CPU getting too hot as we might have suspected and actually the NB or PWM causing the issues. The Gene seemed fine at these inflated temperatures, but we would still highly recommend some kind of air flow over the NB and PWM heat sinks.

We thought we would also log the south bridge temperatures and we are sure glad we did. Despite the additional cooling not blowing directly on the south bridge, the temperature differences between the single and multiple fan setups was as dramatic as the north bridge. With our three fan setup, the south bridge didn't register a single temperature change. With only a single fan the temperature quickly gained 20C and did not recover at all during the closing idle state. These results almost indicate that the south bridge temperature sensor isn't in fact on the south bridge and just a surface mount sensor near the south bridge. It is hard to say either way so we poured a little bit of LN2 onto the SB heat sink and the temperature quickly dropped from 38C to 18C. At the same time, the LN2 simply ran off the heat sink and boiled off on the PCB around the south bridge so if the sensor is simply on the PCB it could have possibly been hit with LN2 as well. Either way, it just goes to show how much stagnant hot air accumulates on the motherboard without appropriate cooling.

Once again, the passive cooling on an X58 motherboard proves that it is really pushed to the limits without some form of active cooling. Without decent case airflow or fans mounted as we have, you can expect high NB, SB, and PWM temperatures from the Rampage II Gene...like every other X58 motherboard we have tested.