Power Consumption and Temperatures

EVGA isn't playing around with the cooling solution. While it did get slight heatsink shrinkage from the Z690 Classified, it still retained the dual fans and heatpipe going to the chipset. The backplate also serves a practical purpose beside some minor passive cooling. It keeps the motherboard rigid, which helps from any PCB flexing that may occur from the weight of larger graphic cards.




For the EVGA Z790 CLASSIFIED thermal testing, one probe is placed along each bank of power stages. A probe is left out to log the ambient temperature. For temperature measurement, a Reed SD-947 4 channel Data Logging Thermometer is used, paired with four Omega Engineering SA1 self adhesive thermal couple probes. All temperatures are presented as Delta-T, which is the recorded temperature minus the ambient temperature as a base. The end result accounts for variation in ambient temperature, including changes over the course of a test.

Tests are conducted over a 30 minute period. For testing, the Intel i7 12700K is overclocked to 5 GHz at 1.35 V to reach 300~ Watts. Next is the Intel 13900K, used in a stock configuration. Tests include two runs; one with a fan placed on the VRM heatsinks to simulate case airflow and one without. If the heatsink has a internal fan, it is not disabled for these tests. The EVGA Z790 CLASSIFIED VRM fans don't have options to adjust their RPM, they were left as-is during testing.


Prime95 has long been used as a torture test for reviewers to keep a sustained maximum load. This program isn't a realistic use-case for the average consumer. Historically, this test has been used to represent a "worst case scenario." However, that has recently changed, as the 13900K can be equally power hungry (if not more) just in a stock configuration compared to a overclocked i7-12700K CPU setup.

Without a additional fan pointed at the VRM heatsink, the built-in fans were keeping busy. We will see in the next chart that the auto configured fan curve is set to around 65 °C target with no ability to adjust this.


By allowing a "unlimited" power target (P1=4096 W), this test actually draws more power. We can see the dips in load as the CPU bounces off the 100 °C limit. Similar to the previous test, the built-in VRM fans seem to have a target of around 65 °C.


The last test here is the gaming scenario. This is an easy way to gauge what the VRM temperatures can be after extended periods of play. The same thing plays out here. Even without any simulated case airflow, the VRM sticks to the 65 °C range. The heat generated will have to exit the computer case at some point, but at least a high-airflow setup isn't required at all.