Thermal Performance

Everything needed (monitor, peripherals, motherboard w/CPU and GPU, radiator, PSU, and so forth) was placed in a sealed, climate-controlled box at 25°C +/- 0.05 °C. Each radiator was connected by Koolance QD3 quick disconnects for easier changing of fans and radiator. The flow rate was held at 1 GPM constant flow rate. The CPU, an Intel Core-i7 4770K at 4.6 GHz, running at 1.3 V, was held at a constant load using a custom Intel XTU profile, and the GPU, an EVGA GTX 780 Ti Classified under load from Unigine Heaven 4.0 at 1080p with extreme HD settings, was overclocked and overvolted in such a way that the total system power draw was 550 W as measured by a power meter.

A near-constant heat load into the liquid loop by the addition of the CPU and GPU helped achieve stable liquid-loop temperatures pretty quickly (as measured by three separate in-line temperature sensors hooked up to an Aquacomputer Aquaero 6 XT). The tubing and fittings were insulated by a sleeve heater, but its heat function was not utilized.

Every single measurement was done twice as a means of verification. Any possibility of running a fan outside of its static maximum RPM was minimized by as much as possible by using a comprehensive series of separate fans (Noiseblocker eLoop, B12-1 through B12-4) to cover a broad RPM range while minimizing fan speed variance and, thus, airflow. The results below are the ∆T (coolant temperature - ambient temperature) in °C.


The Black Ice Nemesis GTR 360 is not low-airflow optimized. I expected as much based on my previous experience with the Nemesis GTR 480 and 560. This is a mid-high airflow-optimized radiator that scales like nothing else I have seen past 1200 RPM or so. Based on my results with the 480 mm version, the closest competition is the EK Coolstream XE, which EK intends its customers to use with their high performance Vardar fans. Among the tested 360 mm radiators, the Nemesis GTR 360 handily overtakes every other radiator past 1500 RPM with the NB-eLoop fans used for testing (which in turn were separately tested to be among 2016's average-performing radiator fans), and the difference only increases from there. Hardware Labs says these do really shine with average-performing radiator fans at a whopping 3000 RPM, but that is in my opinion simply too loud for anyone with this setup a few feet away, unless you have closed noise-isolating headphones on. These are apologetically marketed for maximum performance without much regard for noise, and I myself will use these with fans in push-pull to get that thermal dissipation scaling at lower noise levels than with a single set of fans. 38 mm thick fans are also an option, though they tend to either be expensive or hard to get, or cheap - not a lot of good middle-ground options here. I suspect many will be looking at these for smaller cases where radiator space is a luxury - just make sure to account for the radiator's dimensions as well.