Overclocking and Noise Results
We wondered if the backplate might influence the 380X’s overall performance as we performed our infrared temperature measurements. Based on the GPU clock rate during the gaming loop, AMD's Radeon R9 380X doesn’t throttle quite as much with the backplate in place, even after the card reaches its maximum temperature. The periods of throttling we do observe can be replicated and occur at almost exactly the same times during our taxing load. Subjectively, only the most pronounced dips can be perceived as stuttering during gameplay.
During our torture test, the frame rates we achieve are a bit higher with the backplate installed as well. The plate's benefit isn't profound, but it does show that there is a reason to cool this part of the graphics card.
Overclocking
If you're wondering why we don't give overclocking its own detailed chapter, take a look at the graph at the beginning of this page. Even a moderate overclock to 1100MHz results in so much throttling that the frame times suffer. The average frame rate inches up a bit, true. But more frequent stuttering and the exploding power consumption make overclocking this card beyond its factory setting pointless.
Fan RPM and Noise
Our fan speed measurements are interesting because rotational speed relates directly to the graphics cards’ temperature, clock rate and power consumption. It's also the biggest contributing factor to noise output.
The fans kick in just prior to full utilization, and they start spinning at around 600 RPM. After a few moments, they slow back down to approximately 250 RPM. This first boost is barely noticeable, since the fans are very quiet in this RPM range.
Both versions of our benchmark, the gaming loop and torture test, result in an almost identical curve. The fans don’t have to work quite as hard if the backplate is installed, which we've already seen. The difference isn't profound, but the results are audible.
Just how loud of a graphics card is AMD’s Radeon R9 380X, then? It’s certainly not a coincidence that the company sent us a Sapphire card with an almost perfect cooling solution. It certainly makes both organizations look good.
Considering our results, it’s plain to see just how good of a performer Sapphire's take on the Radeon R9 380X is in this arena. The results in a closed case represent the differences, since overall system noise just isn’t very meaningful.
| Ambient Temperature22°C | Open Bench Table,Gaming Loop | Open Bench Table,Torture | Closed Case,Gaming Loop(Above AmbientTemperature) | Closed Case, Torture (Above AmbientTemperature) | Idle |
|---|---|---|---|---|---|
| Sapphire R9 380X Nitro(With Backplate) | 34.7 dB(A) | 38.1 dB(A) | < 1 dB(A) | <= 2 dB(A) | 0 dB(A) |
| Sapphire R9 380X Nitro(Without Backplate) | 35.4 dB(A) | 38.8 dB(A) | < 1 dB(A) | <= 3 dB(A) | 0 dB(A) |
