Frequency, Temperature & Noise Results
Influence of Temperature on Performance, GPU Boost Clock Rates & Voltages
We’ll start by examining the temperature curves that were generated during the gaming loop and stress test’s 15-minute warm-up phases. The GeForce GTX 1050 Ti Gaming X 4 GB ends up at 62 °C during the gaming loop (63 to 64 °C in a closed PC case), and 64 °C during the stress test (66 °C in a closed PC case). Those are good results.
The curve seemingly cuts off somewhere around 62 to 64 °C. This is primarily attributable to MSI’s default temperature target of 64 °C.
The observed GPU Boost clock rate starts out at a surprisingly high 1835 MHz, but then drops by approximately 40 MHz as the GeForce GTX 1050 Ti Gaming X 4 GB warms up. This difference is barely noticeable when it comes to performance, and can safely be ignored.
We set the fans to a more aggressive profile during our overclocking efforts, hoping to push MSI's card below 50 °C and eliminate thermal throttling. Again, the highest stable overclock we achieved was 1911 MHz, regardless of whether the power target was set to 110 or 125 percent.
The voltage results follow suit. An initial 1.1 V falls to 1.075 V as the card warms up, which is exactly what we'd expect based on the lower GPU Boost clock rate. This means that GP107 operates at a higher voltage than the GP106 GPU used in Nvidia’s GeForce GTX 1060 (we measured a maximum voltage of 1.062 V for that one). It’s interesting to see Nvidia allowing somewhat higher voltages to hit their performance targets; remember that GP107 comes from a different foundry and is manufactured using a 14 nm process.
Temperatures
Our infrared picture shows that the temperature under the GPU's package is in line with the processor's internal sensor.
The voltage regulators and memory modules stay well within a safe range, thermally. The temperature right underneath the memory modules reaches just 60 °C, which is a far cry from their 85 °C maximum. At 70 °C, the voltage regulators are practically cold.
During our stress test, the area around the five GPU voltage regulators hits 73 °C. The hottest memory module tops out just under 62 °C. It's safe to say there's nothing wrong with MSI's cooling solution.
Fan Curves & Noise
The two large fans turn at roughly 900 RPM during the gaming loop and 1000 RPM during the stress test. This is practically inaudible from 50 cm away. Nearer to the card, the fans are more likely to be felt than heard. Ultimately, almost any other active component in your PC will be louder than MSI's GeForce GTX 1050 Ti Gaming X 4 GB.
As usual, the noise measurements are performed using our water-cooled PC in our sound-dampened anechoic chamber built for just this purpose. The setup's practical lower limit is 22 dB(A) due to the measurement system's hardware.
We actually need this low measurable noise level for once. The GeForce GTX 1050 Ti Gaming X 4 GB barely makes any noise, and it would certainly not be loud enough to get clean readings using regular measurement equipment. Levels under 30 dB(A) in a normal environment can’t be distinguished anyway.
For more information, check out Measurement Science: Taking Accurate IR Thermal Readings.
| Test System And Equipment | |
|---|---|
| Microphone | NTI Audio M2211 (with Calibration File, Low Cut at 50 Hz) |
| Amplifier | Steinberg UR12 (with Phantom Power for Microphones) |
| System | Graphics Card Test System with Optimized Water Cooling - Intel Core i7-5930K @ 4.2 GHz, Water-Cooled - Crucial Ballistix Sport, 4x 4 GB DDR4-2400 - MSI X99S XPower AC - 1x Crucial MX200, 500 GB SSD (System) - 1x Corsair Force LS, 960 GB SSD (Applications, Data) - be quiet! Dark Power Pro, 850 W Power Supply Unit (PSU) |
| Water Cooling | - Alphacool VPP655 Pump (Undervolted) - Alphacool NexXxos CPU Cooler - Phobya Balancer - Alphacool 24 cm Radiator - 2x 12 cm Noiseblocker eLoop Fan @ 400 RPM |
| Software | Smaart v.7 |
| Measurement Chamber | Custom-Made Proprietary Measurement Chamber, 3.5 x 1.8 x 2.2 m (L x D x H) |
| Measurement Position | Perpendicular to Center of Noise Source(s), Measurement Distance of 50 cm |
| Measurement Data | - Noise Level in dB(A) (Slow), Real-time Frequency Analyzer (RTA) - Graphical Frequency Spectrum of Noise |
The fans don’t spin when the card is idle, so we're skipping that result. During gaming, we end up with a reading of 28.8 dB(A), which is even lower than MSI's GeForce GTX 1060 Gaming X 3G’s 29.3 dB(A). Across the frequency spectrum, the card peaks on the low end. This noise originates from the bearings. The actual fan noise is drowned out by the card’s overall quiet whisper and ambient noise.
Combined with our temperature results, we can safely conclude that the cooler is large to the point of being oversized. MSI clearly doesn't need such a beefy solution.
MORE: Best Graphics Cards
MORE: Desktop GPU Performance Hierarchy Table
MORE: All Graphics Content