The conventional overclocking by means of an even higher power limit would be a bit pointless, because the feed bowl is already filled magnificently from the factory to the top edge. Without further voltage increase, the card ran smoothly and stable at 1425 MHz and if you give full pipe, even up to 1450 MHz.
The 1475 MHz as the peak also ran in most games, but after a long time it was sporadically free. The memory ran stable until 2250 MHz, but then the temperature of the modules was already at the upper limit of 85°C.
I now tabulate the achieved start and end values for temperatures and GPU clock (boost), but also point out that these clock rates can of course be higher in loops with a significantly lower load.
Initial |
Final value |
|
---|---|---|
Open Benchtable | ||
GPU Temperatures |
44 °C | 74 °C |
GPU clock | 1380 MHz | 1380 MHz |
Ambient temperature | 22 °C | 22 °C |
Closed Case | ||
GPU Temperatures |
45 °C | 75 °C |
GPU clock | 1380 MHz | 1380 MHz |
Air temperature in the housing | 25°C | 44°C |
Overview Graphs: Temperatures vs. Clock
For better illustrations now again the respective courses considering our timeline of a total of 15 minutes each for the warm-up time. Interesting is the clearly visible difference of the heating between the open structure and the closed housing, to which we will come back again in a moment. However, the clock rates remain surprisingly the same, so the 1380 MHz ex works is completely preserved for the benchmarks against his Steam friends.
The Torture Loop doesn't look much different, because open and closed constructions hardly differ. However, the clock always stays above 1300 MHz, which is due to the very high power limit.
Thermal analysis of the back of the board
Finally, we consider the thermal analysis of the respective load states. Since the Beckplate does not have a cooling function, it is easy to remove.
Gaming
We see that the card has no problems with the gaming loop with Witcher 3. With just under 77°C below the package in an open structure, as well as the almost 85°C in the voltage converters, the world is quite fine.
When used in a closed housing, we then interestingly do not measure higher temperatures in the voltage converters and only the area under the package and the memory increases slightly. This is accompanied by the faster rotating fans, because MSI has set the temperature limit rather tightly. Whether this then weighs very negatively in the operating noise or not, see or we'll hear it right away.
Stress
The stress test is significantly higher in terms of power consumption compared to normal gaming, which is also reflected in the temperatures. Well over 30 watts more supplied electrical power also want to be dissipated cleanly as heat converted cleanly.
In the closed housing, this time it is a few degrees more at the neuralgic points – except for the voltage converters. And of course this includes the necessary speeds of the fans – but we'll get to that.
And what do we conclude from this? The cooler cools as it should, despite the immense power dissipation ex works. How loud is that going to be? I'll tell you that right away.
Warming and cooling
The last two pictures show where exactly the heating starts the most and where the board is cooled most effectively.
- 1 - Einführung, Unboxing und Daten
- 2 - Platinenlayout und Spannungsversorgung
- 3 - Gaming-Performance 1920 x 1080 Pixel (Full-HD)
- 4 - Gaming-Performance 2560 x 1440 Pixel (WQHD)
- 5 - Leistungsaufnahme im Detail
- 6 - Temperaturen, Taktraten und Wärmebildanalyse
- 7 - Kühlerdetails und Geräuschentwicklung
- 8 - Zusammenfassung und Fazit
Kommentieren