Cooling, Noise, and Power
For my last set of tests, I like to take a look at different aspects of a card beyond its processing power. Here I test power usage, noise, and cooling performance. All three of these are tests that help show us the difference in cards with the same GPU. So, in this case, I will be focusing a lot on the difference between the EVGA, Zotac, and Founders Edition cards. For my first test, I take a look at power usage. To get the result below I use a Kill-a-Watt on our entire test bench and while running Valley Benchmark I document the higher wattage pulled for the entire system. The result I ended up with was a little surprising. Remember the EVGA GTX 1060 SC has a higher overclock than the Zotac and with that, I expected it to pull a little more wattage. But it actually came in a touch lower than the Zotac.
My next test is our noise testing. Here I use a decibel meter sitting 6 inches away from the fan side of the card while it's on our open air test bench. I test with the fan set to 50% and 100% fan speed and document the result. This test isn’t an exact representation of what you will experience in an enclosed case because the meter is much closer than your ear ever will be, but it does help us catch the differences between cards. In this case, the EVGA with its fan turned all the way up ended up down on the low end of our chart, near the Founders Edition. The 50% result was similar as well. I should point out that the GTX 1060 SC as it was shipped to us had a firmware that the fans weren’t set to turn off at idle loads so unlike most of the other cards our sample could have made some noise at idle, but I never noticed it. I’m told cards that are being made now have the new firmware with 0db turned on.
For my last tests I wanted to test to see just how warm the GTX 1060 SC would run. To heat things up I use Valley Benchmark running on a loop and I test the card both with the stock fan settings and again with the fan turned up to 100% fan speed. This way we can see what you should expect in normal use and also see if the cooler is close to its limits. With the stock fan settings, I was surprised to see the GTX 1060 SC come in with our lowest temperature. When I turned the fan up to 100% there was also still a little room left in it as well. In both tests, it was better than both of the other GTX 1060’s tested. BUT I didn’t leave it at that. When I found out that EVGA would be shipping future GTX 1060 SC’s with the new 0db firmware I downloaded it and put it to the test as well. The new result was significantly worse. We went from 61 at load to 74 degrees at load in the same test. 74 isn’t the end of the world, the Zotac had a similar result but I was surprised at the new fan profile. People on the EVGA forums were as well, it seems the fan profile could use a small tweak. For me I actually went back to the original firmware, being an ITX card I wanted to upgrade our Lunchbox 3 build that my wife uses at LANs and I can’t afford for it to be running that warm. So be warned, some of the cooling results you find on other reviews and even in our chart may not represent the temperatures you will see if you buy a new card.