![]() ![]() Version -B8 was used for AMD (press drivers).Ī separate bench is used for game performance and for thermal performance. We’ve also automated a significant portion of our testing at this point, reducing manual workload in favor of greater focus on analytics.ĭriver version 378.78 (press-ready drivers for 1080 Ti, provided by nVidia) was used for all nVidia devices. We’ve moved to Ultra graphics settings with 0xAA and async enabled, also dropping OpenGL entirely in favor of Vulkan + more Dx12 tests. DOOM, for instance, now has a new test methodology behind it. For all the excitement that comes with a new GPU test bench and a clean slate to work with, we also lose some information: Our old GPU tests are completely incomparable to these results due to a new set of numbers, completely new testing methodology, new game settings, and new games being tested with. This is our first full set of GPUs for the year, giving us an opportunity to move to an i7-7700K platform that’s clocked higher than our old GPU test bed. GPU Testing Methodologyįor our benchmarks today, we’re using a fully rebuilt GPU test bench for 2017. This fan is capable of 0RPM when VRM component temperatures are low enough (seems to be below ~55~60C), but will often keep its RPM low enough that the noise output is effectively 0 – the radiator fan will run louder. A small back-swept blade fan is mounted atop the aluminum heatsink. The plate contacts the CLC coldplate, using a thick layer of thermalpaste between the two copper surfaces and ultimately sinking to the liquid cooler.įor VRM cooling, EVGA is using an aluminum heatsink and spreader that contacts the inductors and FETs directly via thermal pad. We have testing using both ICX sensors and our thermocouples, for those curious about the impact of this copper plate. This adds an extra ~15W to the heat load, but cools the VRAM modules significantly better than without the plate. This isn’t entirely new construction, though EVGA did somewhat recently make the move to start including copper coldplates for VRAM cooling. Here’s a look at the card (from our tear-down video): If framerates are all you care about, check our SC2 review. For this reason alone, there’s no point to testing FPS. The PCB is the same, the clocks are the same, and so the gaming performance is the same. For our testing, we look at thermal performance on various board components (not just the GPU), we look at noise, and we look at noise-normalized thermal performance (every card at 40dBA) for cooling efficiency testing.ĮVGA’s SC2 Hybrid is an SC2 in every aspect except for cooling. The difference is in thermals and noise, and most people don’t go too in-depth on either subject. Once we’ve established a baseline framerate for the actual GPU – that is, GP102 – there’s not going to be a whole lot of difference between most partner cards. Although it may feel like one GTX 1080 Ti isn’t too different from the next, that’s only “true” when comparing the least meaningful metric: Framerate. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |