Furmark
FurMark is a very intensive OpenGL benchmark that uses fur rendering algorithms to measure the performance of the graphics card. Fur rendering is especially adapted to overheat the GPU and that's why FurMark is also a good stability and stress test tool (also called GPU burner) for the graphics card.
Version History
1.20.4.0 (January 21st, 2019)
- added NVIDIA GeForce RTX 2060 support
- updated: GPU Shark 0.12.4.0
- updated: ZoomGPU 1.22.5 (GPU monitoring library).
- fixed the GPU temperature reading on some Radeon GPUs like RX 400 series
- GPU temperature spike filter is now disabled by default.
- removed support of Windows XP.
1.20.2.0 (January 6th, 2019)
- added NVIDIA GeForce RTX (2080, 2080 Ti, 2070) and Titan RTX support.
- added Radeon RX 590 support.
- in GPU monitoring log file, GPU-Z values are used when GPU temperature, GPU core and mem clocks are 0.
- GPU data log polling factor set to 1 (was set to 10 in previous versions). Now GPU data are logged every second when Log GPU data is checked in Settings box. This value can be changed in the config file (log_gpu_data_polling_factor attribute).
- updated: GPU Shark 0.12.2.0 and GPU-Z 2.16.0
- updated: ZoomGPU 1.22.2 (GPU monitoring library).
1.20.1.0 (June 27th, 2018)
- added minimal high-DPI support. FurMark is no longer scaled (blurry effect) on high-DPI devices.
- updated: GPU Shark 0.11.2.0 and GPU-Z 2.9.0
- updated: ZoomGPU 1.21.7 (GPU monitoring library).
1.20.0 (January 23rd, 2018)
FurMark 1.20.0 is a maintenance release and brings the support of new GPUs (NVIDIA Quadro P1000/P4000, GeForce GTX 1060 5GB, TITAN V) as well as 21/9 resolutions. GPU monitoring utilities GPU Shark and GPU-Z have been updated to their latest versions.
The affinity mask of the render thread is now set to 0 by default which means that Windows is free to run the render thread on any logical core of the CPU. In previous versions of FurMark, the render thread was forced to run on the second logical core.