Wallpaper Engine

cherryghost wrote: »

There's surely more headroom in processes used for this. Is it much?

have always hesitated to use live wallpaper applications since they used to be very performance-draining. When researching for perfomance-tests of "Wallpaper Engine" I didn't quite find a review that was useful for me so i decided to get the software and test it myself.
It should be noted that this review will ignore other apsects such as the content of the software.

Hardware Specs:

CPU...............................i5 6400 (no overclock with BCLK)
CPU cooler......................AIO watercooling loop from NZXT
GPU...............................GTX 980TI (MSI - no manual overclock)
RAM...............................16GB DDR4 (GSkill - 2133MHz - dual kit)
Motherboard.....................ASUS ROG Maximus VIII Ranger
SSD................................Samsung 850 Evo (Only for OS, Steam and apps on HDD)
OS..................................Windows 10 Professional

Test:

The software (run in non 64-bit mode) will be tested in two scenarios. Arsenal wallpaper will be used for all tests.

1) Hardware usage will be observed while the System sits idle for 10 minutes without any other applications running in the background. The tool for HW monitoring is "OpenHardwareMonitor". This testing scenario will be repeated for the wallpaper presets Low, Medium and High and a standard windows desktop background as reference.

CPU temperature
CPU load
GPU temperature
GPU load
GPU memory usage

Preset None Low Medium High
CPUTempAvg 21°C 20°C 20°C 20°C
CPUTempMin 18°C 18°C 18°C 18°C
CPUTempMax 25°C 22°C 22°C 23°C
CPULoadAvg 1% 3% 3% 3%
CPULoadMin 0.4% 0.4% 0.1% 0.4%
CPULoadMax 24% 10% 24% 24%
GPULoadAvg 0.1% 7% 10% 15%
GPULoadMin 0.1% 1% 1% 2%
GPULoadMax 5% 11% 16% 21%
GPUTempAvg 35°C 36°C 37°C 37°C
GPUTempMin 34°C 35°C 36°C 36°C
GPUTempMax 36°C 37°C 38°C 38°C
GPUMemoryUse 180MB 185MB 230MB 260MB

Furthermore the RAM usage has been tested, but did not show measureable differences!
Ram usage was always at about 2.2GB.

2) This test will show the impact on gaming performance. Again, Arsenal wallpaper is used, this time on preset High. As there are different options to choose, which define how the wallpaper will react to an application being run in fullscreen (Keep Running, Stop, Pause) a benchmark will be done for every mode and as reference with a standard windows desktop background.
I chose the Benchmark Scene from Rainbow Six Siege with the graphics on ultra preset. Here are the results:

FPS in RB6-Benchmark

RB6S MinFPS AvgFPS MaxFPS
None 102 173 327
Stop 93 172 264
Pause 100 172 296
Keep Running 58 171 299

Conclusion:
As we have been able to see, there are slight differences in terms of hardware usage. I was surprised that the software did much better than expected, but the results may differ, considering community-made wallpapers may use more resources. The CPU temperature values may not be very reliable due to the CPU being watercooled thus these values are at the very low end of the scale.

Wallpaper Engine does also have an impact on gaming performance, which can be kept to a minimum using the "Pause" or "Stop" behaviour for fullscreen applications.

With a hardware config as "enthusiast" as mine, i doubt that there will be serious performance issues with this application, but if you are gaming on a much weaker rig, i recommend turning down the settings of the live wallpapers a bit.

At the end of the day, you will have to decide yourself wether you want to sacrifice resources for cool a$$ (a$$ can be taken literally) wallpapers or not.

I do not regret my 4 bucks on this software and will be using it from now on.