On the Latency of USB-Connected Input Devices

Abstract

We propose a method for accurately and precisely measuring the intrinsic latency of input devices and document measurements for 36 keyboards, mice and gamepads connected via USB. Our research shows that devices differ not only in average latency, but also in the distribution of their latencies, and that forced polling at 1000 Hz decreases latency for some but not all devices. Existing practices -measuring end-to-end latency as a proxy of input latency and reporting only mean values and standard deviations -hide these characteristic latency distributions caused by device intrinsics and polling rates. A probabilistic model of input device latency demonstrates these issues and matches our measurements. Thus, our work offers guidance for researchers, engineers, and hobbyists who want to measure the latency of input devices or select devices with low latency. CCS CONCEPTS• Human-centered computing → Interaction devices; Laboratory experiments; • Hardware → Communication hardware, interfaces and storage.

Yet Another Latency Measuring Device

“…Whereas for capacitive touch screens, input events can be easily triggered by grounding a probe on the screen [8,14], it is not trivial to accurately measure the exact time of a physical button being pressed [13,21,34]. Therefore, we present a method to measure end-to-end latency of traditional desktop setups (button-equipped input device, computer, display) which combines parts of multiple approaches from related work.…”

“…Burke [2] measured the latency of computer mice by detecting the point in time when a button contact is closed and measuring the delay until a USB event is received by a logic analyzer. Wimmer et al [1,34] extend this approach by replacing the professional grade lab equipment with a Raspberry Pi and automatically triggering buttons so measurement series' can be conducted automatically. For this method, it is required to attach wires to a button of the device under test by either soldering or using clamps.…”

In COM we trust: Feasibility of USB-based event marking

“…Although the relatively widespread MCU evaluation boards such as Arduinos can be used for the same purpose, many labs shy away from them because of uncertainty as to whether or not events can be marked with the required precision. The same holds for commercial USB input/output products for which no published timing test results are available (Wimmer et al, 2019).…”

DispLagBox

“…End-to-end latency is comprised of multiple partial latencies such as input device latency, processing time of the computer, network latency, and finally display lag -the latency added by the display. It is important to measure and understand those partial latencies to find bottlenecks influencing the end-to-end latency of a system instead of regarding the whole process from user input to system output as a black box [12]. Wimmer et al have developed a cheap and easily replicable device for measuring the latency of input devices, such as computer mice, keyboards and game pads [2,12].…”

“…It is important to measure and understand those partial latencies to find bottlenecks influencing the end-to-end latency of a system instead of regarding the whole process from user input to system output as a black box [12]. Wimmer et al have developed a cheap and easily replicable device for measuring the latency of input devices, such as computer mice, keyboards and game pads [2,12]. We propose a similar approach for measuring the latency of computer displays in this paper.…”