Sensor and Performance

The ROCCAT Burst Pro Air is equipped with the PixArt PAW3370. According to specifications, the 3370 is capable of up to 19,000 CPI, as well as a maximum tracking speed of 400 IPS, which equals 10.16 m/s. Out of the box, five pre-defined CPI steps are available: 400, 800, 1200, 1600, and 3200.

All testing was done on the latest firmware (1.0008). As such, results obtained on earlier firmware versions may differ from those presented hereafter.

CPI Accuracy

"CPI" (short for counts per inch) describes the number of counts registered by the mouse if it is moved exactly an inch. There are several factors (firmware, mounting height of the sensor not meeting specifications, mouse feet thickness, mousing surface, among others) which may contribute to nominal CPI not matching actual CPI. It is impossible to always achieve a perfect match, but ideally, nominal and actual CPI should differ as little as possible. In this test, I'm determining whether this is the case or not. However, please keep in mind that said variance will still differ from unit to unit, so your mileage may vary.


I've restricted my testing to the four most common CPI steps, which are 400, 800, 1600, and 3200. As you can see, deviation is exclusively positive, highly consistent, and quite low, which is a good result overall. In order to account for the measured deviation, adjusted steps of 400, 800, 1550, and 3100 CPI have been used for testing.

Motion Delay

"Motion delay" encompasses all kinds of sensor lag. Any further sources of input delay will not be recorded in this test. The main thing I'll be looking for in this test is sensor smoothing, which describes an averaging of motion data across several capture frames in order to reduce jitter at higher CPI values, increasing motion delay along with it. The goal here is to have as little smoothing as possible. As there is no way to accurately measure motion delay absolutely, it must be done by comparison with a control subject that has been determined to have the lowest possible motion delay. In this case, the control subject is a G403, whose 3366 has no visible smoothing across the entire CPI range. Note that the G403 is moved first and thus receives a slight head start.

Testing is restricted to 2.4 GHz mode as Bluetooth is not suitable for non-casual gaming applications.

Wired Testing

First, I'm looking at two xCounts plots—generated at 1600 and 19,000 CPI—to quickly gauge whether there is any smoothing, which would be indicated by any visible "kinks." While the 1600 CPI plot does not show any kinks indicating smoothing, the 19,000 plot does, indicating smoothing at this step. Furthermore, both plots show woefully poor tracking.


In order to determine motion delay, I'm looking at xSum plots generated at 1600 and 19,000. The line further to the left denotes the sensor with less motion delay. 1600 CPI does not show any motion delay differential. At 19,000 CPI, a motion delay differential of roughly 2.5 ms can be observed, which is due to the added smoothing at that step.

As of the latest firmware, 1000 Hz wired is no longer stable, which is why wired mode behaves similarly to wireless mode (2.4 GHz) in this regard, and the observations above could no longer be made.

Wireless Testing

Tracking continues to be just as bad as in wired mode.


Once again, 1600 and 19,000 CPI are tested. Determining motion delay in wireless mode on the Burst Pro Air is complicated by the fact that polling is highly unstable. As a result, motion delay is variable, and its degree depends on whether the polling interval is 1 ms, 2 ms, 3 ms, or 4 ms at any given point. The plot below may serve to illustrate this behavior:


With the polling interval being 1 ms, and keeping the motion delay differential in wired mode established above in mind, I can measure a wireless motion delay of roughly 1 ms.

Speed-related Accuracy Variance (SRAV)

What people typically mean when they mention "acceleration" is speed-related accuracy variance (or SRAV for short). It's not about the mouse having a set amount of inherent positive or negative acceleration, but about the cursor not traveling the same distance if the mouse is moved the same physical distance at different speeds. The easiest way to test this is by comparison with a control subject that is known to have very low SRAV, which in this case is also the G403. As you can see from the plot, no displacement between the two cursor paths can be observed, confirming SRAV is very low.

Perfect Control Speed

Perfect Control Speed (or PCS for short) is the maximum speed up to which the mouse and its sensor can be moved without the sensor malfunctioning in any way. I've only managed to hit a measly 5 m/s, which is within the proclaimed PCS range and causes no observable sensor malfunction.

Polling Rate Stability

Considering the Burst Pro Air is usable as a wired mouse as well, I'll be testing polling rate stability for both wired and wireless use.

Wired Testing


All of the available polling rates (125, 250, 500, and 1000 Hz) show severe instability. Polling stability tends to not deteriorate further when using RGB lighting.

Wireless Testing
For wired mice, polling rate stability merely concerns the wired connection between the mouse (SPI communication) and USB. For wireless mice, another device that needs to be kept in sync between the first two is added to the mix: the wireless dongle/receiver. I'm unable to measure all stages of the entire end-to-end signal chain individually, so testing polling-rate stability at the endpoint (the USB) has to suffice here.


First, I'm testing whether SPI, wireless, and USB communication are synchronized. Any of these being out of sync would be indicated by at least one 2 ms report, which would be the result of any desynchronization drift accumulated over time. Polling is unstable to where one has to assume some sort of desynchronization drift is present.


Second, I'm testing the general polling-rate stability of the individual polling rates in wireless mode. Running the Burst Pro Air at a lower polling rate can have the benefit of extending battery life. All polling rates display major instability. Even after a firmware update nominally improving polling stability, stable polling is still nowhere to be seen. On the Burst Pro Air, RGB lighting cannot be disabled fully, only its brightness lowered, which is why determining whether the instability is introduced by the RGB lighting is difficult to impossible. For what it's worth, with all LEDs set to white, which prevents PWM from taking place, stability is slightly improved:


Furthermore, it appears that polling stability is a function of the current velocity of the mouse, which is why moving the mouse at higher velocity will generally reduce the number of off-period polls. When moving the mouse at >3 m/s for the most part and using all-white RGB lighting, the following plot can be achieved, which represents the best case scenario:


I've also conducted the same tests while using a USB extender and a different USB port, and the results are the same.

Paint Test

This test is used to indicate any potential issues with angle snapping (non-native straightening of linear motion) and jitter, along with any sensor lens rattle. As you can see, no issues with angle snapping can be observed. No jitter is visible at 1600 CPI. 10,000 CPI already has smoothing applied and shows minor jitter, which is amplified at 19,000 CPI. Lastly, there is very minor sensor lens movement.

Lift-off Distance

The Burst Pro Air offers two pre-defined LOD levels to choose from, along with the ability to perform a manual calibration. Using the "very low" preset, the sensor does not track at a height of 1 DVD (<1.2 mm). This does not change when using the "low" preset. Keep in mind that LOD may vary slightly depending on the material of the mousing surface (pad) it is being used on.

Click Latency

Most gaming mice use mechanical switches for their buttons. By wiring the switches of the test subject together with the switches of a control subject, I'm able to measure click latency very accurately; i.e., standard error of around 0.05 ms. However, this method is not applicable to mice with non-mechanical switches and wireless-only mice in general. As such, other methods ought to be employed, one of which is NVIDIA's Latency Display Analysis Tool (LDAT). The LDAT allows me to measure the entire end-to-end latency between the mouse click and resulting photon transition on the monitor. By establishing the relative difference to a control subject, I'm able to provide values I consider sufficiently accurate; i.e., standard error of around 0.2 ms. The Razer Viper 8K has been posited as the baseline for being within 0.1 ms of a hypothetical absolute minimum. Many thanks go to NVIDIA for providing me an LDAT v2 device.

Click latency has been measured to be roughly +6.5 ms, with standard deviation being 2.6 ms. The debounce time setting has no effect on click latency on press (button-down). As polling stability is a function of the current velocity, actual numbers may be lower with the mouse in motion. Furthermore, as polling stability is also a function of the currently active RGB lighting effect, using a non-PWM effect such as all-white (Fully Lit) will result in lower click latency (roughly 1 ms lower in my case). Keep in mind that due to the indicated value being neither the absolute click latency nor the measured end-to-end-latency, standard deviation ends up looking disproportionally large. Comparison data comes from my own testing and has been exclusively gathered with the LDAT.