Sensor and Performance

The Pugio II is equipped with the PixArt PAW3335. According to specifications, the 3335 is capable of up to 16,000 CPI, as well as a maximum tracking speed of 400 IPS, which equals 10.16 m/s. Out of the box, four pre-defined CPI steps are available: 400, 800, 1600, and 3200.

Disclaimer: All testing has been done using a new firmware that has been conceived based on my feedback on a previous version. I do not know at which point this firmware will become publicly available.

CPI Accuracy

"CPI" (short for counts per inch) describes the amount 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 vary from unit to unit, so your mileage may vary as well.


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 rather inconsistent. While 400 and 800 CPI are perfectly on point, 1600 and 3200 CPI aren't. Overall, this is still a good result and a massive improvement compared to the original firmware I tested.

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 can only 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.

All wireless testing was done using the regular 2.4 GHz wireless mode as I'm unable to test Bluetooth mode due to the lack of a Bluetooth enabled device. That having been said, the lowered polling rate as well as the properties of the transfer protocol make Bluetooth mode unsuitable for anything outside of office usage anyway.

Wired Testing

First, I'm looking at two xCount plots. Typically, any "kinks" showing up within the plot would be indicative of sensor smoothing at that CPI step. As you can see, such "kinks" are quite apparent in the second plot. We'll have to take a look at the xSum plots in order to determine smoothing levels. In any case, with only minor SPI timing jitter to speak of, tracking is quite clean.


Here, I'm looking at xSum plots generated at 1600, 11,700, and 16,000 CPI. The line further to the left denotes the sensor with less motion delay. At 1600 CPI, motion delay is identical; at 11,700 CPI, motion delay is 2 ms; and at 16,000 CPI, motion delay is still 2 ms. There may be some very minor smoothing even before 11,700 CPI (1 ms), but it's difficult to say for sure. Either way, smoothing is minor across the board and only kicks in late to any substantial degree.

Wireless Testing

Tracking quality takes somewhat of a hit in wireless mode. Both plots show increased SPI timing jitter and the occasional dropped poll, with the effects being more pronounced in the 16,000 CPI plot.


Alright, what are we looking at here? These two plots show the same mouse movement against the G403, plotted as both xCount and xSum. Note the kinks that show up yet again in the xCount plot. If you look closely, you'll see that each of them corresponds to a change in motion delay differential in the xSum plot. In fact, the differential appears to get smaller or larger with every kink. I have no idea what's going on here, but it definitely does not make it easier to determine the actual wireless delay. Depending on which part of the plot I'm looking at, the wireless delay varies between 2.5 and 1.5 ms. After looking at the data extensively, I've come to the conclusion that wireless delay is somewhere between 1.5–2 ms across the entire CPI range, with the delay being closest to 2 ms most of the time. I can only speculate whether the wireless delay would be any lower with a wireless extender/receiver that would reduce the distance between mouse and USB dongle.


For the sake of completeness, here are the specific tests done at 1600 CPI, 11,700 CPI, and 16,000 CPI. The motion delay differential (sans wireless delay) in wireless mode at 11,700 and 16,000 CPI is greater than in wired mode. Additionally, it appears the amount of smoothing is not only increased, but also kicks in earlier. Due to the varying amount of wireless delay, I'm unable to determine the precise values, however.

Speed-Related Accuracy Variance (SRAV)

What people typically mean when they talk about "acceleration" is speed-related accuracy variance (or short SRAV). 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 the G403. As you can see from the plot, no displacement between the two cursor paths can be observed, which confirms that 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 4 m/s (which is within the proclaimed PCS range), at which speed no sign of the sensor malfunctioning can be observed. We can see the same visible kinks in the xCount plots above, however.

Polling Rate Stability

For wired mice, polling rate stability merely concerns the wired connection between the mouse (SPI communication) and the 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/wireless 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. As you can see, several 2 ms reports are visible, which confirms that the polling of the entire signal chain is not in sync.



Second, I'm testing general polling-rate stability of the individual polling rates in wireless mode. Running the Pugio II at a lower polling rate can have the benefit of extending battery life. 250 Hz already shows some variance, 500 Hz a bit more, and 1000 Hz is messy. Furthermore, at the 1000 Hz setting polling frequently drops to 500 Hz (pictured below), which is why I'd recommend sticking to 500 Hz in wireless (2.4 GHz) mode. Looking at the graphs, it becomes clear that the kinks seen in the xCount plots above correspond to polling outliers. These are restricted to wireless mode though, so the Pugio II appears to have issues with keeping a fully consistent polling rate when used in wireless mode.

Paint Test

This test, done in wired mode, indicates any potential issues with angle snapping (non-native straightening of linear motion) and jitter, along with any sensor lens rattle. I'm testing 1600 CPI as a general-use baseline, 11,600 CPI as the last step without any smoothing, 11,700 CPI as the first smoothed step, and 16,000 CPI as the highest CPI step. As you can see, no issues with angle snapping can be observed at any of these CPI steps. No jitter is visible at 1600 CPI, but there is moderate to high jitter at 5000 CPI, which decreases quite significantly at 11,700 CPI, where smoothing is first applied. 16,000 CPI doesn't look much different. Overall jitter levels are very reasonable, especially compared to a sensor with more smoothing, such as the 3389. Lastly, there is no sensor lens rattle.

Lift-off Distance

The Pugio II offers two pre-defined LOD values to choose from. At the "low" setting, the sensor tracks at a height of 1 DVD, but not at a height of 2 DVDs. This does not change when using the "high" setting. Keep in mind that LOD may vary slightly depending on the mousing surface (pad) it is being used on.

Click Latency

Since mechanical switches are being used for the buttons in most computer mice, debouncing is required in order to avoid unintended double clicks. Debouncing typically adds a delay (along with any potential processing delay), which shall be referred to as click latency. As there is no way to measure said delay directly, it has to be done by comparing it to a control subject, which in this case is the Logitech G203. Using the latest firmware, click latency in wired mode has been measured to be roughly +3.0 ms compared to the SteelSeries Ikari, which is considered as the baseline with 0 ms. For whatever reason, I was unable to measure anything in wireless mode. I can report, however, that the main button switches still run at the specified 3.3 V in wireless mode. Comparison data comes from this thread as well as my own testing, using qsxcv's program.