PixArt PAW3395 and PAW3950 Optical Sensor Architecture and Surface Tracking: Fundamental Mechanics and Deep Mathematical Modeling

[Sarah Morris]

So I’ve been pulling my hair over the raw image acquisition rates on these two sensors, and I want to hear from people who’ve actually stress-tested them in competitive settings. The PAW3395 tops out around 5000 FPS in most implementations, while the 3950 claims something closer to 26000 FPS—when you’re running 1080p at 360Hz, does that extra bandwidth actually matter, or are you bottlenecked by polling rate and motion interpolation before the sensor’s native capture speed even becomes relevant? For me, during fast 180-degree flicks in Valorant, I’ve noticed micro-stutter that seems to correlate with surface texture more than with any firmware setting.

On the LOD side, I’ve noticed the 3950 feels like it has a much more aggressive “dead zone” when lifting off my pad compared to the 3395. How are people tuning lift-off distance on the newer sensor, and is the difference purely mechanical, or is there signal processing happening in the interpolation layer that’s discarding low-amplitude motion data?

🎮 Battlestation & Rig: Ryzen 7 7800X3D | RTX 4080 Super TUF | 32GB DDR5 6000 CL30 | 2TB 990 Pro | LG UltraGear 27GR95QE-B 27″ OLED 240Hz | SteelSeries Apex Pro TKL Wireless + Zowie EC2-CW + Zowie G-SR-SE Bi | Beyerdynamic DT 990 Pro (250 Ohm) + FiiO K5 Pro ESS DAC

📍 Munich (Germany)

"Smooth frame time pacing matters infinitely more than peak FPS."

[Author]

Posts