Originally Posted by popups
Isn't the ADNS-3090 a 900 pixel array? I can understand their reasoning for using 800 over 900.
I figure SteelSeries is just scaling the 800, 1600, 3200, 4000 SROM only through their own coding. Logitech scales two setting for their version to cut down coding and level performance. SteelSeries can be doing the same thing. This would make those in between steps perform like the next highest. This is what I have seen in most mice, maybe with the exception of Razer.
Well, data allocation for this particular mouse is a bit awkward compared to typical solution. I'm just noting that those new values are in line with the sensor architecture itself
Agilent had a max registry of 2000 not so long ago via ADNS-6010 (Based on the 3090 architecture)
Logitech is more in line with traditional scaling where the value closest to matrix scale (800) yields highest speed. Alternative 3600 CPI registry emulates a larger scale, ultimately reducing max IPS tracking as the size is 4* bigger. Halved values 400/1800 register identical max IPS vs native reg.
I have a Kone pure Optical here and estimated tracking performance on my Goliathus Speed goes as so. 800 DPI, 4.4m/s > 1200 DPI, 3m/s> 1600 DPI, 2.25m/s and so forth. Would indicate matrix enhancement outside MCU recalculation.
I don't know why the 800 value is choosen, but it seems common to compensate among many similar cmos architectures. PTE also increases a base 800 resolution via similar method ^^Edited by Skylit - 8/18/13 at 10:57pm