Caustic Photon Render Pass

I am using the Bourns non-detented encoder (PEC11-4020F-N0024), with 24 pulses per revolution. I find at a 1ms sample frequency, that the encoders are really "jumpy". When I turn the encoder clockwise very slowly I will see numbers jump, for example: 0,1,2,1,3,4,6,5,6,7,etc I would expect the output to be smooth: 0,1,2,3,4,5,6,7,etc I have tried using various techniques to reduce this bounce with not much success. One I tried was this: http://www.kennethkuhn.com/electronics/debounce.c The integration method he uses seems sound, but I am still getting jumpy output. Do these encoders have too many PPR? Or are my expectations too high? Any ideas?

I have been trying to design an elegant solution for a pedal/damper input that supports both regular "switch" pedals, "half-damper" (potentiometer) pedals, and expression (potentiometer) pedals. See attached. I have a Yamaha FC-3, it has a TRS/stereo phone plug and the wiper is connected to the Ring. I also have a Roland pedal that is a TS/mono phone plug and it is a simple normally closed switch. When a mono plug is inserted into a stereo jack, the Ring is shorted to the Sleeve. So any reasonable input circuit has to take this into account. The problem I have is making the input circuit support both mono and stereo plugs without sacrificing too much "dynamic range" into the ADC, and also not sinking too much current from the power supply rail through a switch-pedal when the pedal is up. My design uses a 1k pull-up resistor between the Sleeve and the reference voltage (3.3V in my case). With a 10k pot inside the pedal, this only sacrifices 9% of the dynamic range (1k res vs 10k pot for 11k total resistance from reference to ground). With a 1k pullup to 3.3V, the switch-pedal will draw 3.3*3.3/1000 = 11mW when the pedal is up. How do the big companies (Yamaha/Roland/Korg) do this?

Hi! First post here! Loving the clean interface of this forum. Great people, keep up the good work! I am thinking of using a 74HC595 as described in the LED rings schematic: 11 LEDs per ring with column drivers, and X number of rings with row current sinks. 1. I believe each pin is sinking too much current. The Fairchild 74HC595 datasheet says 25mA sink or source under Absolute Maximum, MM74HC595 says +-35mA. Given that every 11th LED is on at once, or only 11 LEDs are on at once, this may not be a problem. But it seems like it is awefully close to the Absolute (read danger) rating. 2. It would be cool to apply a PWM signal the Output Enable (pin 13) of the 74HC595 to globally adjust brightness of the LEDs. This should work, right? Thanks again! CausticPhoton