Maybe a reflash of the software is in order. If that doesn't work, I can help you reprogram a different pin for the button input. :S

Grounding seems to work best. I increased the resistance to 10k with no change at all... so now i need a NC button for it to work though I've been using alagator clips to simulate lol. It just seems weird that after fondling with the mod wheel would do such a thing.

I DID however change the 2k resistor to a new one and it worked for maybe a few button presses. it would do 2-3 cycles, then one cycle, then started to go through all of them... wierd...

The input of your microcontroller is hanging open which means it will oscillating between high and low. Grounding it is fine, and pulling it high is fine (granted you're confident in your power supply).

I can only suggest to replace the button, increase the resistor's value, and to check the trace from button to microcontroller. Also make sure you are pulling the trace to the right voltage level. I don't remember if it is high or low off the top of my head.

Hello, havent posted in quite some time
I made an Arduinoboy a few hours ago, mainly playing with the MIDI in function with mGB. Everything worked beautifully, until I played with the mod wheel on my microkorg. I played with it for maybe, 30 seconds, going wild with it. and then the DMG crashed. Naturally I rebooted the DMG, only to find that the Arduinoboy only cycles through modes over. and over. and over. So I took out the pushbutton and replaced it. Nothing. Swapped the 2k ohm resistor for a 2.2k ohm resistor, and nothing (also tried a 1k). I then received results after grounding the switch pin of the arduino (pin 3 i believe). It stopped cycling. Is this safe? Is the atmega fried?