Main > Main Forum
Trackball problems - any tips for me?
markronz:
Ok, one thing to add. In the trackball, there are three rollers, of course. I was watching it closely to see if I could figure out what was happening inside there. To get a perfect upward direction, the one roller with the optical wheel on the top needs to spin. As a result, the diagonal roller (the one without the wheel) should also spin. These two rollers make the cursor move completely straight upwards. BUT what happens when I push straight up is that the third roller, the one that controls left and right motion, also moves slightly. If I put my finger onto the left and right roller to hold it in place and then spin the trackball, the cursor goes straight up, like I want it to.
So I'm not sure what to fix here, but it seems really touchy in there. When I push the ball straight up, I would expect the third roller would stay in place.
Any thoughts here whats going on or how to help me?
RandyT:
--- Quote from: markronz on February 02, 2011, 03:48:33 pm ---What am I doing wrong here guys? Do your trackballs go completely straight up and stay straight when you give them a good push? Should I just give up here? I am at a complete loss.
--- End quote ---
I think you may be "pushing a rope" for the most part. There is a mechanical dynamic that is pretty hard to fight in a 3-roller system. It's not terribly unlike expecting the ball to behave exactly the same way, regardless of the direction you try to spin it. It's just not going to happen. Marble Madness even uses the trackballs in a different than normal orientation, presumably to counter this. Every trackball I have seen so far has a little bit of drift. It makes perfect sense, because there is going to be a differing amounts of friction based on the direction of the spin in relation to the geometry of the rollers.
Making the rollers spin as freely as possible is the best you can hope for. Outside of that, you'd need to design a new trackball.
RandyT
newmanfamilyvlogs:
I haven't thought this all the way through yet... but I wonder if something akin to a 'deadzone' word work... make it dynamically adjusted based on the speed in the other axis... Essentially turning down the sensitivity setting as an inverse function of the velocity of the other, scaled logarithmically so that as the dampened axis' velocity increased proportionally to the non-dampened, the dampening effect is not as strong.
Though I suppose this wouldn't be very 'authentic'.
markronz:
--- Quote from: RandyT on February 02, 2011, 04:49:06 pm ---
--- Quote from: markronz on February 02, 2011, 03:48:33 pm ---What am I doing wrong here guys? Do your trackballs go completely straight up and stay straight when you give them a good push? Should I just give up here? I am at a complete loss.
--- End quote ---
I think you may be "pushing a rope" for the most part. There is a mechanical dynamic that is pretty hard to fight in a 3-roller system. It's not terribly unlike expecting the ball to behave exactly the same way, regardless of the direction you try to spin it. It's just not going to happen. Marble Madness even uses the trackballs in a different than normal orientation, presumably to counter this. Every trackball I have seen so far has a little bit of drift. It makes perfect sense, because there is going to be a differing amounts of friction based on the direction of the spin in relation to the geometry of the rollers.
Making the rollers spin as freely as possible is the best you can hope for. Outside of that, you'd need to design a new trackball.
RandyT
--- End quote ---
Thanks Randy I suspected as much. Any tips for me on how to get the bearings to spin better? They spinned better before I put on the 3-in-1 oil, should I remove that? If not, is it just a matter of spinning it longer? Should I try to rig something up with my drill where they spin for a more extended period of time?
markronz:
--- Quote from: cotmm68030 on February 02, 2011, 04:55:14 pm ---I haven't thought this all the way through yet... but I wonder if something akin to a 'deadzone' word work... make it dynamically adjusted based on the speed in the other axis... Essentially turning down the sensitivity setting as an inverse function of the velocity of the other, scaled logarithmically so that as the dampened axis' velocity increased proportionally to the non-dampened, the dampening effect is not as strong.
Though I suppose this wouldn't be very 'authentic'.
--- End quote ---
I think I followed that. That would work although I'm not sure I would know how to accomplish that. For me, having it work in games is more important than authenticity.