All I read about is problems, problems, problems…
Should I risk it?
I’d like to have four going at once to control my Johnny 5…
I plan on gutting all four controllers and putting the guts into one “master” console…
Thanks
MJL
The forum is full of problem reports? Ok, but honestly they all have to deal with Atom Pro code and pullup resistors needed, Basic Stamp II code just not working, or XYZ brand controller doesn’t work. The Lynx brand PS2 controller works well with the Atom 28 and the Bot Board. Until the Bot Board II is done, stay away from using the Pro. The Basic Stamp II is dead in the water due to command incompatibilities, and there is no way we can provide support for the dozens of makes and models of these types of controllers. That’s the bottom line. Lynx PS2 (Robot Controller) and the Bot Board with Atom 28… 
Hiya Jim,
Thanks for the clarification…
I’ll risk it and get one to play around with, before purchasing the remaining three…
As I understand it - I can use four (wireless) at one time?
Thanks
Yes, I’m not sure how to communicate to 4 controllers with one microcontroller. I’ll bet it has something to do with the ATT (attention) line. Should be a simple matter of sharing the CLK, CMD and DATA lines and putting the ATT from the receivers on seperate I/O pins. You will likely have to start the transmitters one at a time in the same order each time to sync up with the correct receiver.
It appears that the PS2 receiver module is interfaced via an SPI interface. And I think the Atom interfaced with it may be in master mode. Can you tell me the clock rate or any more about interfacing to the (Lynxmotion) PS2 receiver?
I have read Beth’s paper, and ran down a few other supplied references to the PS2. Even saw something for Dios (sp?), but I need to set up a PIC uP or ARM7. I am familiar with SPI on these processors.
Thanks,
Alan KM6VV
This may help.
sophiateam.undrgnd.free.fr/psx/index.html
I’m not sure about the clock rate, and I don’t have much information beyond what’s in the Atom programs for the robots.
That helps. I figured out that the Atom code was not using the SPI hardware of the uP (on the PIC, anyway), the BASIC interpreter is bit-banging the interface.
I just finished writing my own bit-banging code in C for a PIC processor. I haven’t tried it yet (I’ll put a 'scope on it tonight), but the reference you supplied (thanks!) seems to verify what I worked out for myself. Next week I should be able to try it with a Lynx DoubleShock!
If the interface is bit-banged as opposed to using a uP hardware interface (SPI), then the speed is set by how fast the code can run. The hardware implementations I’ve used always require the user to set up a clock for the SPI, as well as several other parameters. I might have to widen the clock pulse width. Good to do anyway; makes it easier to see on a 'scope!
Thanks,
Alan KM6VV
You should be safe with anything up to 100kbps. SPI comes in three flavors(in hardware). 100k,400k and 1meg. There are alot of non standard variants though. It’s not as tightly controlled as I2C. Just a note the Atom code only runs a about 30kbps(shiftin/out is right around here). AtomPro shiftin/out runs upto 100kbps when using the FAST modes. I’ve tested PS2 controllers on both(FAST and NORMAL on the AtomPro too) without problem. The LynxMotion controller is pretty forgiving. Some other ones are less so.
Thanks. It looked like I could run close to 500Khz on the 'scope, guess I’ll be prepared to slow it down a little. I was running close to 5MHz on the ARM7.
My LynxMotion controller just came today! No chance to hook it up yet.
Alan KM6VV