Uno r3 or Beetle CM-32U4 + A4998 max. pulse rate?

Hello:

I know stepper motors but have been avoiding microcontrollers for years.

I just got an Uno r3 and and some A4998 stepper boards. I also got a DFR CM-32U4 to commit to a project once I figure out the basics of making a bipolar stepper do what I want.

I don't know what microstepping rate or speed yet, but the purpose will only be to rotate 180 degrees to open a rotary mechanism (one motor, one axis) then stop, and rotate 180 degrees more in same direction (closing the mechanism) when prompted (both via user pressing a push button).

The other function will be to rotate 360 degrees in response to a second push button, opening and closing a mechanism in one cycle, then stop.

The speed of the 360 degree rotation will whatever I eventually find I can achieve. The microstepping ratio will depend on how smooth (vibration-free) the motion can be made (TBD).

I know there are limitations that depend on the mechanical properties of the system. Since I don't know those yet, I'll just assume as an ideal case I can use accel & decel library code to avoid resonance, skipped steps, etc. Assuming no problems for the sake of reality-checking the electrical limits of the uC,

Does anyone have an idea what either uC and A4998 will limit ‘top speed’ to (best case)?

I think I read 800 pps somewhere and at 1/16 microstepping, that is a 50 Hz or 3000 rpm. 

The half rotation speed is not really important 

The full rotation speed will not be about to be faster than 1/50 second. Assuming 50% duty cycle, the mechanism will be open for half of that, 1/100 second.

This may be unrealistically fast for a stepper and microcontroller.

NOW I can tell you what this is. A motorized drive for a shaft-propelled large camera shutter. It can be operated by twisting the shaft with fingers, which is not a number, I am well aware.

A slower speed may have to be accepted.

My question is whether anyone has already discovered the Uno/Beetle limitation of pulse rate to a stepper motor driver (this would be the best case. Resonance, inertia, vibration & any other mechanical headaches will have to be figured out experimentally. There is no easy math answer).

If it ends up being ½ second instead of 1/50, well, that's life. I just want to keep an eye on reality.

Thanks 

Murray