But the rocket is going to feel weightless in those 22 beautiful milliseconds coasting down the exit tube, every single resonant mode ringing out loudly after dropping that 70,000,000 pound [lateral load] tether before gracefully slipping [axial load if you ain't sideways] into beautiful New Mexico air at MACH-SIX-AND-A-HALF. I don't have the reference materials or the will to estimate supersonic drag, but I know they gotta kill off all rotation with the release mechanism because 3000 to 5000 mph on a 100m centrifuge is 250 to 425 rpm, and a 425 rpm speed flat spin at Mach 6 ends exactly how everyone expects it to. Also "Actuation accuracy <1 millisecond" sounds like a design requirement, not something they found out how to do yet. 1ms off at 425rpm is 2.5° of travel. Much more and you miss the exit.That gets us spit out the end of the centrifuge where we go from vacuum to atmosphere and 10,000g to one (do you like how there's no deceleration there? No accommodation for that transition? 'cuz I do)