Originally Posted by WheelNut

Consider this quote from the Novak website:

"The time required to attain the rotor position from the Hall Effect sensor is on the order of a few micro seconds. If a motor's rotor is turning at 80,000 RPM, that would be 1333 revolutions per second, or 1.33mS per revolution. Compared to a microprocessor running at 20MHz (or 26,600 clock cycles @ 50 nanoseconds per cycle), that is a very long time, which means that the microprocessor is capable of executing many instructions during that time frame."
Source: http://teamnovak.com/tech_info/view_article/24

Theoretically there should be no loss in timing with RPM increases. The example provided by Novak considers a motor at 80,000rpm, which in North America is very uncommon since we mostly all run some type of stock motor . A 17.5t spins about 17,000rpm +/- 2000.