Wiring straight from the encoder does work with some (many) encoders, but not all. You'll need to check the encoder spec sheets and recommended interface. You may need a "line receiver" chip or a pull-up resistor between your encoder and the M-series board.
There are several aspects to the "update speed." First, the actual hardware will track the position instantaneously with each encoder edge. Second, your program will need to make a software call that uses the DAQmx driver to retrieve that position from the board. This is very fast and is unlikely to be problem. A few years back I benchmarked an RT system using the older traditional NI-DAQ driver and recall a time well under 10 microseconds for a call to Counter Read.vi
Third, you mention controlling an actuator based on this feedback. This opens a potentially very big can of worms. If you're using Windows, you simply cannot get deterministic timing from your control loop. You can sometimes get by with control loop rates on the order of 10 Hz or less, but much higher than that and you're almost guaranteed to encounter difficulties. The problem isn't the execution time of the call to retrieve the position data. The problem is when Windows decides to make it.
Finally, your control is likely to work better if you use multiple position samples for calculating your velocity and acceleration. This would lead you toward a hw-timed buffered acquisition of encoder position.
What kind of control do you need to accomplish? How does the actuator work, i.e., how do you set its speed and/or position?
-Kevin P.
ALERT! LabVIEW's subscription-only policy came to an end (finally!). Unfortunately, pricing favors the captured and committed over new adopters -- so tread carefully.
