My hope is that the GPS vertical accuracy is going to get to +/- 30 mm or less but that is pushing the normal GPS vertical accuracy envelope. Using normal GPS survey techniques and then checking the vertical accuracy using optical surveying on control points showed up maximum errors of up to +/- 60 mm here. That was not the accuracy from machine control while moving but GPS surveying involving stopping on the control spot for say around 1 minute with the survey gear then comparing that to optical (dumpy level ) checks. That is a concern for me. What are you finding for vertical accuracy from your experience?
GNSS RTK GPS with a base station within 1km of the field, we think is probably within +/- 25mm most of the time (66% of the time, 1 standard deviation) and +/- 50mm (95% of the time, 2 standard deviations). This means the operator should be on the ball and be aware of the shifts in accuracy.
The critical accuracy is probably during the final trimming operation and the accuracy is better that those given above. The more critical accuracy is the error that occurs over a two minute period which is the time it typically takes to do a final trim from one end of the field to the other.
In this case, the accuracy could be 4x better than what is listed above and in that situation it is very usable and probably better than laser at >300m from the laser transmitter.