Can firmware hacks change the shortcomings of the D600 or is it crippled by hardware?

Marianne Oelund wrote:

Let's think about what it takes to achieve a "good" 1/8000 sec shutter speed:

Time t=0us:  First shutter curtain is released to start moving.

Time t=125us:  Second shutter curtain is released to start moving.

Time t=2500us:  First shutter curtain reaches bottom of sensor.

Time t=2625us:  Second shutter curtain reaches bottom of sensor.

Sounds simple.  However, what would happen if the travel rate of the second curtain was out of calibration by just 1%?  For example, instead of taking 2500us to make the trip, let's suppose the second curtain is 1% slower and takes 2525us to travel across the sensor.  This would cause it to reach the bottom of the sensor 150us after the first curtain, instead of 125us - resulting in almost a 1/3-stop exposure discrepancy across the image.

Keep in mind that the shutter-curtain speeds are established mechanically, by springs and levers.  If a 1% cal error produces a 1/3-stop exposure error at 1/8000 shutter speed, you can appreciate how difficult it is, to assemble and set up these shutter assemblies.  Time is money in production, and adopting a 1/4000 sec speed limit for the D600 probably yields a significant savings. 
- Marianne