Another possible reason for the need to go with an RF laser..... Back in an earlier post (
http://www.buildlog.net/forum/viewtopic.php?f=6&t=36&start=20#p501 I provided the measurements coming out of my DSP controller during a raster. The power level was set to some constant power and the TTL line would bang on and off in order to tell the laser to fire. This TTL bit banging was occuring as fast as 15us on-off. The laser would turn on at the constant set power during the time period of the TTL high. In order to have true analog levels of power output, the DSP controller would need to be putting out the TTL for on-off (and) it would need an analog level coming out for setting power.
I have already proved to myself that I can go full-on and full-off within 15us on the DC CO2 tube. It could then be assumed that the tube can change power from one setting to another within 15us, but I think it is beyond the ability of the present state of the art for the power supplies to handle this control input. Somewhere I thought I read about tens of millisecond response times to a power level command change. I'm sure this number came directly from the PWM to analog conversion time of the Chinese power supplies, basically an RC time constant.
The RF is probably much easier to control. You can adjust the intensity of an RF amplifier in nano-seconds, and turn it on and off with a PIN diode (also within nano-seconds). Perhaps this is the reason the Synrads are so popular (and expensive).
There is good info at the FAQ section of Gantryco.
http://www.gantryco.com/faqs-3d-256-grayscale-graphics.html