I heard talk of these years ago, but have yet to see one. The theory is simple: 1. LEDs have the fastest on/off time (nanoseconds) of any light source. 2. Pulsing does NOT shorten LED lifespan like it might wiht other light sources. 3. It takes a plant so many milliseconds/microseconds to process a photon, so there is no benefit to a continuous light source from the aspect of photosynthesis. 4. Finding the proper on/off time could save lots of power - probably as much as 70%. 5. LEDs driven harder (more light ouput) have shorter lifespans due to heat. 6. Pulsed LEDs have plenty of time to cool off when pulsed; therefore can be driven at maximum output with zero degradation in lifespan. 7. Heatsinks are a large part of the cost of LEDs and could be shrunk 70% or more in size. In the interim, I have accidentally come up with my own pulsed LED. I got a poorly engineered Chinese LED driver (duh!). I think the LED draws maximum current (maximum brightness) which trips the overcurrent protection circuit in the driver which then resets and tries again. So what I have is an LED that turns on and off about once a second- which is nowehere near optimal. It is a 20w 630nm RED LED and it blinded me with four white paper towels over it during testing. Ouch! Because the unit did not work as advertised, I got it for free and figured I might as well use it rather than scrap it. It has been running for several days with no problem - and unsurprisingly runs very cool.