Hopefully, what we'll see in the future is an increased focus on spectral composition as opposed to brute force. I've had more than one highly spirited discussion with reefers that are PAR-obsessed without asctually understanding what PAR measures. The critters we keep do not respond equally to the full range of the PAR spectrum (~400nm - 700nm). Put a lightsource outputting 1200 PAR of yellow light on your tank and get back to me with the results.
As was pointed out earlier, LED's by nature are very narrow band emitters. That's not natural and neither particularly helpful. Our critters have specific (and somewhat varied species-to-species) spectral requirements that align with the zooxanthellae they host and, in some cases, the organism's own biochemistry. Aligning a palette of narrow-band emitters to those needs is not only impractical but probably unachievable.
The LED industry is focused on manufacturing environmental lighting --- warehouses, homes, stores, signage. We appropriate their devices for our (unintended) use. They focus, naturally, on 'white' light. And as was pointed out they do this by doping narrow-band LED's with phosphors --- exactly as is done with T5's and other fluroescent lights. By combining different phosphors, manufacturers give us our different T5 'colors' (MH uses different combinations of gases to sculpt the spectrum).
LED manufacturers can do the same thing. They are just not motivated (yet) to do so. They could give us all manner of 'whites' with spectral compositions that compliment our tanks' needs if they so chose. So for me, it's not about chasing the ultimate high-PAR LED as much as it is giving us a greater range of choices when it comes to spectral composition. Look at the spectral graphs of the three 'whites' from which we can choose and compare them to output plots from various T5 and/or MH bulbs. You'll see what I mean.
