I took a whack at a PWM scheme using a 555, however the first draft of the circuit was non functional, and the 555 got pretty hot so I just assumed I had cooked it and went on looking for other solutions.
I still had resistors and transistors, so I thought to myself "Even though I have never read about ANYONE dimming an LED with a transistor... I know that the brightness of an LED is controlled by the current supplied to it and that a transistor acts as a current amplifier. So, I hooked up a quick test circuit with a potentiometer connected to the base of a common emitter configuration, and lo and behold, that shit dimmed beautifully.
I'm going to pursue this circuit, as some weird stuff happens currently. Right now the board I built is supplied with 9 volts, and the 2n2222a transistors I had on hand have a maximum base-emitter voltage of 6v, so if you turn the pot too far, bad stuff happens. (Oddly, once you go too high, the pot starts to catch fire. WEIRD)
I'm gonna invest in some transistors with a higher max base-emitter voltage, and see if that will clear up the catching fire (and obvious 'you just maxed out the transistor' brightness drop) situation.
Overall it's remarkably robust and offers fairly fine control over the brightness of an LED. If I can get un-lazy I definitely want to look into uC stuff, and digital pots. Make some sweet shit I think. Though, if there was a uC involved PWM becomes a lot easier....