Due to technical limitations of analog LFO's, the waveform amplitude is usually centered halfway.
But, what would be the benefits of having a waveshape that always sits at the edge of amplitude limits?
This drawing shows both cases
(http://i389.photobucket.com/albums/oo339/cpm_04/vault/lfo2.png)
So, usually the LFO is applied to another device or circuit in order to vary some of its behaviour. Some comes to my mind which seems like they will benefit for that fixed bottom reference, regarding to LFO amplitude as the amount we want the secondary circuit to vary.
For example:
FET: usually you have a known voltage threshold from which it starts to increase transconductance. You can start at the cutoff point and go as far as you can
OTA: they're always fed a one-direction current, starting at zero gain, and up from there
LED/LDR: Led brightness looks like a less consistent method, but still theres a thershold point at which the leds starts to get bright
...
i mean, these methods all have a positive scale of action, from zero to its maximum limit, seems like a center pivoting LFO is not the ideal approach
Looks like this "floored" waveform is a win, am i missing anything?
It depends on the effect.
Tremolo, for instance, really needs to be middle based. You want it to be the normal amplitude when depth=0 and to both increase and decrease amplitude with peaks and valleys of the LFO signal. Many tremolos have only signal decrease, and this makes the ear think that the average volume has dropped.
There is an argument to be made for middle based phaser, flanger, and chorus LFO, although it's not as clear cut as for tremolo.
A VCA - sure, you want to start from 0. But that's usually not a simple LFO signal.
Some thought is needed on the details of the effect and what it does to the audio.