So as i understand it (though im not sure if its the same w/ bulletproofs so please correct me if im wrong), a commitment is generated like this;
commitment = (blinding_secret . G) + (value . H)
where G and H are both different generators on the same curve. But why is it necessary to have the second generator, H, instead of just using G twice?
Normally C = xG + aH where a is the amount and x is the blinding factor.
The point of a Pedersen Commitment is to commit you to a certain value of a. If instead you had C = xG + aG, then this simplifies to C = (x+a)G. So you could claim any value of a later by claiming you'd used a different blinding factor. Using H, which has an unknown DL w.r.t. G, prevents you from doing this.
