If you run amplicon specific standards with your amplifications, you can calculate copy numbers directly from the regression curves and need not worry about sequence or size specific inconsistencies with your PCR. 

You do not have to worry about amplicon length, because you are not looking at fluorescence levels per se but at the rise of fluorescence during every cycle. As long as your amplicon is not longer than 200 bp, it will double in amount during the exponential phase and this is the phase you use for quantification. As stated above, you correct for housekeeping genes to correct for variations in the amount of starting material between samples.

Dear all, thanks for your answers. I got the thing that I do the normalisation with the endogenous control, that much I know about PCR ;-) However what if I now besides comparing target 1 in sample 1 with target 1 in sample 2, also would like to compare target 1 in sample 1 with target 2 in sample 2? The thing with the standard curve, I would first have to make plasmids containing my amplicons right? So therefore I would prefer a mathematically approach. Regarding the answer of Christina Vogelaar, what do you mean by rise of fluorescence? One does actually measure when the fluorescence level has reached a threshold and then see which cycle it is, so therefore we measure fluorescence levels each cycle? I think I did not completely get your answer right, unfortunately. Kind regards, Chris