Mike S

had a little search for this in the forum but couldn't find exactly what i was looking for

i have always been curious to know what the difference is between the two methods of adding signals i have shown below

i am working on a project that adds quite a large number of signals together, i was wondering what is the best way (sound quality) to go about adding the signals? i imagine i'll have to do some reading up on dithering too, are there any quick and easy solutions?

adding-signals

Is it? I would say that the first case adds four signals together to which 0 is added: 4 additions. In the second case 3 additions are required. Or: better not to leave the second inlet empty.

or why not use an empty abstraction with only an inlet 

or why not use an empty abstraction with only an inlet 
and an outlet to sum the signals. 

inlet/outlet don't sum the signal. I wouldn't go that way.

no but somewhere they will always go into the next DSP 

it is not that i dont understand the his question. maybe

it is wrong that do that here, but i wanted to point out 

that you do not need *~ at all for summing in practice.

(sometimes i think i should post when bored.)

no but somewhere they will always go into the next DSP 
object, dont they. 

it is not that i dont understand the his question. maybe
it is wrong that do that here, but i wanted to point out 
that you do not need *~ at all for summing in practice.

(sometimes i think i should post when bored.)


let me clarify my first response. The difference between the 2 solutions shouldn't be considered as problematic performance wise. That doesn't mean that summing is not a good idea. For instance, the patch bellow is bad, because the summing only happens in the receive~ object, so the sum is done twice.