Brief question: faster to zero before direct computation?

Hello all, quick question,

in the simplest of examples,

x1a = linspace(-1,1,100);
y1a = zeros(1,100);

y1a = x1a.^3;

Is it a clear computational speed and economy advantage to declare y1a first with zeros, or not, in this simple case?

Is the cube operation one that does not require nor benefit from variable declaration?

Cheers

[per isakson]

• y1a = zeros(1,100); offers no advantage. It adds to the execution time.
• "cube operation" is not affected by "variable declaration"

[Miguel]

this reply is a bit confusing, you are saying it offers no advantage, correct?

[per isakson]

Not just confusing. It was wrong; a "no" was missing. I've edited the comment.

[Miguel]

Ok then, I reckon this answers it, I would like to know, however, if the statement y1a = x1a.^3 is something that builds the y1a variable one by one increasing its size after each value is computed, or, what I am suspecting is the case, applies the operation first to the x1a matrix, and then assigns this to variable y1a?