but it really depends of the purpose,
until your computation need long because of intermediate approximation and precision limit of the float format. but you'll know that only by trying with float before
if it's 'only' to visually rotate a bunch of vectors, i don't see the need and extra memory of long.
DSP computation is another story and can benefit of long for intermediate result, even if the final number will be a float (as usine TPrecision is a float for now).
Statistics: Posted by martignasse — 10 May 2015, 11:39
]]>
]]>
Now I need more sleepless nights.
Statistics: Posted by oli_lab — 08 May 2015, 15:50
]]>
]]>
Statistics: Posted by caco — 07 May 2015, 14:02
]]>
is there a way to easilly do math directly on 2D vectors ?
for exemple multiply a vector by a scalar, add two vectors, normalize a vector ?
or should I do my own fonctions ?
thanx
Olivar
Statistics: Posted by oli_lab — 06 May 2015, 22:35
]]>