Why does Stateflow evaluate the condition [x >= (1/2)] differently from [x >= 0.5]?

Question

Brian B 2012-11-4

1
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/52694-why-does-stateflow-evaluate-the-condition-x-1-2-differently-from-x-0-5

I have created a minimal example (available here: http://dl.dropbox.com/u/99981625/fraction_in_condition.mdl) consisting of a single chart with a single state (A) which increments an output variable x in steps of 0.1. A transition from A to A is conditioned on [x >= y], where y is a literal constant. A conditional action resets x to zero. The output variable x is declared as a real double scalar and is given an initial value of zero. If y is the literal "1/2", the condition mysteriously evaluates true even when the debugger shows that x == 0 (or x < 0.5, in general). The conditional action resets the value of x, causing it to remain zero for all time. On the other hand, if y is the literal "0.5", then x increments from 0 to 0.5 before being reset to zero. I expected both conditions to behave as in this second example. Why don't they?

Thanks for any ideas.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Siddharth Shankar 2012-11-4

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/52694-why-does-stateflow-evaluate-the-condition-x-1-2-differently-from-x-0-5#answer_64228

编辑：Siddharth Shankar 2012-11-4

If I had to guess, I would think that 1/2 results in integer division which always yields 0, and this is then cast to double (0.0) for the comparison, by which time the damage is done. If this hypothesis is correct ;), 1.0/2.0 should show you the same behavior as 0.5.