in the 1070 local time, I find some error when converting
double to int,

int main()
{
double f=0.12;
int t;
t=int(f)*60+(f-int(f))*100;
cout<<t<<endl;
return 0;
}

I expect to get 12, but the result is 11, many compilers
were tested, can anyone tell me how to avoid this?

I am afraid you van do nothing except smth like round
(0.12*100)

You see, because double values are really stored in binary
format, numbers that can't be presented with  a degree  2
denominator (2,8,128,etc) are presented by approximation.
I think 0.12 is really something like  0.1199999????.

Anmd his is the reason