There's a huge concern regarding decimal to float conversion in PHP, because of precision issues.
I wonder why nobody talks about using the decimals as strings and do all the calcs in that way. It seems that PHP works pretty good that way:
"1.123456" / "1.123456"
"1.123456" * "1.123456"
"1.123456" + "1.123456"
"1.123456" - "1.123456"
All those work perfectly fine.
So, what are the cons?
By the way, Laravel's Collection has a method "sum", which might rely on strings to do the operation.
For instance:
// It does the math on MySQL
$orders = OrderModel::sum('total');
// It does the math on PHP, not on MySQL
$orders = OrderModel::get()->sum('total');
Laravel doesn't warn about this. And it seems nobody either. So I might assume is totally reliable.