| ??? 08/05/07 18:51 Modified: 08/05/07 18:52 Read: times |
#142784 - drop float Responding to: ???'s previous message |
Stanley Lio said:
Now I don't understand what the textbook says about the range of the float variable because the calculated value can still fit in the range of a float variable (1.xxxE-38 to 1.xxxE+38), like "tmp" in the code. Drop float if this is for '51 or any other mcu. Try to stick to the smallest type you can. Long is bad enough, already. Stanley Lio said:
So the complier does not look after that, does it? It is new to me, and I have no idea what level of "intelligence" it has. Well, you'd expect a complier would comply to your expectations, woudln't you... :-) (Sorry, I couldn't stand - this typo happens SO often...) There is no more intelligence in any compiler than its author built in - and don't have too big expectations for them, they are usually happy if it conforms to basic standards and does not eat up all the memory and processor time (some don't even care for that). <flame of the week>And, the language of your choice - C - is inherently stupid by definition, so the compiler cannot remedy that. If you want range checking, go for real programming languages.</flame of the week> JW |
