1) If they give you the same value they are defined the same.
2) If they are not supposed to be the same you did something wrong that gave the same value.
3) If you load the same value into a double and a long double, why would you expect 2 different values?
In other words, given your extremely vague description, any answer is possible.
I check the value range of double and long double with vs2010, but it gives me the same value.
Yes, long double is pretty much just a synonym for double to Microsoft's C++ compiler. In fact, Microsoft even recommends against using long double forms of the standard library.
could you please tell me how to use the long double format in vs2010.
The sensible answer is to use a compiler that supports extended precision floating point. You could write conversions to IEEE 80-bit or 128-bit from IEEE 64-bit (the format of double in Visual Studio), but I'd recommend against that if you can avoid it. You could also look for a library that supports extended precision floating point like GMP.