Printable View
Why is bool 1 byte and not 1 bit? Wouldn't it have been easier (and more efficient - less memory needed and so on) with one bit? In the end - it is either 1 or 0 (true or false).
(Have also heard that some versions of Visual C++ define bool with 4 bytes - why did they do that at all??)
You can't allocate less than a byte.
>>define bool with 4 bytes
word alignment, your compiler will/can do this in different scenarios to make access faster. You can give pre-processor directives to force it to pack things though.
Omg, you are totally right ;) Again, I think I must review some memory management basics. Thanx alot.Quote:
You can't allocate less than a byte.
Its 1 bit 0 or .
Pardon?Quote:
Its 1 bit 0 or .
Bit access (ie bit fields) is slower then byte access. Mainly because it still has to access the bit and then figure which bit you are looking for.
Also remember that everything has to be >= ( sizeof(char) == 1 )
word access is faster on word-aligned machines (most all you are likely to see). If you want to save space, you can do your own bit fiddling. If you are interested in a bitmap class, I created one here: http://sol-biotech.com/code/CPE/PrimeCPE.hpp. It is much slower, but uses much less space. The implementation I used it for was to find prime numbers and when compared to an STL vector it is much slower.
Whoa... Anger... Tension...Quote:
Also remember that everything has to be >= (
For what?Quote:
Originally Posted by sean_mackrory
Thantos inadvertently made an angry face out of his greater-than-or-equal-to equation.
If you have a lot of bools, you could use bitwise operations to operate on single bits, thereby make 8 flags out of 1 bool. Of course it might not be worth the trouble for all the good it does. Unless you've got about a hundred or more, which isn't likely.
Thanks alot guys ;) This really got me on the right track again...
You could use a std::bitset, or a boost::dynamic_bitset if you don't know the size at compile time.