Garbage collectors trade space for time. If we collect all the time (which requires a lot of processing time), the GC allocator will use the least memory possible. On the other hand, if we have a lot of space to use, we can collect very infrequently and thus be very fast. In general, with enough extra memory, GC can take almost no extra time extra, when compared to explicit memory management. For example, if we are willing to spend 4x more space than usual, garbage collection can run as fast as explicit memory management; if, on the other hand, we don't give it too much space, it will have to run very frequently and thus will demand much more processing time than
explicit memory management.
Faced with this tradeoff, when should we use a garbage collector instead of manual memory management? Generally, use a garbage collector 1) if you have lots of memory; and 2) if we really want to avoid bugs or extra programming effort related to memory management. If, however, hardware resources are the limiting factor and programmer resources are less of a limiting factor, then manual memory management is probably a better idea.