This is a question to which there is no simple, easy answer.
The dual core Intel does not amount to "5600" or twice the performance of one 2.8GHz core. The reason for this is the instructions making up computer code must be executed in order. This sequence of execution is called a "thread."
When you run several programs on a computer with a single processing core and they appear to be all running at the same time, that is not what is really happening. Each program has at least one thread. The processor quickly switches between threads, executing each one at time. Because threads are switched many times per second, you don't notice it -- just like you don't see individual frames when you watch a movie.
With two cores, you still can't have both cores working on the same thread and getting it done twice as fast. However, both cores can each be working on different threads, at the same time.
Let's say we have a program that uses only one thread and performs some long calculation. If it takes 10 seconds to complete on a single 2.8GHz core, it will still take 10 seconds to complete with two 2.8GHz cores -- because the single thread can run only on one core at a time.
If we run two instances of the same program together, with a single core they both might complete in 20 seconds (the processor time will be divided between them). However, with two cores, they both might still complete in the same 10 seconds it took to run just one.
If we were able to design our program to evenly divide the workload internally and run two threads -- yes, one program can internally use multiple threads -- then a single instance of the program would still take 10 seconds on single core, but might complete in 5 seconds with two cores.
What if we can't separate the workload evenly? Let's say 70% of our calculation has to be done consecutively, in a single thread. Then one instance of the program, with two cores, would take 7 seconds.
Some applications use multiple threads internally, some do not. Some are able to split the workload evenly, some are not. This gets very complicated very fast, after you factor in the Windows operating system itself running multiple system threads along with your applications, at all times -- and the fact threads may have to wait on each other for things like reading and writing files to disk.
Andrew did a good job explaining the present limitations of 64 bit architecture and these are the limitations of dual cores. You are probably now more confused than ever.
If it helps any, the benefit of the second core is usually estimated to be 20-50% depending on the software mix you are running. Going with that figure leaves both of your choices looking about equal.
