QUOTE (Ravor @ Oct 31 2008, 02:31 PM)
I would rule seperate tests because I figure at least some of them are trying the same "basic" tricks.
I would make the same ruling for the reverse reason: they're not coordinating their attacks (that would be a teamwork test), so they're likely to approach the problem from different angles, which means the system's full stability measures apply to each attempt. If they were working together to, say, cause a buffer overflow, then the system would be easily overrun. If they try to overflow different buffers, the success of one is unlikely to affect another's attempt (as he won't know the other buffer is overflowed until it's too late to take advantage of it).