Assume we devise a protocol with the packet size so large that it can carry all chunks of a live or real-time multimedia stream in one packet. Do we still need sequence numbers or timestamps for the chunks? Explain.