To monitor my Xbox Live traffic while playing Battlefield 3, I shared my Internet connection from my laptop to my 360, and then monitored the traffic to my 360 IP address using Microsoft Network Monitor. When playing, I found that hit detection was best when there was little or no RTP traffic, and likewise lag was not as problematic. When RTP traffic increased, hit detection became worse and lag increased. The RTP traffic was identified with multiple different payload types: dynamic, DVI4 Audio, CN Audio, G723 Audio, MPA Audio, and others.
Lag has often been explained as a problem with the end-user's connection, however, this is not the case. A friend and I were playing on the same team on the same map and encountered the same problems at roughly the same time from different Xbox 360s at different physical locations on different networks. The only explanation I have is that he and I were both getting hit with the same RTP traffic from the EA server at the same time.
Seeing as how the problem often comes and goes, as I and several thousand others have reported, I am wondering if the RTP traffic could be the ultimate reason for weird-butt lag issues that just seem to crop up for no readily apparent reason.
Are there any other uber-nerds out there who have done similar testing and seen similar results?
I'll be glad to post an excerpt from my network monitor captures if anyone is interested.