The wiki erroneously states that "it is possible to have false decode problems caused by digital noise at the beginning of transmissions"
What really happens with a 396T on Fleetnet is that when a user keys up his digital radio but doesn't start talking right away, the voice channel he's been given access to goes into an idle/waiting mode until he speaks up. The 396T's digital decoder chip makes the determination that the digital bits coming over the idling voice channel are not P25 audio, so it assumes it's analog audio and passes the racket to the speaker, much to the annoyance of many 396T owners, who refer to it as "motorboating".
Once the user starts talking, the decoder chip detects valid P25 audio and stops passing the raw data to the speaker. Instead, you hear the P25 decoder's output (or nothing if the traffic is encrypted).
The P25 Waiting Time option is a lame attempt to fix this unfixable bug. (It's unfixable because nothing in the scanner is able to distinguish between true analog audio and an idling digital channel)
If you set the wait time to any value large enough to actually have an impact on the motorboating, you'll end up missing the start of many comms. To make matters worse, Uniden made it a scanner-wide option, so if you use it to clean up Fleetnet traffic, it'll chop the start off of the comms on all Motorola systems that you monitor, even all-analog ones that have no use for P25 Waiting Time.
Enjoy...