Hallo John, On Wed, 3 May 2006 06:40:46 +1000GMT (2-5-2006, 22:40 +0100, where I live), you wrote:
>> Also if one of your filters does not have continue processing >> checked it will not proceed to the next filter. JP> Should all filters have this option selected? Definitely not. Among all (100+)of my filters there's no more than one or two that have this option enabled. A message will be compared against the first filter, if it matches it'll be processed and it won't be compared against the next filter (unless the first filter has 'continue processing...' enabled) If it doesn't match, it'll be compared to the next filter and on and on and on, until it finds a match (without 'continue processing...' enabled) or until it runs out of filters. Basically this means that, as filtering takes time and processor power, you should put your busiest filters on top, because that means that you won't waste processor time on comparing lots of messages against filters that won't be triggered anyway. Should you enable 'continue processing....' for all filters you're losing whatever you gained with an efficient filter order. When you look into my sorting office, the first filter you'll see is my the one that acts upon the spam header inserted by my mail server, next you see some filters regarding some heavy duty mailing lists. The first eight filters process more than 95% of my incoming mail. I don't gain anything by comparing all that mail against my other 100+ filters. What complicates this story are subfilters. When a filter is triggered by a message, the subfilters will be compared against that message, when the filter doesn't get triggered, the subfilters will be left alone. -- Groetjes, Roelof Uh, yeah...I MEANT to do that! The Bat! 3.80.03 Windows XP 5.1 Build 2600 Service Pack 2 1 pop3 account, server on LAN OTFE enabled P4 3GHz 2 GB RAM
pgpk4a93meLEW.pgp
Description: PGP signature
________________________________________________ Current version is 3.71.03 | 'Using TBUDL' information: http://www.silverstones.com/thebat/TBUDLInfo.html
