Hello Roy, 

Definitely, the machine with filters is going to be slower (provided
all else is equal), and it gets slower the more filters you have. It's
not a reflection on TB, but on the fact that each message must be
matched against every active filter.

You'll take a speed hit on the complexity of your filters as well. If
you have messages being parsed by some external program, or printing
hardcopys, or whatever, it's gonna be slower than just a straight move
message to folder filter.

Kill filters are slow because of the way that they are designed to
work. Again, it's not a reflection on TB, but of the inherent dual
session requirements. TB must first download all message headers, then
match those against kill filters. Those that are a positive match will
be skipped and then deleted from the server while the rest of the
messages are being downloaded. This happens in the second session TB
has with the mail server.

I would venture to say that even a single kill filter string would
cause the overall filtering system to be slower than twenty move SPAM
to trash filters. I've never tested this, but it seems reasonable.
Then again, someone with a fat data pipe might be able to manage the
D/L of the headers, kill filter, and then D/L the non-killed messages
faster than twenty move to trash filters. Could be wrong, but that's
what theories are for! <grin>

As you said below, it is true that a filter will perform faster if it
has less text to search. i.e. knowing that the string you want to
filter on appears only in the SUBJECT line, and setting your filter up
to search only the SUBJECT line.

If you know where the string is supposed to be, then search only that
area because if you are searching in TEXT OR BODY OR SUBJECT, it's a
good deal of wasted effort by TB.


On Wed, 22 Mar 2000 at 17:22:21 [GMT +0000], you wrote:
RM> I'm wondering what is the impact in terms of speed of various
RM> filters and Leifs piece had me wondering. I've noticed large
RM> differences in speed of download between almost identical machines
RM> where one was filtering and the other wasn't. I've tried deleting
RM> filters that search on all text as I assumed they were the slowest
RM> - but are kill filters that are killing 'on the server' the
RM> slowest?

RM> Cheers
RM> Roy





Leif Gregory 

-- 
TBUDL/TBBETA List Moderator
ICQ 216395 <[EMAIL PROTECTED]>
Web Site   <http://www.pcwize.com>
TBUDL FAQ  <http://www.pcwize.com/thebat/faq.shtml>

PGP Key ID: 
  0x8604279A (DH/DSS)
Fingerprint: 
  9E16 4316 FA42 5DC6 EB1D  D0ED D37A 858A 8604 279A


Using The Bat! 1.42 Beta/6 under Windows 98 4.10 Build 2222 A  
on a Pentium III 500 MHz notebook with 128MB.

Tagline of the day:
Just remember...if the world didn't suck, we'd all fall off.



-- 
--------------------------------------------------------------
View the TBUDL archive at http://tbudl.thebat.dutaint.com
To send a message to the list moderation team double click here:
   <mailto:[EMAIL PROTECTED]>
To Unsubscribe from TBUDL, double click here and send the message:
   <mailto:[EMAIL PROTECTED]>
--------------------------------------------------------------

You are subscribed as : archive@jab.org


Reply via email to