Dossy Shiobara wrote:
The danger here is that you're not only testing the performance of
AOLserver's [ns_share], but you're also introducing Tcl's file encoding
overhead as you're doing [gets] within the test.
A lot of Tcl's encoding overhead can be nullified simply with:
fconfigure $fid -translation binary
which implies -encoding binary as well. That is, of course,
if that would be correct in your program.
If the test were organized such that the file's contents were read into
memory separately, and the timing loop only covered accesses to the
[ns_share]'d variable, it'd be a useful test. Apparently, the encoding
support added to Tcl 8.x wasn't without measurable cost. (Jeff Hobbs
can correct me on this statement if I'm wrong.)
Yes and no. While full utf-8/unicode support did certainly
slow down Tcl when introduced, various optimizations have
been done over time to improve the overall speed of the core.
At this point, 8.4 is actually faster than 8.0, with improved
thread-safety and full unicode support. The best place to
see benchmark data is:
http://wiki.tcl.tk/1611
--
Jeff Hobbs, The Tcl Guy
http://www.ActiveState.com/, a division of Sophos
--
AOLserver - http://www.aolserver.com/
To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]>
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject:
field of your email blank.
