Steve Youngs <sryoungs(a)bigpond.net.au> writes:
|--==> "SJT" == Stephen J Turnbull
SJT> First, did you reboot, then run cold and hot timings?
No. I suppose a fairer test would have been:
xemacs -kill && time xemacs -kill
When benchmarking, I tend to use `xemacs -batch -f kill-emacs'. The
`-batch' is to avoid measuring the time it takes to map an X window.
However, you might *want* to take that into account, so `-batch' might
not be for you. (But even so, a -batch run will show you whether the
difference comes from the X part or not.)
Maybe XEmacs is waiting for the X server a lot? The `:bold t' thing
sounds like a bunch of calls to `font.el', which might not be
optimized for speed.