James K. Lowden wrote:

> I should mention my test was on NetBSD 1.4.2, definitely from the
> pre-vfork era.  A 1.6 Sparcstation yields very different results (with
> apologies to Matthias):
> 
> $ time sh -c 'for i in $(jot 10000); do /bin/echo -n; done' 
> real    1m23.455s
> user    0m2.387s
> sys     0m19.027s
> 
> $ time ksh -c 'for i in $(jot 10000); do /bin/echo -n; done'
> real    2m0.303s
> user    0m6.465s
> sys     0m34.900s

> Well, I hope I'm not beating a dead horse.  Like David, I was surprised to
> find so much difference in what I would have thought is a trivial, well
> understood process.  

Just for the records, here are the results on 1.5.1/vax (VS3100/30):

$ time sh -c 'for i in $(jot 10000); do /bin/echo -n; done'
   750.57s real    28.81s user   446.79s system
$ time ksh -c 'for i in $(jot 10000); do /bin/echo -n; done'
  1020.66s real    66.55s user   715.14s system

I agree that on slow systems (like this one) there is indeed a 
significant difference.  Although it is my understanding that few 
scripts are as bare-bones as the above, executing just a tight loop with 
a rather no-op echo in it; most scripts are doing much more stuff and 
hence a script which does 10000 exec/forks with serious workloads on 
real data probably will run for hours anyways [which makes the extra 5 
minutes that ksh takes vs. sh appear in a very relative light again -- 
if a script is running for 5 hours, I don't really care if it takes 5 
minutes longer.]  Still it shows that ksh could probably be optimized 
somewhat.  Whether it's really worth it, I somehow doubt.

-- 
Matthias Buelow