> 	I came to the same conclusion and posted a tweaked rc.subr which
> 	avoids the run_rc_script overhead:
> 
....

> 	Has anyone on a slow vax had a chance to time it?

I got my slow VAXen finally fixed up over the weekend.

I now have a VS3100/M76, a VS3100/M38, an MVII with scsi and 4 HD's
loaded with 1.2, 1.3.2, 1.4.3, and 1.5 with stock generic and fully
stripped MVII kernels, plus a MVII 4.3 box, likewise stripped or generic.

I was trying to get lmbench to build, and it crapped out needing a
bunch of gnuisms that I will try to compile up tonight.  That thing
seems to be Pentium oriented.  Does it still run up reasonably well
on a VAX or is there some other benchmarking suite I should use?

Send me a direct copy of any scripts you want me to test on 1.5.1A
(Dec 29 sources):

    nbsdbob at weedcon1.cropsci.ncsu.edu

Thanks

Bob.... the idiot with the 0.9 vup VAXentoyz.....(:+{{.....

p.s. You won't believe the times to compile and the kernel footprint
     on a stripped 4.3BSD on the MVII compared to 1.5 stripped MVII!