> > Another effectively useless tunefs parameter is fs_rotdelay - an
> > estimate of the number of milliseconds it takes the disk driver to
> > handle an interrupt. The idea was to optimise the physical placement
> > of the logical blocks of a file. Hopefully the N+1th logical block of
> > a file would appear under the disk heads just after the driver had
> > processed the interrupt for block N. [Unix processes tend to read
> > files sequentially.] Now this was a good thing in the days of 1 MIP
> > CPUs and stupid but simple SMD controllers. Today it's pointless:
> 
> I believe that it does serve a limted but useful purpose in NetBSD.
> NetBSD DOES support those super old, super slow devices, and people
> ARE still running them. If this were an OS geared toward highend Alphas,
> I'd quickly say, "Take it out!" But my guess is that my old hp300 systems
> with the RLL drives benifit from the optimizations.
> 
> Are there any modern optimizations that can be done in software? Maybe
> the file system code could be modified to fit both old and new.
> 
>         Nathan
> 

I totally agree with Nathan if it were not for the supporting of "1 MIPS
CPU's and stupid but simple SMD controllers"  My lab of Sun 3's would be
running SunOS 4.1.1 and not be very happy.  These machines are great
learning platforms most people will be willing to give you the complete
machine if you hall it away or for next to nothing.  I know lots of places
that are still useing older machines that really benifit from the ports to
the old unsupported machines.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Robert Harden                             Phone: 1-888-744-3600 ext 656
Technical Support, SysAdmin               Fax: 1-304-744-8049
NewWave Communications.                   Email: rharden@squid.newwave.net
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=