SCO Unix vs ICS vs Xenix

[Terry Hull]

In article <8909281749.AA03817 at decwrl.dec.com> paine at fungus.dec.com (Willy Paine) writes:
>
>I have contacted several people on private personal opinion on comparing
>between Xenix and SCO UNIX on performance and price.  Now I am trying to
>find out between SCO Unix and ICS 386/xi 2.0+.   I am told that SCO Unix
>is slower or more clumsy than Xenix and couple of SCO Unix owners are so
>UNempressing with SCO Unix performance.   

Performance in what area?  Disk speed? Floating point speed?  Serial I/O
handling?  

>Some said the kernel is too
>big and using too much memory that don't allow enough memory for
>X-sight.   

Both SCO UNIX and Xsight say that 4 MB is the MINIMUM recommended memory
configuration.  It has been my experience that the minimums SCO recommends
are the absolute minimums that you should even consider.  You should have
more.  I found that even running XENIX 4 MB was not really enough memory
if you are running large programs (like GNU Emacs) concurrently with Xsight.  

Did this person have 8 MB in his machine?  That would seem to be a good
starting point for heavy Xsight users running 3.2.  

I have the latest verison Xenix and ICS 386/ix and they are 
>different on size and performance but I am curious about SCO Unix vs ICS 
>386/ix.   One person is saying that ICS 386/ix is better product than
>new SCO Unix but I can not tell if he already test with 386/ix.
>I am still checking on upgrade cost.  

The run-time upgrade from XENIX 2.3 -> UNIX 3.2 cost me $325 as
Softcare supported customer.  I do not know if the price of the
development system upgrade has been set yet.  I do know that the
development system started shipping just recently, so it will probably
be a while before the DS upgrade is available.  

-- 
Terry Hull 
Department of Electrical and Computer Engineering, Kansas State University
Work:  terry at eecea.eece.ksu.edu, rutgers!ksuvax1!eecea!terry
Play:  terry at tah386.manhattan.ks.us, rutgers!ksuvax1!eecea!tah386!terry