[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

DRAFT Agenda for SLUG/SMBX Meeting of 15 December

    I believe maintenance on XL400 costs similar to maintenance on a
    comparable Sun (don't forget to include maintenance on the disk).

Being end-of-year budget time I have been doing extensive investigations
into the comparison between SUN and Symbolics maintenance.  It's really
hard to do an adequate comparison because it really depends on what
"comparable" means and its hard to deconvolve the effects of volume and
discounting.  Perhaps SUN list maintenacne prices are comparable to
Symbolics list prices.  But that observation, even if true (I'm
dubious), is irrelevant to larger sites.

For example, we have 170 SUNs and 27 Symbolics machines.  We actually
paid $80K to SUN last year for repairing parts (we identify bad boards,
disks, monitors, but do not repair them) (the $80K is exchange price of
parts after 12% discount) on those 170 machines and we paid $120K to
Symbolics a year (regular hardware maintenance) for only 27 machines.
This comparison is unfair because it fails to account of labor, etc. but
the disparity is so huge on a per machine basis that something clearly
is wrong.

The basic problem Symbolics faces is that at any site where SUN is in
such a majority, SUN maintenance can be structured so that it will only
cost a small fraction of list.  For such sites (like ours) there is
therefore massive financial incentive to switch to SUN.  Therefore,
Symbolics cannot pattern its maintenance on SUN's.  It must find a way
of pricing its maintenance such that the maintenance costs incurred by
the majority machine (SUN) is comparable to that of the minority machine
(Symbolics) at the same site.  Minority players simply have to adopt
different pricing strategies to survive.

Currently maintenance costs are our dominant impediment to buying more
Symbolics machines.

(It just may be that Symbolics machines fail more often than
"comparable" SUNs.  If so Symbolics is (and we are) in deep trouble.
Does anyone have any statistics on actual costs incurred on T+M for