[comp.sys.atari.st] EST/TT <-> workstation flame

pjd@cwruecmp.UUCP (dr. funk) (07/20/87)

{And while I'm on my soapbox...}

Here's hoping that EST stands for something more than Eastern
Standard Time. And that TT is a reality.

The current movement in the workstation market is really exciting.
Sun and DEC are shooting it out on price and Apollo is finally
going "standard."

Given that I can buy a diskless Sun for $3500-4000 or a node with
70Mb disk for $7000-8000, it won't be long before PC's and workstations
converge. (Some may say that it has happened already!)

Time to put on your skates, boys! (Atari? Atari?) Sun is already
playing your game (low mfg cost) with the one board 3/50. The $2,000
1 mip, 1 mpixel, 2 mbyte machine is gonna be the next battleground
in six to twelve months. With centralized file resources and a fast
network, I can still keep the "per seat" cost low.

After using GEM, Micro C-Shell, Personal Pascal and XPRO Prolog for
the last six months to develop a sizable CAD program, I don't think
the existing environment is up to a challenge from Sun on the high end.
There are too many "unexplainable phenomena" due to the ragged edges
of the GEM/TOS/etc. environment. The lack of multi-processing and
good IPC really hampers the development of certain kinds of systems,
especially those that try to mix programming paradigms (e.g., Pascal
with Prolog, etc.) [I DO know about MT C-Shell; flames to /dev/null.]

And now the message. Before long, I will be deciding where to put
my equipment money next and I will have to take the long term view
(e.g., distributed workstations.) Very shortly, low cost manufacturing
costs and decreasing Si costs will close the hardware gap between
PC's and entry-level workstations. The next purchase decision will
hinge on development and work environment. Please don't be caught
behind the "functionality curve."





-- 

paul drongowski				cbatt!cwruecmp!pjd
case western reserve university		pjd@CWRU.EDU