Is the 3b2 dead?

[Kevin Darcy]

In article <1990Jun5.002131.11389 at maverick.ksu.ksu.edu> terry at eesun1.eece.ksu.edu (Terry Hull) writes:
>sullivan at aqdata.uucp (Michael T. Sullivan) writes:
>
>>:From article <3532 at wb3ffv.ampr.org>, by smarc at wb3ffv.ampr.org (Marc Siegel):
>>> 
>>> I wonder how many other people out there work for companies that
>>> have LARGE investments in 3b2 hardware and software. We have several
>>> 3b2's that may not be supported very much longer. While nobody at
>>> AT&T will actually confirm this, it seems that the 3b2 is a dead
>>> product line.
>
>>Do you mean the whole 3B2 line or the 310/400 models?  I doubt the former
>>and I believe the latter has already been announced.  The low-end 3B2's
>>are slower than 386's so why not phase them out.  However, I have heard
>>nothing to suggest that the rest of the entire 3B2 line is on its way out.
>
>It takes a fair sized 3b2/1000 to be faster than a good '386 box.  
>For the money you spend on the 3B, you can buy several '386s.  In defense
>of the 3Bs though, they are very reliable.   
>
>--
>
>Terry Hull 
>Department of Electrical and Computer Engineering, Kansas State University
>Work:  terry at eece.ksu.edu, rutgers!ksuvax1!eecea!terry
>Play:  terry at tah386.manhattan.ks.us, rutgers!ksuvax1!eecea!tah386!terry

If the demise of the 3B2 is near, I don't think it will be because of pure 
hardware obsolescence so much as a marketing move: the price/performance niche 
can be filled by 386's at the low end (as Terry points out), and true* 
multiprocessing machines a little higher up. And even in its own shrinking 
class, 3B2's face stiff ongoing competition from NCR, Motorola et al.

There is a software concern, as well. AT&T lacks a decent OA product for the 
3B2, and offers only "generic" office packages (e.g. WP, Spreadsheet). Whether
this is the chicken or the egg, is anyone's guess, however. But these are the
bread and butter of a "departmental/branch"-class computer, and it doesn't
help the 3B2's plight in the marketplace.

I can't really comment on the reliability issue: I get to deal with all the 
ugly crashes on our extended network of 3B2's (>130), so my viewpoint is rather 
slanted.

* I hesitate to call the multi-processing feature of the 3B2/1000 "true" - it
  is so asymmetric as to render it useless for almost all real applications.

Speaking only for myself, and not for Chrysler or Chrysler Financial...
------------------------------------------------------------------------------
kevin at cfctech.cfc.com 		    | Kevin Darcy, Unix Systems Administrator
...[mailrus!]sharkey!cfctech!kevin  | Technical Services (CFC)
Voice: (313) 948-4863 		    | Chrysler Corporation
Fax:   (313) 948-4975 		    | 27777 Franklin, Southfield, MI 48034
------------------------------------------------------------------------------