Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site caip.RUTGERS.EDU Path: utzoo!watmath!clyde!bonnie!akgua!gatech!ut-sally!seismo!caip!neuro1!sob From: sob@neuro1.UUCP Newsgroups: net.works Subject: Re: Masscomp Message-ID: <598@caip.RUTGERS.EDU> Date: Mon, 2-Dec-85 19:16:34 EST Article-I.D.: caip.598 Posted: Mon Dec 2 19:16:34 1985 Date-Received: Thu, 5-Dec-85 05:01:38 EST Sender: daemon@caip.RUTGERS.EDU Organization: Rutgers Univ., New Brunswick, N.J. Lines: 26 From: Stan Barber Steve Wall's recent comments about the masscomp do little to explain why his opinion is so negative. His remarks about RTU 3.0 probably are about the beta-test version since the actual released product is only just now getting into the hands of the users. I will have to agree that the masscomp software does have its problems, but much the same can be said of many companies' software (a good example is the Sun and Pyramid f77 compilers). It all depends on the market the machine is aimed at. In the case of the masscomp, the market was scientific, real-time computing originally. The Masscomp is the only machine I know of that was designed to run UNIX in a REAL-TIME environment (Please tell me if I am wrong). It has one of the highest performance/cost ratios and one of the best FORTRAN compilers available in its price range. There are problems as well. It may not run GNU-emacs. It does run Gosling-emacs in the BSD-4.1 style. I have not tried to port GNU as yet. I have tried a number of other 4.2 items, and if you plug in BSD 2.9 or 4.1 networking for those programs that use it, you make very few modifications. Almost all the System V stuff I have ported required no modification at all. As the case with many vendors, some customers are happy and some aren't. We are happy with Masscomp.