Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!uunet!seismo!husc6!rutgers!ames!amdahl!howard From: howard@amdahl.amdahl.com (The Toolmaster) Newsgroups: comp.sys.mac Subject: Re: Drawing in Technical Environment Message-ID: <11292@amdahl.amdahl.com> Date: Sun, 2-Aug-87 05:33:23 EDT Article-I.D.: amdahl.11292 Posted: Sun Aug 2 05:33:23 1987 Date-Received: Sun, 2-Aug-87 20:20:44 EDT References: <3538@sdcsvax.UCSD.EDU> <5277@prls.UUCP> <11158@amdahl.amdahl.com> <1828@batcomputer.tn.cornell.edu> <5305@prls.UUCP> Reply-To: howard@amdahl.UUCP (The Toolmaster) Organization: Amdahl Corp, Sunnyvale CA Lines: 60 In article <5305@prls.UUCP> gardner@prls.UUCP (Robert Gardner) writes: >Other than SYSTAT we have been unable to find anything that will plot >3-dimensional data (hidden line removal, rotations, contour plots, etc.). >This lack almost scuttled our proposal to purchase a reasonably large no. >of Macs for technical use. Anyone seen anything? > >Another product that would be useful would be a package of GKS-style >graphics subroutines (similar to DISSPLA) for use in MPW. Of course, a >lot of this is supplied by the Mac Toolbox, but support for plotting >data in 2- and 3-dimensions would be useful. Have you thought of leaving the bulk of the work on the host and using the Macs for graphic terminal emulation to produce the required plots. A terminal emulator like Versaterm PRO allows Tek4014 and 4105 emulation with ability to zoom and capture diagrams as MacDraw type pictures. This allows a great deal of flexibility and finding something host based that produces Tek style output is probably pretty easy. Of course, if you don't have a "host", forget it. >Although the technical market gets ignored a lot by Mac publishers/ >developers, with Apple trying to push the Mac II into the technical >market, I hope things start to change. (I wonder how the Mac II can >face the onslaught of DEC, Sun, and Apollo, though. Their prices are >now so low for volume purchasers that they were almost price-competitive >to our purchase of Mac SE's!! Software was lacking, though. Anyway, >that's another story for another message...) Buyer beware, personal views follow: The way I see it so far; Sun/Apollo/DEC? workstations are nice machines, but the software is the killer. Those companies appear to be heavily discounting now that they don't fully run the market and the low end machines are edging into the picture. But the so called "low end" machines still enjoy low end pricing of their software. Since a software budget can easily eclipse a hardware budget, for any machine, it still doesn't make economic sense to go with the so called "high end." After all, the software doesn't really matter. The hardware doesn't really matter. Getting the job done is what matters. Now, I'm probably about to start a war of epic proportions so let me just remind everyone that this is comp.sys.mac and if you have comments, mail them. Now, people will say, what do you about expensive software. They'll say that they *give* you UN*X on every machine. I look at that and say that's the problem. UN*X belongs in the backend, not on the workstation. The workstation is a user interface engine and shouldn't be bogged down with running UN*X. All it needs is great graphics and support routines in ROM and a streamlined kernel (preferably with multitasking) to manage user interface resources. Sounds a lot like a Mac (as soon as they work out the multitasking). So, what do I see as today's big challenge? Not to turn everyones workstation into a general purpose computer. Not to share the job among many workstations acting as general purpose computers. The challenge is to share the job among many dissimiliar machines, allowing each machine to do what it does best. And right now, it looks like the Mac is best suited to meet the challenge. -- "The value of knowledge lies not in its accumulation, Howard C. Simonson but in its utilization." E. Green amdahl!howard [ The disclaimer for this message may be found in a previous article ]