Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!convex!poorman
From: poorman@convex.com (Peter W. Poorman)
Newsgroups: comp.sys.cdc
Subject: Re: The Design of the CDC 6600
Message-ID: <poorman.676989854@convex.convex.com>
Date: 15 Jun 91 12:44:14 GMT
References: <25400@well.sf.ca.us> <1991Jun13.175819.573@ultra.com> <204102.21677@timbuk.cray.com> <1991Jun15.012141.1107@ultra.com>
Sender: usenet@convex.com (news access account)
Distribution: comp
Organization: CONVEX Computer Corporation, Richardson, Tx., USA
Lines: 20
Nntp-Posting-Host: convex1.convex.com

In <1991Jun15.012141.1107@ultra.com> shj@ultra.com (Steve Jay) writes:

>No way to duplicate it exactly.  The characters were not bit-maps.  The beam
>actually traced the curves for each character.  You got true curves, and
>each console would be a little bit different.  

What made it particularly interesting was that the beam was slow in making
large position changes.  Unless you inserted small delays into your code the
characters (or pixels -- the tube supported 512x512 dot graphics) would
end up somewhere other than where you intended.

Made for some interesting code to minimize beam movememt in a 4K byte, one
register machine.

(Maybe this should be comp.sys.nostalgia.)

--Pete Poorman
  Convex Computer Corporation
  Houston, Texas
  poorman@convex.com