Browse Prior Art Database

Terminal for Real-Time Viewing of Multidimensional Data

IP.com Disclosure Number: IPCOM000101395D
Original Publication Date: 1990-Aug-01
Included in the Prior Art Database: 2005-Mar-16
Document File: 3 page(s) / 99K

Publishing Venue

IBM

Related People

Todd, SJ: AUTHOR

Abstract

A method is disclosed for computer graphics multidimensional visualization that permits display of many dimensions with real-time manipulation of viewing parameters. Colour and other attributes are treated homogeneously to the space coordinates x, y, z. Implementation is in terminal hardware or microcode. Display list points are vectors of more than three values, display list transforms are wider than the standard four by four and finally colour and other attribute registers are updated with the result of applying display list transforms to display list points.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Terminal for Real-Time Viewing of Multidimensional Data

       A method is disclosed for computer graphics
multidimensional visualization that permits display of many
dimensions with real-time manipulation of viewing parameters.  Colour
and other attributes are treated homogeneously to the space
coordinates x, y, z.  Implementation is in terminal hardware or
microcode.  Display list points are vectors of more than three
values, display list transforms are wider than the standard four by
four and finally colour and other attribute registers are updated
with the result of applying display list transforms to display list
points.

      The method assumes standard high function hardware such as the
5080 and 6090 for embodiment.

      The disclosure proposes that all primitives saved in the
terminal display list (e.g., point, polyline, filled polygon) use
n-dimensional vectors.  A register defining n is supplied by the
hardware (see figure) and is set by the application;  n may be set to
any value, with possibly an upper limit defined by the system.  The
default value for n will be 3, which will give a system equivalent to
a 6090.

      The display transform matrix will be of size (n + 1) * m, set
by software.  A register for m supplied by the hardware will be set
by the software.  Default for m will be 4, equivalent to a standard
terminal.  This will be applied in the normal way to the input
vector, increased to dimension (n + 1) by addition of a 1 in the (n
+ 1)'th position to give an m dimensional output.

      The m values in the output will be mapped onto the registers x,
z, y, w, colour, linetype, and so on, in one of various ways
depending on the hardware and microcode details, as in the methods
below.

      Method 1 The i'th output is mapped into the i'th register.  The
registers are ordered according to how likely it is thought that the
attribute it represents is to be dynamically mapped.  If the
application requires dynamic mapping of the 5'th register but not the
fourth, then the transformation matrix must be set up so that the
fourth output is constant.

      Method 2 A vector register is supplied in the hardware that
tells into which register each output is to be mapped.  This register
is set by the application.  If the same value is required in two
registers, e.g., z and colour, where depth cuing is used, then this
value must be computed twice.

   ...