While a single smart processor like the Glint 300SX can speed up 3-D rendering substantially, there's no substitute for lots of buffer space. Like other Intergraph GLZ adapters, the
24-MB GLZ2
employs a 220-bit-wide memory bus to service 92 video planes consisting of two 24-bit RGB buffers (double buffering for smooth animation) and one 24-bit Z-buffer that caches depth information. Masking, overlay, and image window-control bits account for the remaining 20 video planes.
The GLZ2 uses four custom proprietary Intergraph ASIC subsystems for 2-D and 3-D graphics acceleration. The DMA Engine is the main graphics acceleration processor; according to Intergraph, it touts 3-D speeds of up to 450,000 Gouraud-sh
aded triangles per second. The PCI/DMA ASIC controls vertex data flow (the vertices of surface polygons) up to burst speeds of 4 MBps to and from the PCI bus and the GLZ2's 24 MB of VRAM to the FIFO chip subsystem. The four-ASIC Resolver subsystem controls RGBA (RGB and Alpha channel) pixel and Z-data I/O to the frame buffer. A 256-bit-wide Analog Devices ADV7160 DAC handles color conversion.
Omnicomp's 3Demon SX48 provides substantial acceleration with OpenGL, but the TDZ-40's GLZ2 subsystem is faster still. The SX48 keeps up only in its 12-bit color mode, which isn't a fair comparison.
Flexible C++
Matthew Wilson
My approach to software engineering is far more pragmatic than it
is
theoretical--and no language better exemplifies this than C++.
BYTE Digest editors every month analyze and evaluate the best articles from Information Week, EE Times, Dr. Dobb's Journal, Network Computing, Sys Admin,
and dozens of other CMP publications—bringing
you critical news and information about wireless communication,
computer security, software development, embedded systems,
and more!