Fifth-Generation Workstation Graphics Architecture
The NVIDIA Quadro FX architecture takes application performance to
new levels by featuring three parallel vertex engines, a radically
new line engine, the industrys first on-chip vertex cache, and
eight fully programmable pixel pipelines coupled to a high-speed
DDR2 graphics DRAM bus. Graphics pipeline efficiency is magnified by
NVIDIAs next-generation crossbar memory architecture, enabling
occlusion-culling, lossless depth Z-buffer and color compression.
These elements combine to achieve unprecedented 3D primitive
performance: 80 million lit and textured triangles per second, five
times faster line performance than the NVIDIA Quadro4 graphics
solutions, and massive fill rate powered by eight pixel pipelines.
But the true measure of power is application performance¡Xand the
NVIDIA Quadro FX architecture delivers more than double the
performance vs. the previous generation.
In addition, all NVIDIA products utilize the NVIDIA Unified
Driver Architecture (UDA), which is continually optimized for
performance, quality, and support throughout the life of every
NVIDIA Quadro product. This continual improvement has delivered
performance enhancements of up to 100% within nine months after
product launch.
Third-Generation Vertex and Pixel Programmability
The design cycle is a long, iterative process from concept, to
modeling, and final production. This final production can require
hours of offline CPU rendering. The programmability of the NVIDIA
Quadro® FX architecture empowers the industrys leading OpenGL® and
Direct® workstation applications to now make the production
rendering process an integral part of real-time design. This reduces
design cycles, increases productivity, and accelerates
time-to-market.
Leading this change in workstation application functionality are
the major CAD and DCC application vendors, including: SolidWorks,
Alias|Wavefront, Discreet, Softimage, and more. End users can take
full advantage of the programmable NVIDIA Quadro FX architecture by
enabling sophisticated real-time shaders to simulate a virtually
unlimited range of physical characteristics, such as lighting
effects (fresnel effects, chromatic dispersion, reflection,
refraction, BRDF models*, etc.) and even physical surface properties
(such as casting effects, porosity, molded surfaces, etc.).
For example, car models can be modified on-the-fly to change
their clear coat and metallic reflective paint properties; the sheen
of a characters skin can be adjusted through a dynamic range from
oily to dry by interactively adjusting a few sliders; or objects can
be programmed to dynamically self shadow. Real-time shaders allow
these effects to be combined and modified interactively-impossible
with simple 2D static texture maps.
Full
128-bit Precision Graphics Pipeline
Sophisticated real-time effects typically involve multiple
mathematical operations that demand high precision to maintain image
quality. The NVIDIA Quadro® FX features 128-bit IEEE floating-point
precision, making available millions of colors in a broad dynamic
range. This results in the highest level of accuracy and the
ultimate in visual quality.
High subpixel precision is another major contributor to image
quality, addressing visual anomalies that cause models to "sparkle"
or "crack." The NVIDIA Quadro FX virtually eliminates this problem
by providing 12 bits of sub-pixel precision¡Xfour times higher
precision than the nearest competitive product.
Precision continues to be a critical factor when rendering high
quality antialiased images-for both line and full-scene antialiasing
(FSAA). The NVIDIA Quadro FX architecture accelerates antialiased
points and lines in hardware, and supports up to 16X FSAA. And
unlike other competitive hardware, NVIDIA Quadro FX products drive
FSAA up to a phenomenal 3840x2400 resolution.
¡@