Remember IDF Fall 08. THE top of the line IC7 has a QP path between the GPU and the CPU on the 1366 M/B . Now read about this company Neoptica and what that path between the GPU and CPU does. As for IC7 gaming powers . Until we see it on the platiform its designed for its all guesses. But so far brother in law says everthing is going as plained or better. Thats all I can say about Hi end vs. low.
number of developers who are able to creatively explore the capabilities of hardware systems, which has
historically been the key driver of advancing the state of the art in interactive graphics.
The New Era Of Programmable Graphics
Neoptica has built a new system that moves beyond current GPU-only graphics APIs like OpenGL and
Direct3D and presents a new programming model designed for programmable graphics. The system enables
graphics programmers to build their own heterogeneous rendering pipelines and algorithms, making efficient
use of all CPU and GPU computational resources for interactive rendering.
With Neoptica's technology and a mixture of heterogeneous processing styles available for graphics,
software developers have the opportunity to reinvent interactive graphics. Many rendering algorithms are
currently intractable for interactive rendering with GPUs alone because they require sophisticated per-frame
analysis and dynamic data structures. The advent of programmable graphics makes many of these
approaches possible in real-time. New opportunities from the era of programmable graphics include:
? Feedback loops between GPU and CPU cores: with the ability to perform many round-trip per-pixel
communications per frame, users can implement per-frame, global scene analyses that guide adaptive
geometry, shading, and lighting calculations to substantially reduce unnecessary GPU computation.
? Complex user-defined data structures that are built and used during rendering: these data structures
enable demand-driven adaptive algorithms that deliver higher-quality images more efficiently than today?s
brute-force, one-way graphics pipeline.
? Custom, heterogeneous rendering pipelines that span all processor resources: for example, neither a pure
ray-tracing approach nor a pure rasterization approach is the most efficient way to render complex visual
effects like shadows, reflections, and global lighting effects; heterogeneous systems and programmable
graphics will make it possible to easily select the most appropriate algorithm for various parts of the
graphics rendering computation.
During the past year, Neoptica has built a suite of high-level programming tools that enable programmable
graphics by making it easy for developers to write applications that perform sophisticated graphics
computation across multiple CPUs and GPUs, while insulating them from the difficult problems of parallel
programming. The system:
? uses a C-derived language for coordinating rendering tasks while using languages such as Cg and HLSL
for GPU programming and C and C++ for CPU programming, integrating seamlessly with existing
development practices and environments and providing for easy adoption;
? treats all processors in the system as first-class participants in graphics computation and enables users to
easily share data structures between processors;
? presents a deadlock-free, composable parallel programming abstraction that embraces both data-parallel
and task-parallel workloads;
Neoptica
Programmable Graphics?The Future of Interactive Rendering
Neoptica's Vision
Neoptica published a technical whitepaper back in March 2007. Matt Pharr also gave a presentation at Graphics Hardware 2006 that highlights some similar points.
It explained their perspective on the limitations of current programmable shading and their vision of the future, which they name "programmable graphics". Much of their point resides on the value of 'irregular algorithms' and the GPU's inability to construct complex data structures on its own.
They argue that a faster link to the CPU is thus a key requirement, with efficient parallelism and collaboration between the two. Only the PS3 allows this today.
They further claim the capability to deliver many round-trips between the CPU and the GPU every frame could make new algorithms possible and improve efficiency. They plead for the demise of the unidirectional rendering pipeline.
If ya take the time you will find that a couple of Neoptica's Employees are now working for project offset . So Intels API is going to be really cool. So is this game . Its an intel only game . As no other system will beable to play the game. I know more but its not worth the flames.
So now you know why that IC7 3.2ghz. has the fast QP lane . People say its useless to run it at that speed. People would rather say thats useless not knowing what the hell their talking about . But ATI put a side port on there cards for a reason. ATI just needs a GPU strong enough to get a benefit from it. If your really interested Its said to speed some things up 10-50X. I just reporting what I have read and been told.
These systems "deliver imagery that is impossible using the traditional hardware rendering pipeline, and deliver 10x to 50x speedups of existing GPU-only approaches."
Of course, the claimed speed-up is likely for algorithms that just don't fit the GPU's architecture, so it's more accurate to just say a traditional renderer just couldn't work like that rather than claim huge performance benefits. So in a little over a year we will have a new game and system. That uses all resources of the Nehalem cpu. Thats all 8 threads . plus Larrabee. Ya this is going to be interesting .
Most info here is from the white paper!
Facebook
Twitter