Lucid Hydra 100 multi-GPU scaling demonstrated at IDF

Sunday 24th August 2008, 01:01:00 PM, written by Rys

Lucid have demonstrated working first silicon at IDF this year.  For those unfamiliar with their technology, it's a combination of hardware and software that intercepts graphics calls to the API runtime, before performance workload division to affect performance scaling.

It's that intercept and workload division that sets the multi-GPU scheme apart from what you get already with Crossfire and SLI.  In those existing schemes, the driver is entirely responsible for dividing up the work to be done.  Lucid's approach gets some custom Tensilica-based hardware involved, to help accelerate things and more efficiently drive the GPUs underneath.

Efficiency scaling is achieved at the cost of latency, since there's yet another layer of hardware and software in the way before anything gets drawn on your screen.

The Tech Report have a nice writeup of the demo and some of the technical details, as they attend IDF this year, so if you're curious about how it works it's recommended reading.

TR mention that Lucid are aggressively patenting the key technologys, so some of the finer details are likely to be found in their patent filings.

Discuss on the forums

Tagging

b3d ± tech, report, lucid, latency, graphics, scaling, efficiency, sli, crossfire

Related b3d News

Book reviews: The Magic of Computer Graphics and 3D Engine Design for Virtual Globes
A new approach to graphics performance analysis
RWT Analyzes Bulldozer Benchmarks
Really Techy Worldly Hot Par 2010 coverage
Carmack Rage: either console could look superior
Sub $100 graphics at Tech Report
Broadcom purchase AMD's DTV business
The Khronos Group announce Heterogeneous Computing Initiative
3DMark, the Game
PC's Assassin Creed Official System Requirements: 2GB of RAM