by Anton Shilov
09/23/2009 | 01:12 PM
LucidLogix, a startup working on special-purpose load balancing hardware, at Intel Developer Forum 2009 announced yet another Hydra 200 chip. The developer promises that its new processing engine will offer better functionality than the previous-generation Hydra 100 chip (which has never made to the market) and will enable commercial applications.
“We’ve further refined our Hydra engine and made it faster and more flexible, allowing for a near limitless combination of GPUs. Hydra 200 allows the consumer to get more ‘bang for their GPU buck’ by extending the life of their current GPU investment, providing even faster graphics performance and later upgrading their system with whatever card they choose,” said Lucid vice president of research and development, David Belz.
The Hydra engine sits between the chipset and several GPUs and acts like a dispatch processor within the array of graphics processing units (GPU) to distribute tasks between the chips. The technology drives the GPUs, performing scalable rendering of a particular image or scene, and relies on “unique adaptive decomposition and acceleration algorithms to overcome bottlenecks”. The Hydra engine combines a PCI Express system-on-chip with exclusive software technologies that load-balances graphics processing tasks, delivering near-linear to above-linear performance with two, three or more graphics cards, according to the company’s promises. Lucid claims that even completely different GPUs – from ATI and Nvidia, for example – can work in the same tandem.
“The graphics-vendor-agnostic Lucid HYDRA SoC allows consumers who are tired of complicated graphics solutions to transition towards ultra-responsive, ultra-realistic 3D graphics performance. The Intel and HYDRA 200 solution levels the playing field for users and provides seamless compatibility and interoperability across components,” said Moshe Steiner, chief executive of Lucid.
LucidLogix Hydra 200 Versions
Hydra 200 Model
Number of Supported GPUs
2x16 or 1x16+2x8 or 4x8
It is not exactly clear what is the difference between Lucid’s Hydra 100 and the new Hydra 200. Officially, Lucid claims that the novelty consumes only 6W and is made using 65nm process technology. However, the main difference may be in considerably faster switching between graphics cards, better load-balancing performance and more flexible PCI Express configurations support.
To date, the only commercial product based on Lucid Hydra 200 is Big Bang mainboard from MicroStar International, which also features Intel P55 core-logic and supports Intel Core i5 and Core i7 processors in LGA1156 form-factor. The motherboard features three PCI Express x16 slots and supports up to three graphics cards.