The guys over at Fudzilla have been talking to the people over at Lucidlogix.
LucidLogix, if you did not already know, is the maker of the very highly anticipated Hydra Chip. They asked questions about ELSA Japan's recent usage of the Hydra Chip and were told that ELSA is not using it for graphic performance but as more of a PCI-Control Chip for GPGPU functions.
In Graphics Performance Mode the Hydra changes the traditional way that multiple GPUs deal with rendering. In traditional SLI or CrossfireX setups the GPUs share the rendering burden by rendering alternate items; this is usually broken up in to frames, (AFR) Tiles, and in older SLI cases lines. With Hydra the Hydra Chip controls the rendering, it is capable of sending commands to the GPUs telling them which objects to render in a give frame and then it assembles the distributed objects and send that data to the monitor.
This is a significantly more efficient method. The problem is that no one has seen a truly working demo of this. Perhaps by CeBit we will know more.
Read more here at Fudzilla.
Another interesting aspect is that its possible to cascade multiple Hydra chips, although at the moment this is limited to two chips, but this would allow for more cards to be used. This could at least in theory allow for four GPU graphics cards, but Nir suggested that in the future up to 8 GPU's per Hydra chip could be possible, in theory at least.
For those that are worried that this will be another nForce 200 chip, there's no need to worry as Nir assured us that the Hydra has a TDP of a mere 4W and as such it shouldn't even need a heatsink, although we expect motherboard makers to add one just in case.
The reason why Hydra scales so well is because its the Hydra chip that controls the rendering and Lucidlogix is doing it quite different than the way AMD and Nvidia are doing things. Using multiple graphics cards have so far meant alternate frame rendering, split screen rendering or tile based rendering. Lucidlogix on the other hand is going for object based rendering which is an entirely new approach which means that each GPU get allocated parts of an object of the scene that's being rendered and then the Hydra chip pieces together the small parts into the final frame which is then displayed on the screen.