Nvidia really is chopping and changing its chipsets these days, and the nForce 200 or n200 chip is an interesting little beast, as it is a PCIe 2.0 controller which adds 32 lanes of PCIe 2.0 bandwith, but only communicates with the chipset using 16 lanes of PCIe 1.0 bandwidth.
When you consider that PCIe 2.0 doubles the available bandwidth of PCIe 1.0 per lane, then you have to wonder how Nvidia has managed to pull this one off. From what we have found, their only explanation is that there's enough bandwith there for the n200 to communicate with the chipset, be it the 780i or 780a, without any bottle necks. The reason for this is partly because of the SLI bridge connections and partly because the chipset can handle any overspill SLI data that doesn't use the bridge connection.
One has to question the implementation, though, as the third slot on the 780i chipset is connected to the MCP and not the SPP, and relies on PCIe 1.0. This makes us wonder how much faster the 790i will be in 3-Way SLI, as it should have three native PCIe 2.0 slots. As with the 780a, you'll be limited to either two x16 slots or a single x16 and two x8 slots. In other words, this chipset isn't a top of the line product, according to Nvidia's past classification system which required full x16 bandwidth for this kind of product.
And for those intending to use a 780i or 780a motherboard for some other types of x16 devices than Nvidia graphics cards, it's very possible that there will be some bandwidth issues due to the chipset designs. Corner cutting is never the best way of doing things and it really looks as if the n200 helps Nvidia cut some corners when it comes to product development and R&D expense, but it doesn't give their potential customers what they've come to expect from the company.
Published in Motherboards
The nForce 200 interconnection is a joke
It's all in the bandwidth