Quote from: dust on April 10, 2011, 09:28:03 PM

Quote from: LiquidMiner on April 09, 2011, 06:03:50 PM

Quote from: mrjones on April 09, 2011, 05:50:08 PM

Quote from: LiquidMiner on April 09, 2011, 04:20:02 PM

The cluster will be based on 2xHD6990 per rig and this can be expanded to as many rigs as possible. Income of the rigs will be divided between buying more mining rigs and paying out dividend for each of your shares.

If you're really serious about a cluster, you should use a PCIe expansion chassis designed for GPUs, which can hold 16 cards rather than a bunch of separate computers. For example, the Dell PowerEdge C410x. It works with ATI/AMD cards as well as NVIDIA cards, and can plug right into any computer to add 16 PCIe slots. In a 42U rack you can easily fit three computers, nine of those, and a power distribution unit and you'll find yourself with 144 6990's--which is somewhere around 100 billion hashes per second. All for a cost of under 140 000 BTC.

That is one very good option, but it would be hard to raise 140,000BTC from investors. This can serve more as a long-term goal for the project, and would certainly be sustainable even when the fees will be the main way of financing nodes.

Unless I am mistaken, ATI drivers only support up to 8 GPUs on linux (and up to 4 on Windows), so thats 4 6990s max. I don't think a special expansion chassis would get around that limitation.

Even if the ATI drivers could, I'm betting the miners couldn't.

AFAIK you are completely right