I know many people want to know the fine detail but that's knowledge that we don't want to give away as said previously.
Yohan, not stating anything is completely your preoperative, but I don't know why you're asking for my money then.
At the very least I would want to understand a few metrics in order to evaluate how this compares to what I think Avalon will bring.
a) Ghash/s per $ of capital cost
b) Ghash/s per Watt
c) Ghash/s per Total opex cost (server space, electricity cost, etc)
The technology question is irrelevant I could call it ASIC but that actually doesn't mean anything really. In true meaning a CPU is an ASIC so is actually a FPGA both raw and loaded. I could equally well use a long description which is more accurate but that would confuse all other than the very technically orientated in this forum.
The definition of CPU, FPGA and ASIC relates to how the computation is performed. Not the fact that they are silicon chips. Sorry, calling a CPU an ASIC is a pet peeve of mine, so please excuse the clarification.
- A CPU performs computation as a series of instructions, very inefficient
- A FPGA performs computation in logical LUTs, better than CPU but still inefficient
- An ASIC performs computation directly on transistors, most efficient
We have moved on technology wise from what we did on Cairnsmore1 and we have learned many things doing that project and from running our own mining rigs based on CM1 boards. Generally CM1 was our way to learn about Bitcoin and what works and what doesn't. CM1 was very deliberately limited in the technology that we deployed and that was to meet timescales and do what we promised.
Designing PCB boards that house FPGA chips supplied from Xilinx/Altera is a completely different skill set than designing and fabricating an ASIC where you have to be first time right or all investment is lost. bASIC and BFL are learning this. Just because Cairnsmore1 was a success, does not mean the next engineering project will be. Again it comes down to how can I evaluate if you are capable of building this project, if I don't know what the project is doing?
One last thing that confuses me. You say that we will be able to measure the effect in the network hash rate prior to funding. But that would mean the project is already complete and working, which begs the question, "why do you need my money to fund something that is already completed???"