Are there many individual components of the Xnn series of algos which won't fit on an arria or even a cyclone ?
Yes. Some of the components hardly fit on a 9P. The cubehash example i gave a few posts ago.
Are you saying the algorithm for an individual cubehash pipeline hardly fits on a 9P???... while I havent studied cubehash specifically, its claimed to take 200 cycle on a basic CPU, and I can implement a lot of basic CPU cores on a 9P...
A pipeline, sure, lots of pipelines, if you want to unroll it fully and obtain real (1Gh/s+) performance it would take the entire 9P and it's not clear that a fully unrolled version of it would fit at all.
Ahh ok - I completely understand what youre saying now. I misread as the individual algorithm took the chip.
The way id attack Lyra2Rev2 in hardware is a literal pipeline of chips, sized according to paralyzed throughput. It looks like the whole chain is 256 bit hashes, so 400gbps interconnect could handle your 1Gh+. For chip to hip interconnect on the same board 3 quads of 32 Gbps should be sufficient. The Blake/keccak skein is probably all on the same chip or a much smaller chip.
Your 9P is probably $3000, you could buy 2-3x the luts for pipelines on smaller chips for that...
All that said it looks like the 1GH is worth about $1000/mo right now, so still quite a long payout if youre using $12k in hardware.