The superior GN1 chip has always been ready to push 700+ GH/s, given enough juice.
The incompetent subcontractor for the first board screwed up royally, and the next ones could only do 400+ GH/s until the Yoli and Habanero designs.
Converting that much clean power in such a small area is no small engineering challenge.
The incompetent subcontractor for the first board screwed up royally, and the next ones could only do 400+ GH/s until the Yoli and Habanero designs.
Converting that much clean power in such a small area is no small engineering challenge.
And this is where their failure is (but of course not provable in court). Trying to use so much power in such a small area lead them to the bankruptcy. No other chip manufacturer used this approach and nobody will attempt to try it because it leads to so many problems. Bitcoin mining doesn't need a simple "fastest chip". Bitcoin mining needs a more balanced approach of many things besides speed. In my view HF failed when they chose this approach and also when they never thought of any plan B in case the subcontractor/PCB design fails. The only thing HF was good was spending a shitload of money while not being able to fully deliver their first batch of miners. The other things like refunds and MPP don't matter when you can't even fully ship your first batch of miners. Good ASIC design skills, but very bad management/business decisions. HashFail!
Edit: SolarSilver types faster
