Interesting to speculate on the engineering workarounds if all FPGAs disappeared tomorrow. I'm not sure there's a big long-term impact. Maybe more shift to mor

Interesting to speculate on the engineering workarounds if all FPGAs disappeared tomorrow. I'm not sure there's a big long-term impact. Maybe more shift to more to DSP, make low volume ASICs on older nodes cheaper and cycle faster, put up with more card swaps since you can't update certain things in firmware so easily. You can still do all the things but maybe the design takes longer or you sacrifice some performance (designer gets to choose the tradeoff). My guess is the main "real" impact is cycle times, design retooling and having to pause and rework existing designs suffering from supply chain disruption. Potentially all problems you can fix scalably by throwing enough bodies at them since the money is clearly there. Would be interested in perspectives from anyone working in a related area.