Establish whether fractional-order hardware is practically successful

Establish whether fractional-order hardware can succeed in neuromorphic systems by resolving the remaining bandwidth gap, low-frequency characterization gap, and absence of energy measurements.

Background

The review concludes that the central obstacle is not primarily the available fractional order: fabricated passive devices cover many task-optimal orders. Instead, the unresolved practical issues are whether devices can operate over the required low-frequency band, whether integrable devices can reach that regime, and whether their energy cost is competitive.

Without measurements addressing these issues, the field cannot determine whether fractional-order hardware offers a real advantage over integer-order heterogeneous or adaptive neuromorphic systems.

References

None of those is a proof that the approach fails. All three are reasons nobody can yet say whether it succeeds, and that is a sharper and more actionable position than the one the field currently occupies.

— Fractional-order hardware for neuromorphic computing: Is the order really the problem?  (2609.10882 - Teuscher, 9 Sep 2026) in Section 8, concluding paragraph

Efficiency projections use analytical cost models anchored to published RTL synthesis~\citep{przewlocka2022pot}; open-source logic synthesis (Yosys~\citep{wolf2013yosys}) confirms the structural ordering and bit-width scaling (Appendix~\ref{app:hw_yosys}), and end-to-end RTL/ASIC validation is left for future work.

— Fractional Power-of-Two Quantization for Efficient and Effective Multiplier-Free LLM Inference  (2610.21257 - Wee et al., 15 Oct 2026) in Section 6, “Limitations” (Discussion)