Regarding https://github.com/google/skywater-pdk-libs-sky130_fd_pr_reram/issues/10 and https://groups.google.com/g/xyce-users/c/VjropkUTYyg. If the Verilog-A model isn't working how were the previous projects able to simulate and produce a successful tapeout?

There was only one initial tapeout with only basic support for ReRAM. Since that tapeout has not yet come back from manufacture, there is no way to guarantee that the tapeout is successful. However, the ReRAM process has been under development for some time, so there are people who are aware of how it works whether or not the simulation model is correct. The behavior is hysteresis in the I-V curve with an open eye that gives a pretty large margin for error, so a circuit can be designed around the measured I-V curves and can be made to work whether or not a model exists for the ReRAM device.

@User Thanks

@User: The page I looked at was very scant on information compared to the page that Tim Ansell just posted the link to, below, so probably everything I said about the ReRAM was wrong. I will spend some time trying to understand the form/set/reset stages before I say anything more about it.