Just to be clear, in this very thread of the last 5 hours, I ONLY asked for a number based on (spice) simulation. That was all. I didn't care about the silicon-validation numbers today. In a thread a few days ago on #silicon-validation, it became clear to me that this number can't be measured under the given circumstances, so I explicitply asked for spice simulation numbers today on #openram So I guess I have to fire-up spice simulation in OpenRAM and try to figure out this number by myself. I was hoping that openram designers have played with this parameter already.

We haven't played with that parameter...