Delete comment from: Ken Shirriff's blog
Many thanks for your blogs on the Z-80! I am building a Z-80 replica from discrete transistors (surface mount) in CMOS and this project will be a huge help--especially your test cases.
On the XOR circuit: I built an LT-Spice model of it, and it is pretty sensitive to process and resistance value. The problem is that transistors are either turned on conducting ground or they are high resistance, and the base of following transistor has enough capacitance to be a real effect when you are trying to run at a few MHz. The XNOR output has good, sharp transitions to ground, but charging the RC network up delays and rounds the edge of the rising edges.
No question why designers later switched to CMOS process, but this circuit, as built, would not work except in NMOS process.
Nov 24, 2016, 4:55:05 PM
Posted to Reverse-engineering the Z-80: the silicon for two interesting gates explained