I'm Still In Love With My X! (but, do I want my X to be an optimist, a pessimist, or eliminated?)
Stuart Sutherland, SystemVerilog Trainer · 2013
This paper explores the advantages and hazards of X-optimism and X-pessimism, and of 2-state versus 4-state simulation. A number of papers have been written over the years on the problems of optimistic versus pessimistic X propagation in simulation. Some papers argue that Verilog/ SystemVerilog is overly optimistic, while other papers argue that SystemVerilog can be overly pessimistic. Which viewpoint is correct? Just a few years ago, some simulator companies were promising that 2-state simulations would deliver substantially faster, more efficient simulation run- times, compared to 4-state simulation. Now it seems the tables have turned, and Verilog/SystemVerilog simulators are providing modes that pessimistically propagate logic X values, with the promise that 4-state simulation will more accurately and efficiently detect obscure design bugs. Which promise is true? This paper answers these questions.